Dear All.
You are cordially welcome to join in, with what appears at first sight to be a promising self running device.
http://www.youtube.com/watch?v=Ph2sDN-fmzY&feature=youtu.be
Over the weekend Groundloop has painstakingly converted the schematic into a PCB layout suitable for us to make a, hopefully working device.
T-1000 has said that the transformer is the "Magic" bit of the device. If you study closely the Primary and Secondary windings are in fact separated !! Each wound on a Ferrite half. He has also found out there is a slight air gap created by gluing the two halves together.
I am passing on the data provided by Groundloop to my friend Richard of RM Cybernetics who, I hope will make us a few prototype boards ready for testing.
That's all for now, Cheers Grum.
Dear All.
Here is a rather poor resolution close up of the transformer.
It does however show the coil orientation.
Cheers Grum.
Thanks for the heads-up, Grum!
If your friend makes a few boards, pls count me in if you can -- and I would be most happy to pay for it!
Also, I think I have some ferrite cores a lot like the ones you show... and lots of wire...
I found that Akula0083 is Roman Karnouhov from Russia. I can't find much on him.. does he have youtube, blog, or post on any forums.. I would like to see more of his work.
Found the thread on Akula and his videos.. very interesting.
http://www.overunityresearch.com/index.php?topic=2227.0 (http://www.overunityresearch.com/index.php?topic=2227.0)
Do you guys know of any successful replications of any of his circuits?
Edit: To answer my own question; x_name41 may have and that's his video in the OP.
Thanks!
I found this 1W version over at OU.com
Edit: Attached higher res a few post down in the thread.
@Grumage
It is good you opened a thread on this Akula guy.
Because the 30watts video is showing a self-runner, which is a great statement of OU if true, I decided to take a look at his video before I consider spending any time on it myself.
Sadly, I am not convinced this video is real.
I am showing a few images.
Image 1 shows wire going under the battery.
Image 2 shows wire that was under the battery once the battery was removed.
Image 3 shows same wire from another angle.
Image 4 shows a second wire going through a hole that he tried to cover with some gadget.
So the circuit is spiked from under the table coming up directly under the circuit. You can see other small holes were drilled probably the first holes that he decided did not position well the circuit and transformer on the table.
So consider the circuit position on the table is anchored or locked in with two wires holding it from underneath the circuit. This explains why the circuit DID NOT MOVE while he was manhandling it when he was.........
1) placing the hot wire from the battery to the circuit input diode.
2) removing both the red and black clips when he removed the battery.
3) taking the transformer in his hand, again the circuit did not move any more then a circular angle movement or pivot that is consistent with the circuit being anchored from underneath.
Anyone else in this world that would handle the circuit in such a manner would have surely moved the circuit from its position on the table. I have cursed more then once when that happens. Now that you have this information, go back and look at his video and look at how he compensates his movements especially when he realizes he videoed the two wires, he moves the camera away. Then look when he manipulates the connections or removes the battery.
Sorry to break your bubble but if you guys are to spend another ion of time working on this, Akula will have to make a much better 30 watts video with less crap.
I get the very uncomfortable feeling that these guys build something, then they know it is not a self-runner but the trick their videos to make it seem like a self-runner to entice replications and if by chance one of the replicators works out some miraculous final bug to make theirs a self-runner, Akula will grab all the glory.
REPLICATORS BEWARE.
wattsup
Ya, Ya, I know. It's that damn @wattsup again with his Wire X. But WTF man. You want to spend your life running after shadows or do you want to spend your life working on real devices. These Russian guys make their videos, never show input and output results, always make videos from hell and then we have to lick their asses for some crumbs. Not a good OU research strategy at all.
I think this exposé would merit a response from Akula to make a better video if his device is real because right now, it smells of faked juice.
Dear Wattsup.
I will open this reply with the point that the video WAS NOT presented by Akula. However your comments and pictures are noted, and DO seem valid !!
What more can I write ?? Groundloop has gone to a lot of trouble for us, so I think the least I can do is to carry on regardless as the parts are of minor cost. And see what happens!!
I will however state that this schematic is a follow on from some earlier work from a Russian named Pantiuchov. I built a device from the schematic that self ran for 30 sec's before the storage cap dramatically exploded !!
So I think there may be some merit to this one, time will tell.
Cheers Grum.
Oh sorry, I didn't save the image correctly. Here is a higher res image of the 1W version.
Seems Akula started big and has been working on scaling down.. just the opposite of what we normally see, but easier to replicate.
I wish that there was some detailed instructions on how to make the transformer.
I found a video on the 1W.
Here is the schematic in the video. I think this is the blog in the schematic, but not Akula. http://renuar.blogspot.com/ (http://renuar.blogspot.com/)
We all know Wesley right? The guy who went to visit Kapanadze in person. Akula is one of his friends (or fellow experimenter) who Wesley has worked with a lot. In my book Akula has high credibility and I don't believe what wattsup is presenting here is any proof at all of fakery. Where wattsup sees wires and holes seems like imagination getting carried away to me. Sorry wattsup but I am not buying this hidden wire conspiracy this time either. Could we ask Akula to do a better video? Sure but then someone would come along and say he edited out the hidden wires. Not hard at all to do with a video editor. I'm not against asking for another better video though ....
I'll add a couple more thoughts too. Most people are terrible at doing video's. Most move the camera around too fast and too much. Also don't you think if someone was going to go to all the trouble to build that circuit and then to fake a self runner they would have done a much better job of hiding wires and holes if those were really part of faking a self runner? I know I wouldn't want to get caught making a fake and stand losing all credibility .... only makes sense such a person would cover all obvious sources of fakery if they were faking it. My opinion anyway ....
Quote from: e2matrix on 2014.03.03, 18:26:29
We all know Wesley right? The guy who went to visit Kapanadze in person. Akula is one of his friends (or fellow experimenter) who Wesley has worked with a lot. In my book Akula has high credibility and I don't believe what wattsup is presenting here is any proof at all of fakery. Where wattsup sees wires and holes seems like imagination getting carried away to me. Sorry wattsup but I am not buying this hidden wire conspiracy this time either. Could we ask Akula to do a better video? Sure but then someone would come along and say he edited out the hidden wires. Not hard at all to do with a video editor. I'm not against asking for another better video though ....
Well, for Wesley you can ask himself in PM and I also have Roman's contact in skype. The latest video you seen is just another replica and Akula was not there :)
If you did lots of research over last few years you might notice about people doing some "magic" to transformers in way how they are made and connected in order to reach unconventional effects. For example, there is no single conventional circuit showing connected BEMF in series with charged cap and dumping that back to power source. Also when transformer is asymetrical and bloch wall moves inside of it that cause all other effects and it is interesting topic to do experiments on. You might find lots if different people refered by this but there is something interesting for sure...
Cheers!
Quote from: T-1000 on 2014.03.03, 19:22:57
Well, for Wesley you can ask himself in PM and I also have Roman's contact in skype. The latest video you seen is just another replica and Akula was not there :)
If you did lots of research over last few years you might notice about people doing some "magic" to transformers in way how they are made and connected in order to reach unconventional effects. For example, there is no single conventional circuit showing connected BEMF in series with charged cap and dumping that back to power source. Also when transformer is asymetrical and bloch wall moves inside of it that cause all other effects and it is interesting topic to do experiments on. You might find lots if different people refered by this but there is something interesting for sure...
Cheers!
Can you ask Roman for detailed instructions on how to make the toroid transformer?
Thanks
I see Itsu worked on the 1W a bit.. anyone else? I noticed the schematics he used didn't have the transistors marked, but the one in this thread that I got from OU.com does.. maybe that will help Itsu.
http://www.overunityresearch.com/index.php?topic=2227.msg36309#msg36309 (http://www.overunityresearch.com/index.php?topic=2227.msg36309#msg36309)
@e2matrix
Regardless of who made the Akula0083 video, the video IS faked. There is simply no way a circuit board would stay in one place like that with all the manoeuvring he did.
So to confirm this, I did a small screen capture using my virtualdubmod program to show you directly. Because the file is too big to post here and I did not want to upload this video into my youtube channel, I put it on my ftp web site here...
http://purco.qc.ca/ftp/possible-fakes/akula0083/
Sorry if I said Akula made that video. Look, I don't know why people do things like that but I do know that other people wind up wasting a load of time and money. This is one reason why I no longer do any replications based on a video before I do any due diligence. Time + Effort + a dab of luck = OU. Wasting your time = no OU.
wattsup
Dear 4Tesla.
I have noticed that you have posted a slightly different schematic for the 1 Watt device than that of the one Itsu used !! Already there is one of two conclusions we can draw!! Is the later one correct ? Or just another round of disinfo !!
I do agree with some of wattsup's arguments but as I wrote earlier as the device should not cost more than £50.00 to build then , what the hey, let's give it a go !!
There is a lot of chatter over on OU.Com about Akula seeming to now have more expensive test gear. They presume that he is making money from his videos. But as stated earlier, this was not him, just another guy !!
Cheers Grum.
Hi People,
Quote from: Grumage on 2014.03.02, 19:20:40
T-1000 has said that the transformer is the "Magic" bit of the device.
If you study closely the Primary and Secondary windings are in fact separated !!
Each wound on a Ferrite half.
He has also found out there is a slight air gap created by gluing the two halves together.
My two cents:
What is a trafo with (physically) separated Primary and secondary?
IMHO, it is a (small, indeed, but real) capacitor.
You can call it a Coilcap.
As I (curiously) keep on seeing (non conventional) capacitors everywhere, I recently
made these (attached) 2 short pdfS.
-------------------------
I, of course, did not discovered this by myself but while reading Vladimir Utkin's texts
(a Russian guy, BTW).
Beware the Russians! :) OU-wise they are genius! (IMHO).
I should have learnt the Russian language instead of the German one when I was young.
Morality: 2 years of German course and the result appart from
"Die Kleine Vögel fliegen im Himmel" is: nichts, nothing, nada, niente, rien, nihil, etc...
-----------------
About Vladimir Utkin, you can consult:
http://freenrg.info/4OUR_F/V_Utkin_FREE-NRG.pdf (http://freenrg.info/4OUR_F/V_Utkin_FREE-NRG.pdf)
http://freenrg.info/4OUR_F/V_Utkin_Free_energy2.pdf (http://freenrg.info/4OUR_F/V_Utkin_Free_energy2.pdf) (about Don Smith's SEG)
http://freenrg.info/4OUR_F/Vladimir-Utkin_newest.pdf (http://freenrg.info/4OUR_F/Vladimir-Utkin_newest.pdf) (not the newest)
http://freenrg.info/4OUR_F/Vladimir_Utkin2.pdf (http://freenrg.info/4OUR_F/Vladimir_Utkin2.pdf) (about the 'Electro radiant Effect')
http://freenrg.info/4OUR_F/Vladimi_Utkin_june_2012.pdf (http://freenrg.info/4OUR_F/Vladimi_Utkin_june_2012.pdf) (the very newest)
These are different versions of basically the same text.
Beware! Vladimir Utkin is not the Bible.
I notably beg to differ with the page 57 of "Vladimir_Utkin_june_2012.pdf".
----------------------
More to come, should anybody be intersted in my considerations.
All the Best,
Jean
Quote from: wattsup on 2014.03.03, 20:41:53
@e2matrix
Regardless of who made the Akula0083 video, the video IS faked. There is simply no way a circuit board would stay in one place like that with all the manoeuvring he did.
So to confirm this, I did a small screen capture using my virtualdubmod program to show you directly. Because the file is too big to post here and I did not want to upload this video into my youtube channel, I put it on my ftp web site here...
http://purco.qc.ca/ftp/possible-fakes/akula0083/
Sorry if I said Akula made that video. Look, I don't know why people do things like that but I do know that other people wind up wasting a load of time and money. This is one reason why I no longer do any replications based on a video before I do any due diligence. Time + Effort + a dab of luck = OU. Wasting your time = no OU.
wattsup
Hi wattsup,
I appreciate you analyzing ou claims on video. I agree it is very frustrating to see all those fake ou videos and hard to get excited about new ou devices when we've had Mylow, RomeroUK, Steorn and so on. I hope that Akula's work is real. I haven't seen enough evidence to convince me either way. At least these lower watt schematics won't take too much time to try out.. if we can get more details on the transformers.
Quote from: Grumage on 2014.03.03, 21:21:02
Dear 4Tesla.
I have noticed that you have posted a slightly different schematic for the 1 Watt device than that of the one Itsu used !! Already there is one of two conclusions we can draw!! Is the later one correct ? Or just another round of disinfo !!
Yes it is slightly different and has the components labeled. I have no idea where GeoFusion over at OU.com got it from. Here is a link to his post.
http://www.overunity.com/12736/kapanadze-cousin-dally-free-energy/msg390297/#msg390297 (http://www.overunity.com/12736/kapanadze-cousin-dally-free-energy/msg390297/#msg390297)
Dear Jean.
Many thanks for your, as always, detailed observations. Verpies has stated many times that the transformer is the most least understood electrical / electronic device that we can use !! So many winding configurations !! Also note that the Russian transformers that appear in so many OU videos are usually separate limb types. Not like the European concentric wound!! ?? Could there be something in that single observation ?? ;)
With the mention of GeoFusion by 4Tesla earlier on I remembered that it sparked off a series of experiments for me, starting with the video posted below. I found that if you place a coil across a gap somehow the frequency doubles from that of the input !!
https://www.youtube.com/watch?v=SPV89JWVdtY
I have quite a few more in the Ferrite saga, as I called it!! Showing various interesting acoustic phenomena !!
On a lighter note I am glad that this thread has sparked a little more interest than the last one I started !! ;D
Cheers Grum.
I found the forum that Akula posts on.. only if I knew Russian
http://realstrannik.ru/forum/user/3683-akula0083.html (http://realstrannik.ru/forum/user/3683-akula0083.html)
http://translate.google.com/translate?hl=en&sl=ru&tl=en&u=http%3A%2F%2Frealstrannik.ru%2Fforum%2Fuser%2F3683-akula0083.html (http://translate.google.com/translate?hl=en&sl=ru&tl=en&u=http%3A%2F%2Frealstrannik.ru%2Fforum%2Fuser%2F3683-akula0083.html)
Hi Grumage,
Thanks for you feed back.
Quote from: Grumage on 2014.03.03, 22:45:29
..............
Verpies has stated many times that the transformer is the most least understood
electrical / electronic device that we can use !! So many winding configurations !!
I Do agree with this very statement! :)
-----------------------------------------------
About the vid: " Split core transformer"
Splitting the core is adding another (very small but more important) capacitor. No? :-\
While we are at it, some new experiment with trafos. 2 pictures.
Sorry if it is not new for you.
(http://freenrg.info/4OUR_F/Exp_With_Trafo1.jpg)
(http://freenrg.info/4OUR_F/Exp_With_Trafo-v21.jpg)
---------------------------------------------------
About my 2 PdfS:
My reasoning is very very simple, as usual. I'm not a scientist. :P
Eric P. Dollard and Jean Louis Naudin (JLN) seem to get very interesting (OU?) results
using what it is called a "Longitudinal 'Magneto Dielectric'" configuration.
This, with just a kinda analog computer (for JLN).
To get this called "Longitudinal 'Magneto Dielectric'" configuration it sounds like
that you just have to 90° CCW rotate the called "Transverse 'Electro-Magnetic'" configuration
(the classical Transmission Line).
Hence, the famous 'multiplying by the square root of minus one'...
Please, notice that I can wrongly understand this Eric P. Dollard statement! :-X
Then, firstly, comes Konstantin Meyl with a similar (Transverse) configuration (his open capacitor) that
IMO, resembles the 'Transverse' stuff. Not to mention some Tesla patents.
After that, appears Vladimir Utkin, with, indeed, very interesting inputs. No?
So (#1), I'm just wondering whether (one of) the (main) 'secret' might be that simple...
So (#2), I'm also just asking me: what if all these 'inventors' have just figured out
one of the essential 'secret' and, after some experiments, were able to design strange devices?
Without, of course, telling from where their inspiration came...
Just my 2 cents,
All the Best,
Jean
Found more info on Roman.
http://www.overunity.com/7679/selfrunning-free-energy-devices-up-to-5-kw-from-tariel-kapanadze/msg363790/#msg363790 (http://www.overunity.com/7679/selfrunning-free-energy-devices-up-to-5-kw-from-tariel-kapanadze/msg363790/#msg363790)
Looking at related youtube vids, came across this by Wasaby:
https://www.youtube.com/watch?v=3u-i3wdESxE&feature=c4-overview&list=UUM82bm9KN7DMBkTfR-psyqQ
Screen shot shows some detail, but no ckt diagram is given:
I read the last few pages of the thread "Selfrunning Free Energy devices up to 5 KW from Tariel Kapanadze" from page 1417 over at OU.com. Good stuff, but no way I can start from the beginning.
http://www.overunity.com/7679/selfrunning-free-energy-devices-up-to-5-kw-from-tariel-kapanadze/21240/ (http://www.overunity.com/7679/selfrunning-free-energy-devices-up-to-5-kw-from-tariel-kapanadze/21240/)
Quote from: PhysicsProf on 2014.03.04, 07:32:03
Looking at related youtube vids, came across this by Wasaby:
https://www.youtube.com/watch?v=3u-i3wdESxE&feature=c4-overview&list=UUM82bm9KN7DMBkTfR-psyqQ
Screen shot shows some detail, but no ckt diagram is given:
I hate it when they say "What is your opinion? is it reall or fake?" makes me guess fake.
Quote from: 4Tesla on 2014.03.03, 19:54:59
I see Itsu worked on the 1W a bit.. anyone else? I noticed the schematics he used didn't have the transistors marked, but the one in this thread that I got from OU.com does.. maybe that will help Itsu.
http://www.overunityresearch.com/index.php?topic=2227.msg36309#msg36309 (http://www.overunityresearch.com/index.php?topic=2227.msg36309#msg36309)
Hi 4Tesla,
thanks for this new diagram, its as Grumage mentioned again different.
None of the mentioned transistors there seems to resemble the big 2n3055-like transistor on the heatsink though.
They do however match the description of a red/brown capacitor i made in the other thread when looking at the video of the device.
Biggest problem as i see it is not the transistors, but the make up of the coils which in your diagram has an added 4th (dual) coil.
Thanks for looking into this, regards Itsu
Dear All.
I am providing a link to a Firm (UK Based) that has supplied me on many occasions with Ferrite products.
http://powermagnetics.co.uk/calculator
From their very useful inductance calculator you can go on to the shop!! Today I bought ETD49-CF139 TIMES 2 And 1 by ETD49-90794-20P. All under £10.00 inc shipping !! Probably dearer for other countries!!
I decided to go with a gapless core as it is easier to make a gap than fill one !! :)
I have not heard from my friend yet as regards the PCB's, but the info is with him.
I have a good feeling with this one. As I have heard some whisperings !! ;)
Cheers Grum.
Quote from: 4Tesla on 2014.03.03, 23:20:45
I found the forum that Akula posts on.. only if I knew Russian
http://realstrannik.ru/forum/user/3683-akula0083.html (http://realstrannik.ru/forum/user/3683-akula0083.html)
http://translate.google.com/translate?hl=en&sl=ru&tl=en&u=http%3A%2F%2Frealstrannik.ru%2Fforum%2Fuser%2F3683-akula0083.html (http://translate.google.com/translate?hl=en&sl=ru&tl=en&u=http%3A%2F%2Frealstrannik.ru%2Fforum%2Fuser%2F3683-akula0083.html)
While it doesn't always translate 100% clear just use the Google Chrome browser and it will automatically translate any page for you.
Quote from: PhysicsProf on 2014.03.04, 07:32:03
Looking at related youtube vids, came across this by Wasaby:
https://www.youtube.com/watch?v=3u-i3wdESxE&feature=c4-overview&list=UUM82bm9KN7DMBkTfR-psyqQ
Screen shot shows some detail, but no ckt diagram is given:
The first thing I would ask is what kind of light bulb that is made to fit your common 120VAC socket can run off a single AA battery as he shows at one point directly hooking it to the battery? Of course the answer is it's a very modified light bulb. I can think of some ways this is faked and am fairly certain it is fake.
Quote from: e2matrix on 2014.03.04, 19:30:50
The first thing I would ask is what kind of light bulb that is made to fit your common 120VAC socket can run off a single AA battery as he shows at one point directly hooking it to the battery? Of course the answer is it's a very modified light bulb. I can think of some ways this is faked and am fairly certain it is fake.
O0
too fantastic to be real see also the same bulb in the fan video and the magic effect of the magnet !!.
who is this FE man Houdini?
https://www.youtube.com/watch?v=jiAhiu6UqXQ&list=UUM82bm9KN7DMBkTfR-psyqQ&feature=c4-overview
Quote from: wings on 2014.03.04, 19:57:39
O0
Thank you for this thread and please concentrate on Akula schematic. I have an idea of the principle of working of this circuit and it was previously posted on ou.com as a overunitymethod.pdf. Akula seems to be slightly modified implementation of this principle. The collapsing magnetic field in coil sustain current flow in forward direction throught both diodes and back to the source capacitor while inducing again in transformer proper current . Note the important factor of 1:3 transformer ratio stepping down the flyback spike. I believe this circuit is genuine and working after some corrections. I will check maybe next week. Good thing if this circuit works there is only very few possible principles involved...
Dear forest.
I have to agree with you, my own feelings are these circuits are closely related to the ones posted below. In fact when I joined OUR there was some interest in the bottom schematic.
Cheers Grum.
Here is the document I referenced.http://www.google.pl/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CCwQFjAA&url=http%3A%2F%2Fwww.overunity.com%2F12736%2Fkapanadze-cousin-dally-free-energy%2Fdlattach%2Fattach%2F132482%2F&ei=RTsWU7rlJKfD4gSt84DoDg&usg=AFQjCNES-8ssChEay0jk98QgNDpoADkonA&sig2=F8E7fwOKwqB_8ekLTBIEbg&bvm=bv.62286460,d.bGE&cad=rja
Have to go.. when I get back on the weekend, I'll try and study the schematics (all versions 1 - 80 watt).. find what is common in them. The frequencies used, windings of the transformers, etc. We need the basic understanding.. some here might know, but I don't.
:)
http://192.211.49.220/-%20Free%20Energy/Akula/
Ciao a tutti,
One more PDF in attached file. Should anybody be interested in.
Actually, my aim is to write one single document about "Caps and possible 'OU'?"
------------------------
A personal and apparently off topic message to PhysicsProf:
You must remember this Prof. "Nenad Savic" boiler claimed 'OU' very apparatus that
you get and I tested, a couple of years ago, without success. :(
Failure is mine, of course, but I wish I were more helped by this dear Nenad.
It is not complicated! I soon as I asked Prof. Savic a good question, I was not answered.
Finally, this gave me some clues by not enough at the moment.
My last unanswered question to Prof Savic was something like:
"Is it of any importance that one part of the apparatus is ferromagnetic?"
This actually is the case.
I was also told about some cryptic "standing wave".
Now (thanks to Vladimir Utkin) my understanding is, basically:
- This Nenad Savic's device *is* a capacitor.
- Charging and discharging a cap with the grid cost almost nothing. Right?
- But, between the plate of a cap there is a "displacement current" and a variable electric field. Right?
- This variable electric field generate a variable magnetic field at right angle. OK?
- This variable magnetic field is able to move (vibrate) the ferromagnetic capacitor "plate". Right?
- The second non ferromagnetic "plate" acts as a "base".
- This vibrating "plate "should be able to heat water. No?
This for practically no power consumption. Hence, the "OU" behavior, should any "resonance" (between one "plate" of the device and the grid freq) is achieved.
Now, to behave as a "real" capacitor there must be no "real current" between the capacitor plates.
So, this device must only work in distilled water. This was not specified.
Unless the "standing wave" and the following picture?
I, in fact, observed this "Water Rim" but it was a transient phenomenon,
(http://freenrg.info/4OUR_F/Stand_Wave.jpg)
Any feedback, PhysicsProf?
Cheers,
Jean
Excellent detective work, Jean. O0
I like your reasoning and want to put this to a test.
We left home yesterday, now traveling to visit family, but when I return to my home lab, I will give it a try -- OK? and I will use distilled water.
I wonder though- if the ferromagnetic cap vibrates, it seems there must be SOME power consumption.
Worth some testing!
I am also very interested in the Akula device.
Quote
My last unanswered question to Prof Savic was something like:
"Is it of any importance that one part of the apparatus is ferromagnetic?"
This actually is the case.
I was also told about some cryptic "standing wave".
Now (thanks to Vladimir Utkin) my understanding is, basically:
- This Nenad Savic's device *is* a capacitor.
- Charging and discharging a cap with the grid cost almost nothing. Right?
- But, between the plate of a cap there is a "displacement current" and a variable electric field. Right?
- This variable electric field generate a variable magnetic field at right angle. OK?
- This variable magnetic field is able to move (vibrate) the ferromagnetic capacitor "plate". Right?
- The second non ferromagnetic "plate" acts as a "base".
- This vibrating "plate "should be able to heat water. No?
This for practically no power consumption. Hence, the "OU" behavior, should any "resonance" (between one "plate" of the device and the grid freq) is achieved.
Now, to behave as a "real" capacitor there must be no "real current" between the capacitor plates.
So, this device must only work in distilled water. This was not specified.
Quote from: NerzhDishual on 2014.03.05, 21:52:20
You must remember this Prof. "Nenad Savic" boiler claimed 'OU' very apparatus that
you get and I tested, a couple of years ago, without success. :(
The moment I saw your picture, It reminded me of Peter Daysh Davey and his device. Back in the fifties (I think), he has two bicycle bells which are tuned to the AC frequency being fed to them (or an octave - but not a harmonic).
The tuning is crucial.
[I'm not sure how this slots in to this thread - but I offer it anyway]
.
Hi Paul-R,
Yes, Peter Daysh Davey... :)
I made a lot (un successful (OU-Wise)) experiments with a lot of these kind of devices.
Yes, you can heat water very quickly.
A non insulated Davey-like device seems to be more efficient than a mere non insulated too electric kettle.
Should it be possible, achieving "obvious OU" (I mean: a repeatable COP of, saying, > 1.5 ) seems to be another story.
I agree with you: this must be a matter of very very precise crucial tuning.
What to tune is another question... No? I finally gave up with my Daveys-like cups.
----------------------------------
Then came Prof. Nenad Savic.
I was sent a device claimed to be tuned for 220 Vols and 50 Hz.
I had a very interesting, but hard time, dealing with Prof Savic (in Googled Serbian language, :-X )
and his very apparatus.
My results were absolutely fuzzy, unpredictable, confusing, deceptive, non reproducible, and sometime
very encouraging (for a short while)...
I gave up, once again and dispatched the Savic's thing to PhysicsProf.
---------------------
Now, I'm very very interested in these Capcoils? CoilCaps?
Or:
Transpacitor? or Capaciformer? (ewizard)
http://www.energeticforum.com/renewable-energy/11451-transformer-core-capacitor-plates.html (http://www.energeticforum.com/renewable-energy/11451-transformer-core-capacitor-plates.html)
A very interesting (short) dead thread, that I stumbled upon today. BTW...
Now, I see capacitors everywhere. :o
-----------------------
The point is that Prof. Nenad Savic always insisted that his apparatus, resembled the Davey's things, indeed,
but was actually based on another principle and quoted Tesla (of course).
That is why I though of the condenser about the Savic's device and designed a pet theory about it.
Of course, I can be wrong. :'(
One simple way to figure this out is to test the Savic's device with distilled water.
This is why the experiment that PhysicsProf (to whom I give my regards, by the way) is proposing
can be of great help.
--------------------
IMHO (= my intuition) is telling me that:
Tesla
Akula
Kapanadze
Don Smith
Konstantin Meyl
Eric Dollard
Duncan?
Same things...
Meilleurs Sentiments,
Jean
About transformer used in 30W self runner:
(http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/dlattach/attach/134541/image//)
I might be wrong but here is my point of view on how it is made:
1) Take core with same spec as it is in any switching power supply;
2) Wind 30 turns (15 forward 15 on second layer) of L2;
3) The rest 15 turns(8+7) of L2 goes to the middle point.
4) The L1 15 turns(8+7) are winded in opposite direction to the middle completing layer.
5) Leave very small gap between halves when joining core back.
When it is done, it is ready for testing as per circuit connections in first post. By the way, any 555 timer driver circuit would do the job.
Hopefully someone can pick that up and do a test... :)
Quote from: NerzhDishual on 2014.03.06, 19:18:27
I agree with you: this must be a matter of very very precise crucial tuning.
What to tune is another question...
.
You tune to the frequency of the AC going to the bells (or an octave - often 3 octaves up, 400hz if the AC is 50hz).
.
Hi Paul-R,
OK...
Some years ago, I made some (laborious) freq. measurements with my cups.
http://freenrg.info/Sonettes_Davey/Freq/ (http://freenrg.info/Sonettes_Davey/Freq/)
It is not so easy to determine the main freq of *one* single cup, not to mention
the "main freqs" of 2 assembled cups.
I'm not (at all) a machinist neither a metal fitter nor an organ pipe tuner... :-X
IMHO, these devices could work and show "OU" but they are not for a layman like me
(even with some musical smatterings) ... :'(
Now, all these useless cups stuffs are dumped in a nice cardboard box in my cellar. :P
-----------------
Meanwhile, I can not help to keep on repeating about "CoilCaps".
3 Pictures from and edited pdf (in an attached file (Mislavskij.pdf)).
(http://freenrg.info/4OUR_F/Mislavskij-Transformer-P1.jpg)
(http://freenrg.info/4OUR_F/Mislavskij-Transformer-P2.jpg)
(http://freenrg.info/4OUR_F/Mislavskij-Transformer-P3.jpg)
All the Best,
Jean
Buonasera, Добрый вечер, Good evening,
'In' Topic.
Just in case you have not yet consulted another famous forum (emphasizes are mine):
----------------
http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg391515/#msg391515 (http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg391515/#msg391515)
From Gyulasun
"I think It must be a rectangular metal plate positioned to a certain, a few cm distance from coil L2.
I believe there is a capacitive coupling between this plate and the coil....
............
I have seen such plates positioned near to Tesla coils in some Russian youtube videos."
------------
http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg391536/#msg391536 (http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg391536/#msg391536)
From: a.king21
"It is a one turn plate surrounding the transformer. It has a gap in it like a 9/10 turn.
It's purpose is to introduce the Tesla's radiant energy receiver into the circuit.
It is an electrostatic device and follows the rules of electrostatic induction.
It is therefore not subject to the laws of electromagnetic induction which is always cop <1.
It is subject to the laws of electrostatic induction which is cop = 2 minus system losses.
The naysayers should learn their physics."
----------------------
http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg391553/#msg391553 (http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg391553/#msg391553)
From: a.king21
"A simplified king-grumage schematic is enclosed so you can see what is going on."
(http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/dlattach/attach/134578/image//)
----------------------
Do not make short CCT.... :o
I have already read such recommandation in some Vladimir Utkin's text.
--------------------------
http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg391559/#msg391559
From: MenofFather
"I remove not important, not needed parts in this circuit."
(http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/dlattach/attach/134581/image//)
----------------
Amazing. No?
All the Best,
Jean
Quote from: NerzhDishual on 2014.03.07, 18:46:38
-----------------
Meanwhile, I can not help to keep on repeating about "CoilCaps".
3 Pictures from and edited pdf (in an attached file (Mislavskij.pdf)).
...
(http://freenrg.info/4OUR_F/Mislavskij-Transformer-P3.jpg)
All the Best,
Jean
Thanks for this, Jean. I've always been puzzled or fascinated by the displacement current in Maxwell's equations.
And now this "real device" of the Mislavskij xformer -- so he has done experiments, you say. Very clever, but I would like to see
RESULTS, measurements of course. Input power, output power for the different configurations. Have you built this xformer?
So, are there any test data? (If so, perhaps you should start a separate thread.)
Also, is there any connection between this xformer and the Akula device(s)?
Thanks, my friend - fun to think about the "displacement current transformer."
Hello PysiscsProf,
Sorry, I have no information or any results about the Mislavskij transformer.
All I got was from Vladimir Utkin.
Anyway, a google search with Mislavskij transformer seems to give some urls.
I also have not yet done any experiments.
For the moment, I'm trying to connect some dots.
In the AKULA's device it also seems to have a sort of "loose" capacitor (what Gyulasun calls
"a capacitive coupling between this plate and the coil"). You can consult my previous post.
Now, I just have stumbled upon another "dot" to connect: "Charging By Induction".
-------------------------------------
In the following vid, quoted here:
http://energeticforum.com/renewable-energy/6686-electrostatic-induced-generator-free-energy.html (http://energeticforum.com/renewable-energy/6686-electrostatic-induced-generator-free-energy.html)
"Charging By Induction"
http://www.youtube.com/watch?v=o0hXYfvnND0 (http://www.youtube.com/watch?v=o0hXYfvnND0)
I can hear, at 1:00:
"The easiest way to neutralize a charged object is to connect it to the earth
with a good conductor called the ground. If the object is negatively charged, the excess
electrons flow into the ground. until it is neutral. If it is positively charged, electrons flows
from the ground into the object, until once again it is neutralized.."
I can also hear, at 7:16:
"Since there was no exchange of charges between the rod and the electroscope, the
rod can be used to charge any number of electroscopes. Each electroscope will acquire a
charge opposite to that of the road."
-------------
Indeed, discharging electroscopes (even charged for free) to the ground will not solve
any energetic issue. :P
But, what about discharging/charging "something" (a cap ?) via and inductor to/from the ground?
Tariel Kapanadze?
Nilson Barbosa and Cleriston Leal?
Cheers,
Jean
yes, you need a magic rod ;D as curious as it heards it may be a real solution indeed O0 for example in Tesla patents he used special bulb radiating radiant energy
Dear All.
I have opened a thread over on OU.Com. I thought it might help if I posted the coils arrangement proposed by T-1000 here for some constructive comments. Also if any are interested in purchasing ready made PCB's !!
"T-1000's suggested scheme for the coil. This would put the first layer of 30 turns of L2 right in the air gap !!
At this stage I am waiting for R M Cybernetics to get back with a price for the manufacture of the PCB's. I myself do not feel confident enough to build a working circuit. Perhaps if any members are interested in purchasing the boards direct from RMC, they could make it clear here. A run of 10 or more would certainly be cheaper than one or two !! "
Cheers Grum.
I found Akula's Youtube channel, but looks like he deleted (or made private) most of his OU videos.
http://www.youtube.com/channel/UC2aHNMDJzRW7YDd145_Pa2w (http://www.youtube.com/channel/UC2aHNMDJzRW7YDd145_Pa2w)
Quote from: Grumage on 2014.03.09, 21:53:40
At this stage I am waiting for R M Cybernetics to get back with a price for the manufacture of the PCB's. I myself do not feel confident enough to build a working circuit. Perhaps if any members are interested in purchasing the boards direct from RMC, they could make it clear here. A run of 10 or more would certainly be cheaper than one or two !! "
Cheers Grum.
Why not try on breadboard first?
Quote from: 4Tesla on 2014.03.09, 22:23:03
Why not try on breadboard first?
Dear 4Tesla.
As quite a few members here know I am more at home with a Lathe and Milling machine, than a breadboard !! :)
No it's not for me I would rather wait. Groundloop has made such a nice job of the PCB I am really looking forward to putting the transformer through it's paces !!
Cheers Grum.
Hi Grumage,
I'm very often game to purchase things.
But, a PCB is just a printed CCT. No?
You need extra components. No?
Where can we find them?
Are you sure that this AKula's device will work?
---
This said, what I can propose you is the following:
You purchase the PCB and all the components.
You test the AKula's device. So, all the initial (hard) work is on you.
---
In compensation, I agree to send you, via Paypal, saying $100.
If enough persons are doing the same, you should be able to purchase all the stuffs
and make the tests.
For my part, should you fail, I would not ask you anything...
I have already given some money for some vanishing projects in the 4Th dimension.
If, as I do wish to you, you succeed, then it would be the time to think about any
'kit' or whatever.
I am mot kidding at all.
Please, send me a private (or not) message with your "Paypal address". Please, creates it if needed.
I will dispatch you $100.
So, perhaps? This will start something?
Bien cordialement,
Jean
Привет (Priviet) (Hello) Forest,
Thanks for your comment.
Do you have any idea of what such a "magic rod" could look like :o
You seems very cryptic and, as far as I can feel it, a little bit scoffing towards me.
RU not C.C
---------------------
Off topic, as often - but while I'm at a post-what about these statements:
"Static" electricity is almost useless" (or something like that).
Yep! Just have a look at the Testatika device...
"Displacement current is fictitious.
Yeah!
"Displacement Current Example 2"
http://www.youtube.com/watch?v=Tnyl8Gj-plk (http://www.youtube.com/watch?v=Tnyl8Gj-plk)
7:51: "..the actual current in the circuit and the displacement current in the circuit
are exactly the same..""
"The effects of centrifugal force are not real". Really?
You can see the following picture (E.I. Linevich's device) and also
the attached pdf (Kanarev_Correcting_The_First_Law_Of_Dynamics.pdf)
(http://freenrg.info/4OUR_F/EI_Linevich_Apparatus.jpg)
Etc...
Now, you can also consult William Beaty "A 'GERM THEORY' OF EDUCATION"
http://amasci.com/miscon/germ.html (http://amasci.com/miscon/germ.html)
Sounds intresting. Not fully read. :-X
до свидания (do svidaniya) Bye,
Jean
-------------
PS: I'm fed up not being able to understand the Russian comments/explanations in some circuits/sketches.
So, instead of complaining of my ignorance, I have recently decided to get some smatterings of this
very 'bizarre' language and the strange alphabet thereto.
Hence some trials.
I do not expect at all to speak any Russian (too many declensions).
I would be happy to be able to consult a Russian/French-or-English dictionnay without too much pain.
Being capable of catching some words in a Russian vid would be a real bliss.
Jean
your Model for proceeding has merit [to underwrite the builders].
of course this would be up to Grum he seems to have a few concerns ,however I would offer to assist there also [financially].
or perhaps he can subcontract to another builder up to the task?
thx
Chet
So I studied the different schematics by Akula.. I don't know electronics well enough to completely understand them. Basically I see it as a signal generator feeding a special "magical" transformer that somehow recaptures more energy than is put in and I am assuming that the frequency is the resonance of the transformer.
Can anyone offer a clearer understanding of these devices for me?
Thanks!
I'll match Nerzh's $100 to you for your effort, Grum:
Quote from: NerzhDishual on 2014.03.09, 22:57:55
Hi Grumage,
I'm very often game to purchase things.
But, a PCB is just a printed CCT. No?
You need extra components. No?
Where can we find them?
Are you sure that this AKula's device will work?
---
This said, what I can propose you is the following:
You purchase the PCB and all the components.
You test the AKula's device. So, all the initial (hard) work is on you.
---
In compensation, I agree to send you, via Paypal, saying $100.
If enough persons are doing the same, you should be able to purchase all the stuffs
and make the tests.
For my part, should you fail, I would not ask you anything...
I have already given some money for some vanishing projects in the 4Th dimension.
If, as I do wish to you, you succeed, then it would be the time to think about any
'kit' or whatever.
I am not kidding at all.
Please, send me a private (or not) message with your "Paypal address". Please, creates it if needed.
I will dispatch you $100.
So, perhaps? This will start something?
Bien cordialement,
Jean
Could you be more patient ? Why don't you wait a moment for any replicators to stand up and say : yes, it works! ..or... do not bother, it's fake ! Time, my friends and patience. I'm sure a lot of people work on it right now. I think the correct procedure would be to replicate 2-4 times, ideally 2 devices by one person (to learn what is important) then the second person can succesfully build the same using own parts (own made transformer) with the correct method of adjustment.
I hate looking people throw money to the trash can.
Forest
Much time and resource has already been spent,now it goes to production ,the idea of offering to help cover some of the costs has little to do with Patience and more to do with gratitude.
To burden a few for the benefit of many is unfair.Also to "hope" that someone will share their results when we have an offer on the table to contribute towards a well organized "replication"and spread the cost around for the benefit of all.
Sounds like a good plan.
Thanks for all you do Grum and Groundloop and others....
And You too Forest!
Chet
Dear All.
I have spent most of the day thinking and coil winding too !! :)
Chet also phoned me. Your offers to help financially are really generous but I feel I must decline.
I would rather have Mr RMC get things moving and each member could then purchase a ready built board from him. We now have the first price come in !! £5.00 per PCB for a minimum of 10. Not bad really !!
I passed on the component list to Farnell and am waiting on their reply.
Earlier on I put the transformer to a PWM but I found nothing of any consequence to report. I wound the transformer to the schedule posted below.
Once again, thank you all.
Cheers Grum.
Hi Grum,
May we please see some photos of your transformer? What frequencies did you run through it? 5khz?
Even though I have never heard of making PCB without testing circuit first, I'll buy one if I can use Paypal.
Thanks!
Grum
when I was growing up we had a nursery rhyme that they taught us,it was the story of the Little red hen.
I can't physically help you with your chore [if I could you know I would] and my offer of assistance holds no unseen cost or obligation [no strings attached],I just feel lead to "help" ,
so know that the offer is there and know that we appreciate your efforts "come what may".
some may think us a bit crazy or silly but one thing is for certain this example of cooperation between yourself and Groundloop ,two men from completely different countries and backgrounds will bare fruit way beyond the outcome of this project,for it is selfless behavior such as this with a commitment beyond just typing and talking which will ultimately bring about the change we all seek.
thanks for your wonderful example and all your hard work.
with all respect and appreciation
Chet
Dear 4Tesla , Chet and all.
Ok here is a photo of the transformer that I wound yesterday. I used ECW as there was no way you could get the 30 turn first layer with any other wire. This might have a crucial bearing on how it is wound in relation to the photo that I posted on page one !!
cheers Grum.
Quote from: Grumage on 2014.03.11, 16:08:49
Dear 4Tesla , Chet and all.
Ok here is a photo of the transformer that I wound yesterday. I used ECW as there was no way you could get the 30 turn first layer with any other wire. This might have a crucial bearing on how it is wound in relation to the photo that I posted on page one !!
cheers Grum.
Looks good! Thanks for posting!
The 1W reminds me of this "Micro TPU" circuit.
Component Values:
2.5 hour configuration
R = 1 Mohm and C = 1.5 uF
The main electrolytic is 1000 uF.
Ferrite core 2 inches
Number of turns:
15 for Recapture coil,
30 for Collector coil (2 coils of 15 turns each),
7 for Trigger coil
Dear All.
My apologies !! I have been a little busy over on the thread I started at OU.com.
I would first like to welcome dllabarre to OUR and I would also like to mention, and welcome RichMo who is the MD of RM Cybernetics.
I have been "bashing the transformer" with square wave pulses of varying frequency and duty cycle. To those that don't visit OU.com I have attached a scope shot of a linear ring !!
Not much further news to share. But looking forward to seeing some results from other experimenters.
Cheers Grum.
Dear All.
This appeared on OU.Com overnight.
http://www.youtube.com/watch?v=Vpwm2-eJiiM
We are way behind !! :)
Cheers Grum.
Great! Anyway we can get this guy to tell us how he wound the transformer?
Edit: uhg.. not the original author of video.
Edit2: asked for link to original video.
In post 30 by T1000 there's a picture of a transformer, if this is what you are trying to replicate i can see it is standard enameled copper wire but it has high temperature white sleeving covering the wire, it must be required to run hot.
If that's a MNZN core you maybe looking at NAR as a method of energy production 8)
Quote from: Peterae on 2014.03.15, 16:08:28
In post 30 by T1000 there's a picture of a transformer, if this is what you are trying to replicate i can see it is standard enameled copper wire but it has high temperature white sleeving covering the wire, it must be required to run hot.
If that's a MNZN core you maybe looking at NAR as a method of energy production 8)
Dear Peter.
Your simple statement just kicked a memory response !! Old Age is creeping in !! ;D
T-1000 mentioned in passing, having spoken to someone that the transformer ran very hot !! A clue ??
Cheers Grum.
Quote from: Peterae on 2014.03.15, 16:08:28
In post 30 by T1000 there's a picture of a transformer, if this is what you are trying to replicate i can see it is standard enameled copper wire but it has high temperature white sleeving covering the wire, it must be required to run hot.
If that's a MNZN core you maybe looking at NAR as a method of energy production 8)
I've seen all we have. I don't think it is accurate. We need more info.
Quote from: Grumage on 2014.03.15, 16:20:36
Dear Peter.
Your simple statement just kicked a memory response !! Old Age is creeping in !! ;D
T-1000 mentioned in passing, having spoken to someone that the transformer ran very hot !! A clue ??
Cheers Grum.
Grum, I would say that is a big clue. I have always said that when dealing with RF at most frequencies, you can develop huge eddy currents, just a simple loop antenna can create hundreds of amps on the surface, the trick is knowing how to utilise them ??? You can see on the circuit there is an RF blocking choke on the return before the voltage regulator, so this must be generating RF.
My two pence worth ;D
regards
Mike 8)
QuoteYour simple statement just kicked a memory response !! Old Age is creeping in !! Grin
Old age, I'm blaming all those hamburgers i ate during the 80's, BSE if my legs start giving way i will really start worrying ;D
The TPU was also supposedly known to run hot, strange thing is copper enamelled wire can be good up to 250 Degrees C or less depends on type.
That fiberglass sleeving is good up to 600 Degrees C although i would not think he would go above the enamel temp without clouds of smoke and a nasty smell, it cant be for extra insulation because the sleeving is woven.
Maybe he wants to run the core cold and wire hot, and is using the sleeving as a sort of high temp insulation
Can you explain why my R3 on Akula circuit burned ?
I have not been following this too much, which post is the circuit diagram?
Dear Peter.
Here is a replacement !!
Cheers Grum.
OK i did post a reply, but then realized i was wrong, so i will try again.
I am not sure how this circuit is working, my theory is not good enough with this circuit.
If your circuit is built correctly, no wrong connections then i would say the resistor is too low wattage, so what wattage are you using?
Is there an image of the PCB where this resistor is located, it may give you an idea what wattage it's meant to be.
I notice the designer has used 2 lines in the diagram for some resistors, 1 line for others and no lines also.
So no lines will be standard resistor wattage IE 1/4 watt
1 Line maybe 1-3 watts at a guess
2 lines must be higher, i am not aware of this drawing convention of lines to denote wattage.
A pcb picture may reveal the answer.
Try a 10Watt resistor, it's running temperature will give you an idea of power dissipation and allow you to refine it's wattage
Dear All.
Short video showing how the manipulation of the air gap alters the output waveform.
https://www.youtube.com/watch?v=u2ThpfhQqTQ
Let's see if we can improve to a ring up ??
Cheers Grum.
I used 2W resistor here and 0.22 I replaced by 0.18ohm 2W also. Anyway, the strange thing is that they heat up very fast and generate smoke while I measured max 4 amps from my 12V battery. Mosfet is cold when 0.18ohm resistor is used but when replaced by short thick wire mosfet and diodes heats a lot.
All that imho is not important because I used low wattage load (10 ordinary leds from garden solar lamps) but I was surprised to see it.
I'm not sure what the fet switching is doing but you could have pk currents of 10's of amps and R3 seems to be an energy recovery method, i would definitely go for 10watt or even higher until good measurements can be done, this probably explains the high temp fiberglass sleeving.
I think the circuit is very interesting, you could scope across R3 and try to get a mean average of the voltage, the average mean power will then be calculated by P=V*V/R or V^2/R same thing.
I see on the circuit diagram that the transformer has a thick black line and some Russians writting, this looks like some sort of screening built into the transformer, you need to find out how that is made, if it's a screen it may slow down the flux changing in the core, a bit like dropping a magnet down a copper tube and the magnet falls slowly.
Алло (Hallo) Hello, Gentlemen,
Some off-topic -should I dare to say 'as-usual'- considerations?
In most of case, in the average life: You succeed. You are rewarded (often) with money.
You carry off the prize.
Could we imagine another way of 'behavio(u)ring'?
What I did with my only son when he was a bit more than 18:
I told him: You have to pass the French "baccalauréat" ( =/= General Certificate of Education).
I will offer you -without condition and before the examination- such amount of money to buy a car.
A small brand-new one or second-hand bigger one. That will be your choice.
Why? Because I thrust you. Evidently, I would prefer an examination success. He succeeded.
IMHO, it is not: If S then R. If Success then Reward.
Could it not be the contrary? : If Reward then Success.
Please, think about the first US pioneers/colonists. They were given land for free. Were they not?
Just because they were considered as being able to thrive from this free land and
to expand the economy. They, also, were not asked for any result.
Cheers,
Jean
@NerzhDishual
Not land for free. Most had to work for x amount of years for a current land owner to then get their allocation.
hehehe
While my two boys were in school, from grade 1 onward, each subject on the tri or bi yearly report card showed their percentage plus the class average. I would take their score, minus the average score and pay them 2 bucks per percent higher or reduce 2 bucks per percent lower then pay them final result. They would generally make over a 100 bucks each. This really got their interest and they never failed a course. This puts a direct link on the impact between their efforts and rewards. Best money I ever spent.
wattsup
Dear All.
I am writing to report a serious error I have just found with the linear ring signal !!
It would appear that it was noise from my switch mode PSU that was being manipulated and not the air gap as first thought. I have deleted the video in question and would like to offer my sincerest apologies for the mistake.
So it's back to the drawing board !!??
Cheers Grum.
Quote from: Grumage on 2014.03.16, 19:18:38
Dear All.
I am writing to report a serious error I have just found with the linear ring signal !!
It would appear that it was noise from my switch mode PSU that was being manipulated and not the air gap as first thought. I have deleted the video in question and would like to offer my sincerest apologies for the mistake.
So it's back to the drawing board !!??
Cheers Grum.
No worries; all good!
You found a problem, that's great O0
It's not easy publishing work when working on the unknown, the only other option is to post nothing and then no one would get anywhere.
So my advice is keep making mistakes so we can all learn and build on what does work out
Thanks for your time & work Grum O0
Hi Grumage,
i bought the same core/holder as you have and trying to wind the coils as shown in your drawing.
But there is a problem as:
first of all, the picture shows 2 wires ending on each side, and not as your drawing 3 at one end, and 1 at the other end.
Secondly, i am using 0.75mm wire (like the diagram), and i am unable to lay 30 turns as the first layer for L2, it just won't fit.
Max. is about 16 which seems more logical as we can see that the 2e layer of this L2 coil and the 1st layer of L1
(both 8 turns = total 16 turns) fills up the second layer.
My conclusion is that the first 2 layers of L2 should be forth AND BACK (2 x 16 = 32 turns), then continue as normal.
This will bring 2 ends of the coils to 2 sides of the core (like the picture) and it better fits the number of turns.
Any thoughts?
Regards Itsu
Quote from: Itsu on 2014.03.16, 22:07:30
Hi Grumage,
i bought the same core/holder as you have and trying to wind the coils as shown in your drawing.
But there is a problem as:
first of all, the picture shows 2 wires ending on each side, and not as your drawing 3 at one end, and 1 at the other end.
Secondly, i am using 0.75mm wire (like the diagram), and i am unable to lay 30 turns as the first layer for L2, it just won't fit.
Max. is about 16 which seems more logical as we can see that the 2e layer of this L2 coil and the 1st layer of L1
(both 8 turns = total 16 turns) fills up the second layer.
My conclusion is that the first 2 layers of L2 should be forth AND BACK (2 x 16 = 32 turns), then continue as normal.
This will bring 2 ends of the coils to 2 sides of the core (like the picture) and it better fits the number of turns.
Any thoughts?
Regards Itsu
Dear Itsu.
I have re read T-1000's suggestion and yes you are fully correct. 15 turns forward then 15 back that makes two full layers and then an 8 to the halfway line and 7 back. At the opposite end it is just an 8 turn to the halfway and 7 back but wound in the opposite direction, I think ??
I am really glad to read that you are having a "bash" at this one. I shall rewind and we can compare notes !! O0
Cheers Grum.
Quote from: Itsu on 2014.03.16, 22:07:30
Hi Grumage,
i bought the same core/holder as you have and trying to wind the coils as shown in your drawing.
But there is a problem as:
first of all, the picture shows 2 wires ending on each side, and not as your drawing 3 at one end, and 1 at the other end.
Secondly, i am using 0.75mm wire (like the diagram), and i am unable to lay 30 turns as the first layer for L2, it just won't fit.
Max. is about 16 which seems more logical as we can see that the 2e layer of this L2 coil and the 1st layer of L1
(both 8 turns = total 16 turns) fills up the second layer.
My conclusion is that the first 2 layers of L2 should be forth AND BACK (2 x 16 = 32 turns), then continue as normal.
This will bring 2 ends of the coils to 2 sides of the core (like the picture) and it better fits the number of turns.
Any thoughts?
Regards Itsu
I think your logic makes sense. Good work! Please post pictures. Are you going to use copper foil? If so on the core or on the outside of windings?
Thanks
12v 1w LED
http://www.amazon.com/Lighting-EVER-Downlights-Voltage-Recessed/dp/B004BLKZ8Q/ref=sr_1_1?ie=UTF8&qid=1395027208&sr=8-1&keywords=12v+1w+led (http://www.amazon.com/Lighting-EVER-Downlights-Voltage-Recessed/dp/B004BLKZ8Q/ref=sr_1_1?ie=UTF8&qid=1395027208&sr=8-1&keywords=12v+1w+led)
12v 10w LED
http://www.amazon.com/LENBO-White-Waterproof-Outdoor-Lights/dp/B008XZAPV8/ref=sr_1_8?ie=UTF8&qid=1395027365&sr=8-8&keywords=12v+10w+led (http://www.amazon.com/LENBO-White-Waterproof-Outdoor-Lights/dp/B008XZAPV8/ref=sr_1_8?ie=UTF8&qid=1395027365&sr=8-8&keywords=12v+10w+led)
And this one :)
http://www.parts-express.com/portable-10w-cob-type-super-bright-led-work-light-rechargeable-flood-light-lamp-yellow--361-361 (http://www.parts-express.com/portable-10w-cob-type-super-bright-led-work-light-rechargeable-flood-light-lamp-yellow--361-361)
My pick for MOSFET:
Power MOSFET FDP025N06 265A 60V 0.0019Ω 395W x2
http://www.ebay.com/itm/Power-MOSFET-FDP025N06-265A-60V-0-0019-395W-x2-/181187045315?pt=LH_DefaultDomain_0&hash=item2a2f96e7c3 (http://www.ebay.com/itm/Power-MOSFET-FDP025N06-265A-60V-0-0019-395W-x2-/181187045315?pt=LH_DefaultDomain_0&hash=item2a2f96e7c3)
Datasheet:
http://www.fairchildsemi.com/ds/FD/FDP025N06.pdf (http://www.fairchildsemi.com/ds/FD/FDP025N06.pdf)
Quote from: 4Tesla on 2014.03.17, 03:53:16
My pick for MOSFET:
Power MOSFET FDP025N06 265A 60V 0.0019Ω 395W x2
http://www.ebay.com/itm/Power-MOSFET-FDP025N06-265A-60V-0-0019-395W-x2-/181187045315?pt=LH_DefaultDomain_0&hash=item2a2f96e7c3 (http://www.ebay.com/itm/Power-MOSFET-FDP025N06-265A-60V-0-0019-395W-x2-/181187045315?pt=LH_DefaultDomain_0&hash=item2a2f96e7c3)
Datasheet:
http://www.fairchildsemi.com/ds/FD/FDP025N06.pdf (http://www.fairchildsemi.com/ds/FD/FDP025N06.pdf)
Oh, that's a monster! Pity I cannot get it here cheaply, and Farchild does not provide sampels to Europe I think
Quote from: forest on 2014.03.17, 08:08:22
Oh, that's a monster! Pity I cannot get it here cheaply, and Farchild does not provide sampels to Europe I think
Yep it is. Do any of these sites work for you?
http://uk.farnell.com/fairchild-semiconductor/fdp025n06/mosfet-n-ch-60v-120a-to-220/dp/2083327 (http://uk.farnell.com/fairchild-semiconductor/fdp025n06/mosfet-n-ch-60v-120a-to-220/dp/2083327)
http://uk.rs-online.com/web/p/mosfet-transistors/7599664/ (http://uk.rs-online.com/web/p/mosfet-transistors/7599664/)
http://www.digikey.co.uk/product-detail/en/FDP025N06/FDP025N06-ND/1878806 (http://www.digikey.co.uk/product-detail/en/FDP025N06/FDP025N06-ND/1878806)
4Tesla
He he sure, but shipping cost kill me. No problem man, that's for taking time to help me O0 I have irf3205 and it works fine (but heats a lot if 0.22 resistor is omitted)
Quote from: 4Tesla on 2014.03.17, 03:08:50
I think your logic makes sense. Good work! Please post pictures. Are you going to use copper foil? If so on the core or on the outside of windings?
Thanks
Thanks.
I am also puzzled by this L2 (should it not been called L3?) coil.
Was somebody able to translate what it says next to it?
My thoughts are that it must be some kind of shielding around the L1/L2 coil, probably from aluminium or copper tape.
A single turn, just not touching its ends and connected to ground.
I don't think its inside the coil (around the core) as we would have seen the connection or wire coming out.
But this is just guessing.
A picture or video of the device working would reveal this.
Regards Itsu
Slightly different device, but in this video you'll see what appears to be foil, probably aluminum:
Quote from: Itsu on 2014.03.17, 12:18:58
Thanks.
I am also puzzled by this L2 (should it not been called L3?) coil.
Was somebody able to translate what it says next to it?
My thoughts are that it must be some kind of shielding around the L1/L2 coil, probably from aluminium or copper tape.
A single turn, just not touching its ends and connected to ground.
I don't think its inside the coil (around the core) as we would have seen the connection or wire coming out.
But this is just guessing.
A picture or video of the device working would reveal this.
Regards Itsu
I didn't notice this earlier but it was translated over on OU.. I guess it is aluminum on the outside.
OOPS both posted similar times
From the diagram it would appear it's L2 only that's shielded, someone needs to get maybe T1000 to translate the Russian text on the diagram which is related to this shielding if not already done
Quote from: Matt Watts on 2014.03.17, 14:44:53
Slightly different device, but in this video you'll see what appears to be foil, probably aluminum:
Thanks! Confirmed that may be what Akula (Roman) used.
Quote from: Peterae on 2014.03.17, 16:13:28
OOPS both posted similar times
From the diagram it would appear it's L2 only that's shielded, someone needs to get maybe T1000 to translate the Russian text on the diagram which is related to this shielding if not already done
Translated in the image I attached a few posts up. O0
I think L2 is referencing the coil and not the foil as it has number of turns and wire type listed under it.
So maybe we now know all about the transformer?? 8) O0
Edit:
I guess there are more questions...
Air gap in core and if so how much
L1/L2 not same size wire in picture, yet labeled same in schematic
Following Itsu's logic, this is what I imagine the winding of the transformer looks like. Sorry not good with graphics.
I think that he used .75mm (20 AWG) for L2 and 1mm (18 AWG) for L1.
This might be a stupid question, but stranded or solid wire?
QuoteI think L2 is referencing the coil and not the foil as it has number of turns and wire type listed under it.
Agreed but he has drawn the line to the right of L2 coil which to me means it's only L2 that screened otherwise he would have drawn the line to the left of L1 as well if both were screened separately.
Dear 4Tesla and all.
Perhaps you missed this post ?? http://www.overunityresearch.com/index.php?topic=2358.msg36789#msg36789
Yes I do believe that I had, at first misunderstood T-1000's coiling schedule. I will be rewinding later and then looking for anomalies.
Attached is version 2 !! O0
Cheers Grum.
Dear All.
There has been so much stuff posted on both threads that I now can't remember who suggested the coils might be Litz wire in a heat proof sheathing. ??? :)
I am sure they will step forward !!
Cheers Grum.
Quote from: Grumage on 2014.03.17, 17:47:55
Dear 4Tesla and all.
Perhaps you missed this post ?? http://www.overunityresearch.com/index.php?topic=2358.msg36789#msg36789
Yes I do believe that I had, at first misunderstood T-1000's coiling schedule. I will be rewinding later and then looking for anomalies.
Attached is version 2 !! O0
Cheers Grum.
Cool!! Your graphic a lot nicer than mine! :) 8)
Quote from: Peterae on 2014.03.17, 17:44:42
Agreed but he has drawn the line to the right of L2 coil which to me means it's only L2 that screened otherwise he would have drawn the line to the left of L1 as well if both were screened separately.
But L2 does go under L1.. so I think all windings?
Quote from: 4Tesla on 2014.03.17, 16:15:54
Translated in the image I attached a few posts up. O0
I think L2 is referencing the coil and not the foil as it has number of turns and wire type listed under it.
4Tesla and Peterae,
i have to agree with you, L2 is not the shield, but the larger (45t) coil, well spotted.
But how do we shield only this L2 coil as it is half covered by L1, only the the L2 part visible?
Questions. questions......
Regards Itsu
I wrapped a piece of aluminum foil almost all the way around the outside, on top of all the windings, but I do not let the 2 ends of the aluminum touch each other.
I think this aluminum is not really a shield but more like an antenna to pickup "something" being sent from the coils.
Its just my guess until I can prove it.
DonL
Quote from: Itsu on 2014.03.17, 18:18:44
4Tesla and Peterae,
i have to agree with you, L2 is not the shield, but the larger (45t) coil, well spotted.
But how do we shield only this L2 coil as it is half covered by L1, only the the L2 part visible?
Questions. questions......
Regards Itsu
I don't see any shielding in the picture.. so I think we have it down to two options.. either just the visible L2 or everything.
Quote from: dllabarre on 2014.03.17, 18:26:25
I wrapped a piece of aluminum foil almost all the way around the outside, on top of all the windings, but I do not let the 2 ends of the aluminum touch each other.
I think this aluminum is not really a shield but more like an antenna to pickup "something" being sent from the coils.
Its just my guess until I can prove it.
DonL
Thanks for posting! Welcome to the forum. O0
But isn't L1 on the left of the image and L2 on the right, then wrap the alu over the right half of the transformer.
Here's some food for thought, in Nuclear Acoustic Resonance during beta decay aluminium is used to shield against fast moving particles, but if heavier metals are used then it possible for the fast particles to cause Gamma radiation & xrays, i hope that's right anyway, Verpies warned against using heavier metals. O0
Also note that in the transformer picture that it appears that L1 looks thicker, this is not so, with L2 he has pulled the woven fiberglass sleeving tight to make it longer and have less diameter, he's probably done this to get more turns for L2
I would try to get some of that sleeving from an electrical shop because it will affect your windings spacing and therefore change the transformers operation quiet a bit. i think it's sold by the meter O0 , you will note on the diagram that both L1 & L2 are both using the same thickness wire.
It maybe worth videoing yourself building the transformer and put that up on youtube and post a link to it on Akula's feed, he may pipe in and say whats wrong.
QuoteIt maybe worth videoing yourself building the transformer and put that up on youtube and post a link to it on Akula's feed, he may pipe in and say whats wrong.
It would be cool if Akula would chime in, but he seems to be avoiding communication as I and others haven't received replies to emails and pms.
Well, I have to go. I'll try to follow on my phone.. I'll be back on my PC on the weekend.
Quote from: 4Tesla on 2014.03.17, 16:14:16
Thanks! Confirmed that may be what Akula (Roman) used.
I'm not sure of the varying characteristics between a Pot-Core and an E-Core, but something to note is the foil isn't the width of the entire core and is centered over what would be the gap. It's also not clear whether this foil makes a complete loop or not, but in this circuit it does connect to ground, denoted as X4 in the attached schematic.
Quote from: dllabarre on 2014.03.17, 18:26:25
I wrapped a piece of aluminum foil almost all the way around the outside, on top of all the windings, but I do not let the 2 ends of the aluminum touch each other.
I think this aluminum is not really a shield but more like an antenna to pickup "something" being sent from the coils.
Its just my guess until I can prove it.
Added capacitance centered on the gap... Could it be electrons are being ejected in that particular location and funneled to ground? Or the other way around, the foil is a negative capacitor plate causing positive charge to build up in the gap? It's there for a reason and probably important. If one of us can get a working device, we can build replicas and found out.
Quote from: Grumage on 2014.03.17, 17:53:10
There has been so much stuff posted on both threads that I now can't remember who suggested the coils might be Litz wire in a heat proof sheathing. ??? :)
Grum, can you tell if this is a solder scorching of the insulation or just rosin residue?
Quote from: Matt Watts on 2014.03.17, 19:47:49
I'm not sure of the varying characteristics between a Pot-Core and an E-Core, but something to note is the foil isn't the width of the entire core and is centered over what would be the gap. It's also not clear whether this foil makes a complete loop or not, but in this circuit it does connect to ground, denoted as X4 in the attached schematic.
Added capacitance centered on the gap... Could it be electrons are being ejected in that particular location and funneled to ground? Or the other way around, the foil is a negative capacitor plate causing positive charge to build up in the gap? It's there for a reason and probably important. If one of us can get a working device, we can build replicas and found out.
I just noticed you're new to the forum. Welcome to the forum! O0
Shopping for transformer supplies..
ETD49 Bobbin
http://www.ebay.com/itm/Bobbin-Coil-Former-for-ETD49-core-Epcos-Germany-20pin-/181078749380?pt=LH_DefaultDomain_0&hash=item2a292270c4 (http://www.ebay.com/itm/Bobbin-Coil-Former-for-ETD49-core-Epcos-Germany-20pin-/181078749380?pt=LH_DefaultDomain_0&hash=item2a292270c4)
ETD49 Cores
http://www.ebay.com/itm/ETD49-Transformer-Cores-Power-Ferrite-500mT-Bsat-x2-/181078743358?pt=LH_DefaultDomain_0&hash=item2a2922593e (http://www.ebay.com/itm/ETD49-Transformer-Cores-Power-Ferrite-500mT-Bsat-x2-/181078743358?pt=LH_DefaultDomain_0&hash=item2a2922593e)
32mm Aluminum Shield (I don't know if I trust this supplier but don't see any negative reviews)
http://www.chipsetsupplier.com/htm/729/129-729-1141-32mm%20x%2040%20meters%20x0.1mm%20Adhesive%20Aluminum%20Foil%20Papper%20Tape,%20High%20Temperature%20Withstand,%20EMI%20Shielding,%20Waterproof.htm (http://www.chipsetsupplier.com/htm/729/129-729-1141-32mm%20x%2040%20meters%20x0.1mm%20Adhesive%20Aluminum%20Foil%20Papper%20Tape,%20High%20Temperature%20Withstand,%20EMI%20Shielding,%20Waterproof.htm)
20 AWG (.75mm) Red Stranded Wire for L2
http://www.ebay.com/itm/20-AWG-Copper-Hook-Up-Wire-Red-Lead-Wire-100ft-Length-600-Volt-Stranded-/321267925529?pt=LH_DefaultDomain_0&hash=item4acd0f8219 (http://www.ebay.com/itm/20-AWG-Copper-Hook-Up-Wire-Red-Lead-Wire-100ft-Length-600-Volt-Stranded-/321267925529?pt=LH_DefaultDomain_0&hash=item4acd0f8219)
18 AWG (1mm) Blue Stranded Wire for L1 (slightly larger as shown in picture)
http://www.ebay.com/itm/18-AWG-Copper-Hook-Up-Wire-Blue-Lead-Wire-50ft-Length-600-Volt-Stranded-/321159876551?pt=LH_DefaultDomain_0&hash=item4ac69ecfc7 (http://www.ebay.com/itm/18-AWG-Copper-Hook-Up-Wire-Blue-Lead-Wire-50ft-Length-600-Volt-Stranded-/321159876551?pt=LH_DefaultDomain_0&hash=item4ac69ecfc7)
Even found the wire in the colors of transformer diagram :)
I haven't ordered anything yet.. does this all look like correct supplies to you all?
Thanks!
This document describes the importance of the gap between the two pieces of ferrite. I believe it was originally authored by Andrei Melnichenko and translated.
Quote from: Grumage on 2014.03.17, 17:53:10
Dear All.
There has been so much stuff posted on both threads that I now can't remember who suggested the coils might be Litz wire in a heat proof sheathing. ??? :)
I am sure they will step forward !!
Cheers Grum.
@Grumage
Yes, it looks like Litz wire because of that black residue on the solder. Very common because of all the wire coating that melts and carbonizes during the solder process. Litz is very often used as a primary coil.
Man all this information should have been available from Akula so why are all the questions still pending when he could have resolved them all in a few sentences. Maybe someone needs to contact Wesley.
Ultimately, if there is no direct information from Akula on this, you will have to test a second transformer wound normal like I have mentioned, 15 turns on left half, 45 turns on right half, keep the gap free, otherwise the if you are basing your results on only this @T1000 method, then you may be working in the wrong direction from the start.
wattsup
Quote from: Matt Watts on 2014.03.17, 20:21:30
Grum, can you tell if this is a solder scorching of the insulation or just rosin residue?
Dear Matt.
A good question !! I have also noticed how, as with each presentation the picture quality gets worse !! I think Peterae's analysis is probably the best, ECW in a heat proof sheath the one being a little stretched to cover the greater amount of wire.
BUT, I seem to remember when the photo first appeared, seeing what appeared to be multi strands of tinned copper at the termination !! So it could also be Litz wire !! Which is ideally suited to high frequency applications and, more particularly, TPU's !!
Itsu and I spent an interesting hour via Skype last night comparing notes on not quite identical transformers !! I ordered non gapped Ferrites, Itsu 2mm gapped Ferrites. The difference in inductance was huge !! milli Henries for me, micro Henries for Itsu. Both bobbins were wound to the revised schedule I posted yesterday.
There would be more but I managed to fry my Mosfet !! But a point of note. Using my improved PWM from RMC (Non Mosfet) very different scope shots were seen !! Noise might play a big part in this device !!
Cheers Grum.
Litz wire or just stranded wire? I did notice many very fine copper strands. It can be seen in original closeup on page one. My conclusion was just stranded wire, but maybe it is litz. Also, L2 solid, stranded, or litz? Does anyone know of good suppliers for cores and litz wire?
If it is litz we know it is about 1mm total thickness and very fine strands.
What type of litz has been used in these devices? There are 8 types.
Well I fired-up DipTrace and prepared a test circuit based off the schematic we currently have. We should be able to evaluate various transformer designs. If this looks useful, give it a good inspection for obvious errors and I'll correct them.
Keep in mind the start switch is momentary. If voltage across C11 bleeds off immediately, something isn't right. I would think using a low frequency, low duty cycle signal source at first would be best. Little impulses should show the effect we are after. Ease things up gently. If you happen to get a good wind on the test coil with a proper gap on the core, it may take off and self destruct quickly. Be ready to kill the signal source if you see the voltmeter across C11 jump up beyond 12 volts.
I would have to guess R2 and C4 need to match the test transformer so the Akula values would only be a starting point. I'd like to try little larger values for R1 and R3 to keep things from blowing up, but I'm just not sure whether the transformer will respond by doing that.
It is going to be tricky to work with a looped system like this because there are no independent variables that I can see. Everything seems to depend on something upstream of it and around the loop we go.
I found this on litz wire types. I'm assuming Type 1 is what has been used in these devices (kapa/tpu).
http://www.litzwire.com/litz_types.htm (http://www.litzwire.com/litz_types.htm)
Quote from: Matt Watts on 2014.03.18, 14:58:19
Well I fired-up DipTrace and prepared a test circuit based off the schematic we currently have. We should be able to evaluate various transformer designs.... If this looks useful, give it a good look over for obvious errors and I'll correct them.
Nice work! 8)
Quote from: 4Tesla on 2014.03.18, 14:47:55
Litz wire or just stranded wire? I did notice many very fine copper strands. It can be see in original closeup on page one. My conclusion was just stranded wire, but maybe it is litz. Also, L2 solid, stranded, or litz? Does anyone know of good suppliers for cores and litz wire?
If it is litz we know it is about 1mm total thickness and very fine strands.
Dear 4Tesla.
I am providing a link to a UK based firm that supplied both Itsu and myself with the cores and bobbin. Itsu told me they were rally fast with their despatch and postal costs !!
http://powermagnetics.co.uk/calculator From their very handy calculator you can move to the shop and find the following items.
Ferrite ETD core (ungapped). ETD49-CF139 Bobbin ETD49-90794-20P
Manufacturer: Cosmo Ferrites.
Material Grade CF139 (Equivalent to Cosmo CF138, Ferroxcube 3C90 and TDK PC44).
Price is for a half core - (I.e. you will need to order 2 pieces to form a pair)
Dimensions:
Length: 48.5mm
Width: 16.7mm
Height: 24.9mm
Wt: 58g
Magnetic Characteristics:
uiac = 2100
AL Value = 4100
(+ 30% / -20%)
RichMo from RM Cybernetics also supplies Litz wire to order. http://www.rmcybernetics.com/shop/index.php?route=product/product&product_id=160
I hope the above is of some help??
Cheers Grum.
Hi all,
If it is litz wire I'm guessing it is 10/30 or 15/30 (AWG) Type 1
30 AWG = .25 MM
http://www.litz-wire.com/New%20PDFs/Litz_Config/Litz_Magnet_Wire_AWG_28-37_Circular_Mil_Feet_per_Pound.pdf (http://www.litz-wire.com/New%20PDFs/Litz_Config/Litz_Magnet_Wire_AWG_28-37_Circular_Mil_Feet_per_Pound.pdf)
Looks like we'll have to try all configs.. adding to my list to buy if I can find.
Quote from: Grumage on 2014.03.18, 15:15:50
Dear 4Tesla.
I am providing a link to a UK based firm that supplied both Itsu and myself with the cores and bobbin. Itsu told me they were rally fast with their despatch and postal costs !!
http://powermagnetics.co.uk/calculator From their very handy calculator you can move to the shop and find the following items.
Ferrite ETD core (ungapped). ETD49-CF139 Bobbin ETD49-90794-20P
Manufacturer: Cosmo Ferrites.
Material Grade CF139 (Equivalent to Cosmo CF138, Ferroxcube 3C90 and TDK PC44).
Price is for a half core - (I.e. you will need to order 2 pieces to form a pair)
Dimensions:
Length: 48.5mm
Width: 16.7mm
Height: 24.9mm
Wt: 58g
Magnetic Characteristics:
uiac = 2100
AL Value = 4100
(+ 30% / -20%)
RichMo from RM Cybernetics also supplies Litz wire to order. http://www.rmcybernetics.com/shop/index.php?route=product/product&product_id=160
I hope the above is of some help??
Cheers Grum.
Thank you Grum! Do you know if http://powermagnetics.co.uk ships
internally internationally? I'm in US.
Quote from: 4Tesla on 2014.03.18, 15:34:01
Thank you Grum! Do you know if http://powermagnetics.co.uk ships internally? I'm in US.
Dear 4Tesla.
I assume you meant internationally ?? As I certainly would not want a CF139 fitted internally !! :D ;D
Cheers Grum.
@all
I think while you guys are working on that device, maybe a good EEer can identify a zener diode that could be added before and/or after the mosfet to bleed any juice above a critical level. Bleed it into a tanked cap with a led will show when the system is in a mosfet critical mode.
Litz serving of Type1 with total size anywhere from 18 to 21 AWG. These can have hundreds of strands of fine wire. If you try to find some, make sure it is solderable. Many sellers on ebay will sell small lengths.
The only question I have is about the windings on the E core. You guys are fine with the @T1000 method but no one has provided any theory as to why this should work or even a link to a web site that shows such a gap overlapping as a standard or preferred method. Did @T1000 get this confirmed by Akula or is this simply his gut feeling. Important questions should be put up front before more weeks go by. If it is a gut feeling that @T1000 has, then you may as well say you are working blind. If you are working blind, then you have to feel your way forward and try all possibilities meaning when you buy or make an E core, make two, one like @T1000 and one like the standard method. Then test them both, otherwise, I get this crazy feeling that Murphies Law is getting ready to make another move and more mosfets will be blown because that core does not respond well to an overlapped primary.
wattsup
Quote from: Grumage on 2014.03.18, 15:42:23
Dear 4Tesla.
I assume you meant internationally ?? As I certainly would not want a CF139 fitted internally !! :D ;D
Cheers Grum.
Yep internationally.. lol
Dear Wattsup.
Wise words from an OU veteran !! O0 Thank you !!
I will try to speak with T-1000 later. He is basing this assumption on the work presented by GeoFusion and the physical way his transformer was wound. I too noticed an anomaly here. https://www.youtube.com/watch?v=SPV89JWVdtY
Also with the secondary winding placed across the split in the core !!
I have attached a photo of the latest transformer wind !!
Cheers Grum.
Hi all,
I found this litz wire that might work.
http://www.aliexpress.com/item/0-1X100-strands-10m-pc-Litz-wire-stranded-braided-enamelled-copper-wire-woven-wire/1685017155.html (http://www.aliexpress.com/item/0-1X100-strands-10m-pc-Litz-wire-stranded-braided-enamelled-copper-wire-woven-wire/1685017155.html)
I find this conversion table useful:
http://www.engineeringtoolbox.com/awg-wire-gauge-d_731.html (http://www.engineeringtoolbox.com/awg-wire-gauge-d_731.html)
Also if solid wire is used.
http://www.ebay.com/itm/like/171261352963?lpid=82 (http://www.ebay.com/itm/like/171261352963?lpid=82)
Hi,
As I mentioned before, the secret is in way transformer is made, not in electronics.
If you attach standard transformer there won't be anomalies at all.
Also if you have analytical mind and gone over all Free Energy devices claims there are coming things what can be groupped and may be put as assumption to not explained devices including the one is this thread about.
So few things I can list here for deeper understanding where to look for clues:
1) E. Leedskalnin approach to magnetism and his configuraitons on magnetic fields - http://www.alexpetty.com/index.php/2011/06/01/flux-flow-analysis-of-the-edward-leedskalnin-flywheel/
2) Coil shorting - that is from N. Tesla time: http://home.dmv.com/~tbastian/images/CorumDesign.gif
And it is also in Aviso car secret: http://www.youtube.com/watch?v=wC6aglJ6a_8
That one is most likely utilized on thread circuit in unique approach for making current spikes in series of BEMF with energy recovery.
3) Asymetric magnets with moving bloch wall:
Floyd Sweet FE device - http://www.cheniere.org/misc/sweet.htm and Bearden's explanation on it in http://www.youtube.com/watch?v=QyO2wWsToKk&t=2m18s
So it takes only determination to achieve results with knowledge available from other people...
Good luck! :)
Thanks T-1000 for the links! Saturating the core causes a shorting effect and is key to device? Do you have any input on build of transformer?
Thanks.
I thought this was a good post from lost_bro over at OU.com
Quote
Good evening Grumage:
OK, in relation to the really HOT!!! MOSFET; this is some info that I use when winding GDT's:
Saturation flux density is how much magnetic flux the magnetic core can handle before becoming saturated and not able to hold any more. This depends on several factors including ferrite type, temperature and electrical and magnetic conditions on the transformer.
When the ferrite saturates, the transformer no longer acts like an inductor with a linear increase in current over time. Rather, the magnetic field cannot increase further and current is limited by the source impedance of the power supply and the resistance of the transformer wire.
This leads to very large currents- Very HOT MOSFETs and blown devices. This is a Bad Thing.
Ferrite typically saturates at a flux density (B) about 0.3 Tesla but this depends on temperature and the ferrite type. A typical design might have as a target a B = 0.25T at 125ºC which gives plenty of operating margin to the limits.
The formula for flux density is
B = ( V x t ) / ( N x Ae )
where:
B = flux denstiy in Tesla
V = applied Voltage to the winding in Volts
t = time that Volts V is applied for in seconds
Ae = cross sectional area of the core in m^2 - obtainable from the core datasheet
If this were an inductor carrying a DC component, the DC current would play a significant part in calculating the saturation.
I usually use a DC blocking capacitor in the primary side of the GDT to eliminate the DC component and thus avoid alot of the issues involved (flux walking).
The main thing to take from this equation is that for a given core and frequency there is a minimum number of turns that can be used without saturating the core. So, to avoid saturation, use more than the minimum number of turns.
So I have wound many GDTs that have overheated the driver circuit just for the above mentioned fact: When the core saturates, it acts as a DEAD short and MOSFETS don't like that..... yes it is a balancing act between minimizing the leakage inductance, the minimum number of windings permitted and core saturation and of course maintaining an acceptable wave form...
So looking at the plan for the winds on the XFMR, could be that for the type of Ferrite involved, you don't have sufficient windings on it to avoid saturation.
Just a thought.
Take care, peace
lost_bro
Quote from: 4Tesla on 2014.03.19, 00:22:38
I thought this was a good post from lost_bro over at OU.com
Interesting. So with this:
B = ( V x t ) / ( N x Ae )
B = flux denstiy in Tesla
V = applied Voltage to the winding in Volts
t = time that Volts V is applied for in seconds
N = number of turns
Ae = cross sectional area of the core in m^2 - obtainable from the core datasheet
So to avoid saturation, we can reduce voltage, reduce duty cycle, increase the number of turns or use a heftier core. Seems like the first two options would be the easiest to manipulate.
I enhanced the photos on the schematic. This is about as clear as I can get them.
Dear All.
I spent another fruitless evening looking for anomalies, nothing to report.
If I have correctly read the schematic, the Mosfet switches in series with the transformer and the load is fed after the primary winding?? In other words current is drawn through L1 to light the LED's ?? So could this be why I am not seeing anything unusual ?? The simple reason being I am not pulling enough current through L1 to create a strong magnetic field ??
For the EE's, can I use my PWM connected as in figure 4 from the attached PDF ?? To create the condition I have explained above ??
Thought or comments gratefully received.
Cheers Grum.
Grumage
IMHO The circuit seems to regenerate or capture energy to feed itself, if possible i would try to stick with the prescribed circuit otherwise how can you tell if it's feeding itself and creating the correct drive.
Quote from: Peterae on 2014.03.19, 13:23:42
Grumage
IMHO The circuit seems to regenerate or capture energy to feed itself, if possible i would try to stick with the prescribed circuit otherwise how can you tell if it's feeding itself and creating the correct drive.
Dear Peterae.
Thank you for your comments. I was wondering if I could use the PWM in that way to see if there were any anomalous signals on L2 by driving a load. Do you think it is possible ??
Cheers Grum.
Dear All.
A guy called MenofFather over on OU.C just posted this.
"Do like in this picture and you, I think, have bigest chance to make this divice). This schematic is from author.
He say, that first wound primary coil of 15 turns, turn to turn wound, then piesce of paper and wound secondary coil 45 turns, also turn to turn. Coils wound like show in schematic in oposite directions.
Also not wery clearly say, but something like this, that after primary coil is short turn of ecran (one turn of tape of specific metal, maybe cuper is good) and after secondary coil is short turn of ecran. So wound primary, then tape of paper, then ecran short, then secondary, then ecran short. Load 5-20 W LED 12 V. "
Attached was this schematic.
Cheers Grum.
So primary wind first and secondary surrounded upper and lower with foil. Can we suspect only the outer foil is grounded and the inner foil is floating? Also, can we assume the foil is a complete loop? Or does "short turn" mean less than a complete loop?
Did a little search for ecran film and it appears to be some sort of protective material for touch screen devices. So I don't know if this is actually metal or what it is. Very likely it is clear so it would be very difficult to spot when wrapped on a transformer core.
QuoteDear Peterae.
Thank you for your comments. I was wondering if I could use the PWM in that way to see if there were any anomalous signals on L2 by driving a load. Do you think it is possible ??
Cheers Grum.
Without knowing what the effect we are looking for it's hard to say and because of this i would try to stick with the intended circuit if possible.
Did you get a PCB made up of the circuit or is that in progress.
Quote from: Peterae on 2014.03.19, 15:26:33
Without knowing what the effect we are looking for it's hard to say and because of this i would try to stick with the intended circuit if possible.
Did you get a PCB made up of the circuit or is that in progress.
Dear Peterae.
The lovely people at Farnell are dragging their feet !! Perhaps my little order of 10 set's of components is not large enough to make them move faster !! So at this moment in time we are still waiting for a quote, from this we can then get an overall price for a ready built PCB.
Looking at it another way the finished circuit can always be used as a PWM, with of course a handy reclamation circuit built in !! :)
Cheers Grum.
Quote from: Grumage on 2014.03.19, 14:16:40
Dear All.
A guy called MenofFather over on OU.C just posted this.
"Do like in this picture and you, I think, have bigest chance to make this divice). This schematic is from author.
He say, that first wound primary coil of 15 turns, turn to turn wound, then piesce of paper and wound secondary coil 45 turns, also turn to turn. Coils wound like show in schematic in oposite directions.
Also not wery clearly say, but something like this, that after primary coil is short turn of ecran (one turn of tape of specific metal, maybe cuper is good) and after secondary coil is short turn of ecran. So wound primary, then tape of paper, then ecran short, then secondary, then ecran short. Load 5-20 W LED 12 V. "
Attached was this schematic.
Cheers Grum.
Perhaps he could draw a graphic. Where did he get his info? Did he talk to Roman.. Many others, including myself, never got a reply.
IMO.. your v2 graphic is what it looks like in the photo.
Dear All.
MenofFather has just placed this on OU.C.
Cheers Grum.
Grum
Would you like me to do the parts list from farnell, do you have a parts list ;)
Quote from: Grumage on 2014.03.19, 17:54:49
Dear All.
MenofFather has just placed this on OU.C.
Cheers Grum.
Not the same as in photo, but still might be worth trying.
Dear All.
I have just finished speaking with T-1000.
Direct translation of Russian text alongside L2. " ONE TURN NOT SHORTED STANDARD SHEILD " Ecran being Russian for shield !! Not a material !!
Shorting this, as I have proved for myself, just creates a massive current draw !! T-1000 also stated that we should be looking at the creation of noise, just like my findings with the negative only connection from my PSU creating the non ringdown !! ;)
This is the way forward !!
Cheers Grum.
Quote from: Peterae on 2014.03.19, 18:30:54
Grum
Would you like me to do the parts list from farnell, do you have a parts list ;)
I have this list, not positive of its accuracy.
Quote from: Grumage on 2014.03.19, 20:04:20
Direct translation of Russian text alongside L2. " ONE TURN NOT SHORTED STANDARD SHEILD " Ecran being Russian for shield !! Not a material !!
Great detective work Grum! O0
Great work Grumage thanks to T1000 for the help.
I will work on the parts list tomorrow O0
Dear All.
Thanks' guys. Peterae., I have sent you two emails.
Now to business. The way to go is by driving the transformer with the load applied !! I am now seeing very different output signals.
Keep on trucking !!
Cheers Grum.
@Itsu
I was doing a current trace of the 1W circuit and I believe that the first version was an open circuit. It was missing a resistor. I circled here. Did you get LEDs to light? I can't see how in the first version.
Edit: I've run out of time.. got to go again, but I noticed there may be a path through VT2, the PNP.. now I got to start over when I get a chance.
On phone.. Otherwise I would quote very good post with great video by Les over at ou. Worth checking out.. Maybe someone could copy to here. 8)
Les could have been more thoughtful.. I don't always agree with all posts.. But don't call them out unless a troll.. Feel bad for MenofFather.
Quote from: 4Tesla on 2014.03.19, 23:23:11
@Itsu
I was doing a current trace of the 1W circuit and I believe that the first version was an open circuit. It was missing a resistor. I circled here. Did you get LEDs to light? I can't see how in the first version.
Edit: I've run out of time.. got to go again, but I noticed there may be a path through VT2, the PNP.. now I got to start over when I get a chance.
Hi 4Tesla,
yes i have seen that new diagram, and indeed in my test setup this resistor was missing.
I never was able to get any lights in the leds (no oscillations), only when i hooked up my FG and was sweeping through the frequency the leds would light up now and then,
but powered by my FG.
I will configure the circuit according to this new diagram and see if this resistor will make any difference.
I notice that also the transformer is different, it has another 4th coil to it!
That makes it more complicated.
Thanks, regards Itsu
Dear All.
As requested. Copy and paste of Les Banki's post from OU.C.
Cheers Grum.
Hi guys,
I have been monitoring this thread from start.
Here is what I see:
The 'Akula0083 circuit' you are trying to duplicate is essentially a simple device which
(for simplicity), can be divided into two (2) sections:
a. Control ("exciter")
b. "Free energy"
I comment mainly on the control section which you have thoroughly messed up!
Admit it or not, most of you have contributed to this.
I am particularly critical of some of the 'technical' comments made!
They are WAY OFF!
OK.
The control section is based on the TL494 PWM IC which has been around for over 30 years.
It became an industry standard, so to speak.
In the past, most computer SMPS used this IC.
It is well understood and works well.
If you care to watch (and pay attention!) to the video where Akula explains all three (3) versions of his MOSFET drivers, you would not be confused and would stop 'butchering' the circuit!
For example, arguments are still raging about that Ground wire connection to pin 16 of TL494.
Know this:
If you remove that connection (like I have seen in some diagrams in this thread), you will also lose the Ground connection to the indicator LED HL11 AND the Ground connection to voltage divider R9-R11-R10 which will result in error amplifier A1+ (pin 1) will be at the supply voltage (+12V) potential trough R9 and part of R11 (pot)!
This has already been pointed out by one poster ('lost_bro' on page 6) and yet, even as I write this, yet another diagram appeared ("Akmodedit.png") where that Ground connection is MISSING!
This diagram has been REPEATEDLY posted!
In other words:
By removing that Ground connection in the diagram, you DISABLE not only the indicator LED but also DISABLE BOTH error amplifiers, A1 & A2!
If you pay proper attention to the details in this video:
http://www.youtube.com/watch feature=player_embedded&v=EwCS15pRtH0&list=PLC7684829E98CAD74
you will see that the disputed Ground link (connection) is PRESENT in all 3 versions of the diagram.
No, you don't even have to understand the language!
By the way, the author announced a few days ago that English subtitles are coming soon!
Then there is one other point which everyone have missed, including Akula it seems!
Irrespective of which (of the 3) MOSFET drivers you use, there is only ONE MOSFET in this device and the drive is single-ended.
Period.
Yet, the output logic control (pin13) is tied to pin 14 which is the reference voltage. (+5V, ±5%)
Therefore, the output stage is operating in push-pull mode.
(Only one poster commented on this, in RED, written over the diagram.)
Sure, it still works this way BUT the output frequency is only HALF the oscillator frequency.
If pin 13 was connected to Ground (0V), the output would operate in the single-ended mode which gives the option to parallel the two output transistors, thus gaining significant current drive to the MOSFET, either directly or through the complementary NPN/PNP driver, or through the dedicated Mosfet driver IC (IXDD609 or similar).
Lowest rise/fall times are achieved with dedicated Mosfet drivers (they are MOSFET based).
However, unless I need super fast switching, I prefer the complementary npn/pnp driver since I found that the MOSFET drivers are too "touchy" for experimental work (meaning they are too easy to blow up!).
To those who intend to try to duplicate this circuit, my advice is:
Make sure the control section works properly before attempting to experiment with the free energy section (which the transformer is part of).
Also, don't fall for all that nonsense (yes, nonsense!) about input/output power measurements as some posters promote!
Why do I make such a statement?
Simple.
You are NOT dealing with a device with a claim of marginal OU in the order of a few percent which could perhaps be a measurement error!
If you want to know the input power to this device, I can give it to you right now:
The TL494 IC, running on a 15V supply has a maximum current draw of 10 mA.
(average 7.5 mA)
That translates to 150 mW (0.15 W). (maximum)
Add the 12V regulator's power consumption of approx. 96 mW (maximum) and you have a
TOTAL of 246 mW (0.246 W)
Now, just for the hell of it, DOUBLE it and round it to 0.5 W.
However, this device will produce 30 W (or more) depending on the LOAD and the transformer
you use.
I hope you can see the stupidity of those who blindly INSIST on input/output measurements!
Don't get me wrong.
Sure, you will be curious (as you should be!) about the output power of your device!
With devices like these, you may face a very different kind of problem!
Avalanche. Yes, avalanche.
I hope you know what that is.
(It is a runaway condition which, if not controlled, can cause DESTRUCTION of great magnitude.)
Don't believe it?
Fine. But don't complain if something blows up in your face! (or worse)
Oh...'MenofFather' posted suggestions to the winding method, including SHORTED turns!
Guys, with all due respect, with that kind of suggestions you will never get anywhere!
Guaranteed!
Cheers,
Les Banki
A little more playing with DipTrace. Probably not quite as good as GroundLoop did with Eagle, but I think it would run.
Took me a while to find components and make patterns so that everything fits correctly.
Again, this is just a test circuit for the transformer; you would have to provide your own signal input. I will say though, if you find the magic touch, this board would easily power an Arduino or whatever you choose to use to build the pulse train.
Hi Matt,
Nice design, but it will not work. The D1 diode in your drawing will block all
current from flowing through the Q1 transistor. Also the IRF3205 was
upside down.
Attached is a proposed correction.
GL.
Quote from: Groundloop on 2014.03.20, 20:42:52
Hi Matt,
Nice design, but it will not work. The D1 diode in your drawing will block all
current from flowing through the Q1 transistor. Also the IRF3205 was
upside down.
Attached is a proposed correction.
GL.
Ha! Good eye! Of course me being blind, that isn't saying much.
I'll proceed with a redo. And a design for the Andrei Melnichenko version too.
Grum, If I'm going south with your thread in any way, just a swift kick is all I need. My intent is to lay out some solid test beds we can use to evaluate coil designs and maybe get us a runner.
Quote from: Matt Watts on 2014.03.20, 20:59:45
Ha! Good eye! Of course me being blind, that isn't saying much.
I'll proceed with a redo. And a design for the Andrei Melnichenko version too.
Grum, If I'm going south with your thread in any way, just a swift kick is all I need. My intent is to lay out some solid test beds we can use to evaluate coil designs and maybe get us a runner.
Matt,
No problem, I do such errors all the time. Right now I'm working on a GPS based auto pilot to a boat. Did
add a small pizeo buzzer today. The buzzer will make a sound if the GPS loose tracking. Added some code
to my program to control the buzzer, and, voila, f..ked up my code timing and did loose the GPS input. :-)
GL.
Corrected Akula DipTrace (http://www.diptrace.com/) project (schematic, board and custom lib) attached.
Concerning the used HL 15V leds, perhaps he means something like this:
http://www.hero-ledstore.com/heroled-3w-led-c-2_174.html?zenid=8f71hoj8k9s3tbefmn7kn8ej72
or like Peterae mentioned:
http://www.ebay.co.uk/itm/50x-3W-Cool-White-High-Power-Super-Bright-LED-Light-3-Watt-Lamp-Bulb-Chip-/140968579330?pt=UK_BOI_Electrical_Components_Supplies_ET&hash=item20d261d102
Regards Itsu
Hey guys, I'm new to OUresearch and this thread and just wanted to to say hello. :)
concerning LEDs: I have a LED that looks very similar like the one just posted by ITSU, but rated at 16,4 V @1000mA, think they are available with different specs. See eBay link for mine:
http://www.aaaa-electronic.de/products.php?OSRAM-Semiconductors-LECWE2B-Uni-Color-White-Chip-LED&cPath=5&language=en_US (sorry this was link from their shop only, here is the ebay link too:)
http://www.ebay.de/itm/OSRAM-Semiconductors-LECWE2B-Uni-Color-White-Chip-LED-/360697925575?pt=Bauteile&hash=item53fb458fc7
When running at lower voltage than 16 v, amperage drops dramatically, think at 10v was only around 20mA, so maybe at 15v its exactly the 3W. Sadly cant do a quick test right now as its already damaged. But I anyhow wanted to order a new these day then i will check on that.
kind regards
From what I can decipher from YouTube's translation, this isn't all that difficult:
Quote from: Matt Watts on 2014.03.21, 13:43:43
From what I can decipher from YouTube's translation, this isn't all that difficult:
Hmmm, he is using the diagram (up side down) with the floating pin16 (which prevent the led from being lit), see video at the 3:26 minute mark:
Regards Itsu
Quote from: Itsu on 2014.03.21, 14:25:15
Hmmm, he is using the diagram (up side down) with the floating pin16 (which prevent the led from being lit), see video at the 3:26 minute mark:
Regards Itsu
The comment in YouTube indicates he is using this schematic:
http://savepic.org/5111643.jpg
Which is the original single transistor driver circuit and it does have the ground connected.
What I do not see is the fifth wire (the ground shield) on the coil.
Quote from: Itsu on 2014.03.21, 14:25:15
Hmmm, he is using the diagram (up side down) with the floating pin16 (which prevent the led from being lit), see video at the 3:26 minute mark:
Regards Itsu
Good day all; thanks for the invite to participate here on OUR.....
Hello Itsu:
Yes, I noticed the same peculiarities with the video also:
Below is a post I made at OverUnity regarding the same video in relation to Schematic that is posted as a direct link below the video on YouTube.
The Schematic posted as a Link is not the same as the one in the video which as you pointed out is MISSING the Ground connection to the Voltage divider, Pin #16 error amp and LED indicator for the 12 volt. reg.
Re: Akula0083 30 Watt Self Running Generator.
« Reply #144 on: March 11, 2014, 09:00:51 PM »
Quote
Quote from: From other Planet on March 11, 2014, 08:25:06 PM
btw, this seems to be the schematic he used, he posted link in comments: http://savepic.org/5111643.jpg
Hello From other Planet:
Yet another mystery to be dealt with:
The posted schematic from the U-tube channel does NOT match the Blurry Schematic shown in the Video of the Self- running TPU.
I could not guess why one schematic is shown in the video and another very different one is shown as a linked reference.
Thanks, peace
lost_bro
So my question is How is the indicator LED in the video circuit lit-up if if has NO Ground connection? >:(take care, peace
lost_bro
Good day All:
Just wanted to share an interesting simulation using LTspice. The original file comes from the same Russian forum that I reference earlier.
I took the liberty to change only the schottkey diodes in the circuit, as the diodes that came with the original file were Russian I suppose and I did not have them in my LTspice library.
All else are the same values as from the Russian Author of the simulation. I take no credit.......
Please see the observation that I made regarding Voltage at the output of the L2 coil (feedback) and the input which in the circuit for simulation purposes is 3 volts even.
I believe you will see the there is an EXCESS of .16Volts coming from the L2 coil.... Of course, This is a perfect simulation and does NOT take into account many REAL life losses (ie; Voltage regulator losses etc).....
AND, the voltage is the PEAK pulsating Voltage NOT constant..... But none the less it is encouraging.
I will post the original LTspice file if anyone is interested in exploring further this anomaly. O0
take care, peace
lost_bro
Dear All.
I wold first like to welcome, From other Planet and lost_bro. Consider yourselves "poached" !! O0
T-1000 shared this video via Skype to Itsu and myself last night. Please enjoy !!
http://www.youtube.com/watch?v=afPQtDbn_3s
You will note that the transformer is a pair of TV yoke cores and the driver is a Push Pull style with frequency controller chip. Ah La GeoFusion !! "Say no more " Nod Nod Wink Wink........... Sorry I grew up with Monty Python !! ;D
I think the point I am trying to make here is that it does not matter what form the drive takes, it's the way the transformer is configured !!
T-1000 is convinced that we are looking at a Parallel input resonance and a Series output resonance and the above video is achieving this at around 14 KHz!!
Comments??
Cheers Grum.
Hi Lost_bro
Welcome O0
Thanks for the work on the sim, I'm not a sim man, is there anything they could have done to the files to fool the simulator, i ask this because i seem to remember a sim floating around by an Indian guy and it also showed OU, but if i recall correctly he fiddled with something which increased the output and fooled the sim.
Maybe someone else will remember more details.
EDIT i am not sure it's a good idea to use a digital meter to measure current on the output.
Dear Peterae.
Quote "EDIT i am not sure it's a good idea to use a digital meter to measure current on the output. "
Was that for me ??
I am in total agreement !! Particularly when most DVM's have a max current of 10 Amps !!
But the lamp was pretty bright nonetheless !!
Cheers Grum.
Quote from: Peterae on 2014.03.21, 18:46:38
Hi Lost_bro
Welcome O0
Thanks for the work on the sim, I'm not a sim man, is there anything they could have done to the files to fool the simulator, i ask this because i seem to remember a sim floating around by an Indian guy and it also showed OU, but if i recall correctly he fiddled with something which increased the output and fooled the sim.
Maybe someone else will remember more details.
EDIT i am not sure it's a good idea to use a digital meter to measure current on the output.
Good Day Peterae
Too answer your question in short: NO
This is because LTspice uses library files, and the library files that I use are in my computer and are standard files.... I only used the Schematic from the Russian forum.
This is akin to using a map. A map is NOT the territory.
So, using my files from my computer I ran his schematic as a simulation. All of the .op directives are listed on the screen shot that I uploaded and nothing is out of the ordinary with the command lines.
I am NOT saying that this IS OU. But it definitely shows promise and this is Using Standard Accepted Physics.
Of, course for this scheme to work (ie: OU..) we must now add the secret ingredient which will probably NOT be part of the LTspice program.
I know that the secret to this IS in the coil configuration, just as the work of Stanley Meyers was and is the secret of the coil configurations.
take care, peace.
lost_bro
@Grumage
QuoteWas that for me ??
Just a comment that came to me while watching the video.
for instance, if we say use a 100Watt bulb and we know the current I=35.5A then V=P/I
Bulb Wattage Bulb Voltage
100W 2.82V
50W 1.408v
10W 282mV
So it takes a bulb voltage rating of 2.82V to light a 100 Watt bulb to the brightness in the vid when 35.5 amps current are flowing
I've never heard of a 282mV 10 Watt bulb
So there's something fishy going on.
@Lost_bro
OK Thanks that's interesting then O0
Received my gapless core today (2 days, super fast), see picture.
With the gapped one i now can make severall combinations of gaps.
The gapless one measures like this:
L1 500uH
L2 7mH
Now of to build the control circuit.......
Regards Itsu
Nice one Itsu ;)
Dear Itsu and all.
A very nice transformer and almost identical inductances to mine !! O0
Please see attached screenshot from the recent Russian video I posted.
Comments ??
Cheers Grum.
I can't explain this hunch, so I'll just show it. Something about that little RC circuit tells me it's there to slightly offset the phase between the primary and secondary. And a phase offset could be the key to the resonance or a way to match any transformer you hook up.
Quote from: Matt Watts on 2014.03.21, 20:41:30
I can't explain this hunch, so I'll just show it. Something about that little RC circuit tells me it's there to slightly offset the phase between the primary and secondary. And a phase offset could be the key to the resonance or a way to match any transformer you hook up.
Dear Matt.
T-1000 made a strong reference to that "LITTLE" corner !! ;)
Glad you brought it up !!
Cheers Grum.
Quote from: Grumage on 2014.03.21, 18:45:18
Dear All.
I wold first like to welcome, From other Planet and lost_bro. Consider yourselves "poached" !! O0
T-1000 shared this video via Skype to Itsu and myself last night. Please enjoy !!
http://www.youtube.com/watch?v=afPQtDbn_3s
You will note that the transformer is a pair of TV yoke cores and the driver is a Push Pull style with frequency controller chip. Ah La GeoFusion !! "Say no more " Nod Nod Wink Wink........... Sorry I grew up with Monty Python !! ;D
I think the point I am trying to make here is that it does not matter what form the drive takes, it's the way the transformer is configured !!
T-1000 is convinced that we are looking at a Parallel input resonance and a Series output resonance and the above video is achieving this at around 14 KHz!!
Comments??
Cheers Grum.
Good Day Grum, thanks for the invitation to be part of the 'Group'.
OK, funny thing you should mention the Frequency controller chip.
The TL494CN is a PWM controller chip. It regulates via feedback thru the error amps to control Pulse width of the signal, not the frequency of signal.
I thought this was strange from the start because the only frequency regulation on the schematic is from potentiometer R15.
This adjustment is NOT automatic ( as the error amps only control Pulse width) so this Chip will NOT track frequency, if that was the original idea.
From my experience..... the frequency will drift with temp. change of the circuit..... and I would assume that the operating frequency IS critical to the workings of the circuit.
Do you know which chip is used in the video?
Take care, peace.
lost_bro
Hi Gentlemen,
My 2 cents... Once again this clapped-out old drivelling French guy? :-X
Yes! :-[
IMO:
- In the
Shark (акула/Akula)'s device, the secret is in the transformer.
Of course, it is very usefull to have the "driving circuit".
- But, if we do not understand what is going on in this apparatus
the likelihoods of success are very weak. R they not?
So, post like the Matt Watts's one about Andrei Melnichenko and this
Transgeneration_Discovery.pdf could be of great help.
Should I also quote:
Quote from: T-1000 on 2014.03.18, 20:53:31
Hi,As I mentioned before, the secret is in way transformer is made, not in electronics.
............
And
Quote from: lost_bro on 2014.03.21, 19:40:32
..........
I know that the secret to this IS in the coil configuration, just as the work
of Stanley Meyers was and is the secret of the coil configurations.
..........
For my part, I keep on thinking that:
- The Akula's "transformer" is more than a mere transformer.
- The Kapanadze "coil(s)" is/are more than (a) mere coil(s).
Now, 2 pictures:
1) This one:
(http://www.freenrg.info/4OUR_F/Double_condo.jpg)
2) The modified MenofFather's picture from OU dot com (already quoted here)
http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg393505/#msg393505
(http://www.freenrg.info/4OUR_F/Wound-Mod4.jpg)
------------------
If (Voltage across a Capacitor)
Then
If (Variable Voltage)
Then
variable electric field
Variable magnetic field at right angle
end if
else
Va revr gant al laboused bihan, me zo 'vont d'ar gêr ;D
end if
--------
Cheers,
Jean
Quote from: NerzhDishual on 2014.03.21, 22:16:22
IMO:
- In the Shark (акула/Akula)'s device, the secret is in the transformer.
Of course, it is very usefull to have the "driving circuit".
- But, if we do not understand what is going on in this apparatus
the likelihoods of success are very weak. R they not?
So, post like the Matt Watts's one about Andrei Melnichenko and this
Transgeneration_Discovery.pdf could be of great help.
Should I also quote:
And
For my part, I keep on thinking that:
- The Akula's "transformer" is more than a mere transformer.
- The Kapanadze "coil(s)" is/are more than (a) mere coil(s).
Now, 2 pictures:
1) This one:
(http://www.freenrg.info/4OUR_F/Double_condo.jpg)
2) The modified MenofFather's picture from OU dot com (already quoted here)
http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg393505/#msg393505
(http://www.freenrg.info/4OUR_F/Wound-Mod4.jpg)
------------------
If (Voltage across a Capacitor)
Then
If (Variable Voltage)
Then
variable electric field
Variable magnetic field at right angle
end if
else
Va revr gant al laboused bihan, me zo 'vont d'ar gêr ;D
end if
--------
Cheers,
Jean
Hello Jean
Yes, I remember from my days of pursuing the S.M. HHO experiments that Stanley Meyers talked about 'distributed Capacitance' in his bifilar coils used in his patents.
He apparently used it to his benefit.
Tesla always talked about reducing 'distributed capacitance as if it were the enemy. He shows many different ways to add capacitors across his coil configurations to eliminate his version of distributed capacitance.
Yes, we are dealing with something very different here.
take care, peace.
lost_bro
Hi Lost_Bro,
Thank you for your comment. I was talking about "purposely" added capacitance.
Actually, I had forgotten to post this picture:
(http://www.freenrg.info/4OUR_F/Usual_And_Not_Usual_Caps.jpg)
-------------------
According that, in an AC circuit, a capacitance (and/or a coil) does not consume (and produce) nothing.
Notably: take a "tank circuit" (parallel resonance).
Seen from the source, it has an huge impedance. It behaves like an almost open circuit.
So, very few power is consumed from the source.
But, in this tank circuit, amperes are circulating.
Unfortunately, these amperes are considered as useless.
Fortunately, some witty guys (not me!) had the idea of designing non conventional caps and coil.
Actually, a kinda mix of cap and coils.
One question is: does Nature care (NRG-wise) whether or not a tank circuit is made of "normal"
caps and coils or tricky (non-conventional) ones? Seems not.
Another question is: does the AC Source will figure out (and react (Lenz law)) if
we try to get some power from a non-conventional tank circuit?
Usually, trying to get back some power from a "normal" resonating tank circuit
(with a secondary coil, for ex) leads to break his resonance and the efficiency is < 1.
Now, can we (without a secondary coil, for ex) get back some free power?
Could we cheat the source?
I have not the answer. I came across these issues reading Valdimir Utkin.
I have to make some experiments.
Gwella soñjou,
Jean
Quote from: NerzhDishual on 2014.03.22, 00:22:21
Hi Lost_Bro,
Thank you for your comment. I was talking about "purposely" added capacitance.
Actually, I had forgotten to post this picture:
(http://www.freenrg.info/4OUR_F/Usual_And_Not_Usual_Caps.jpg)
-------------------
According that, in an AC circuit, a capacitance (and/or a coil) does not consume (and produce) nothing.
Notably: take a "tank circuit" (parallel resonance).
Seen from the source, it has an huge impedance. It behaves like an almost open circuit.
So, very few power is consumed from the source.
But, in this tank circuit, amperes are circulating.
Unfortunately, these amperes are considered as useless.
Fortunately, some witty guys (not me!) had the idea of designing non conventional caps and coil.
Actually, a kinda mix of cap and coils.
One question is: does Nature care (NRG-wise) whether or not a tank circuit is made of "normal"
caps and coils or tricky (non-conventional) ones? Seems not.
Another question is: does the AC Source will figure out (and react (Lenz law)) if
we try to get some power from a non-conventional tank circuit?
Usually, trying to get back some power from a "normal" resonating tank circuit
(with a secondary coil, for ex) leads to break his resonance and the efficiency is < 1.
Now, can we (without a secondary coil, for ex) get back some free power?
Could we cheat the source?
I have not the answer. I came across these issues reading Valdimir Utkin.
I have to make some experiments.
Gwella soñjou,
Jean
Hello Jean,
Yes, I understood that you were referring to induced or added Capacitance in the coil, just as S.M. did.
You are speaking of Reactive AC power, as that is the definition:
If the circuit contains reactive components, the voltage and current waveforms will be "out-of-phase" by some amount determined by the circuits phase angle.
Reactive power is produced when the current waveform is out of phase with the voltage waveform due to inductive or capacitive loads.
Current lags voltage with an inductive load, and leads voltage with a capacitive load......
(ELI the ICE man)
Only the in phase component of current with voltage produces real or active power that does real work.
Reactive Power can best be described as the quantity of "unused" power that is developed by reactive components, such as inductors or capacitors in an AC circuit. Yes, it bounces back and forth in the circuit, just recirculating..............
If the phase angle between the voltage and the current is at its maximum of 90 degrees, the volt-amp product will have equal positive and negative values.
In other words, the reactive circuit returns as much power to the supply as it consumes resulting in the average power consumed by the circuit being zero, as the same amount of energy keeps flowing alternately from source to the load and back from load to source.
This is what you are getting at: tap into the reactive power , which is created by the interaction of Inductance & Capacitance...............................An analogy used to explain these terms is that if Apparent Power is a glass of beer, Reactive Power is the foam that prevents you from filling it up all the way, so that you are left with less beer or True Power. In other words, the thirst-quenching portion of your beer is represented by KW in the pic. below. The foam is represented by KVAR. The total contents of your mug, KVA, is this summation of KW (the beer) and KVAR (the foam). In our beer mug analogy, Power Factor (P.F.) is then the ratio of Beer (True Power) to the entire volume of the mug (beer plus foam or Apparent Power.)
P.S. stole the pics. from Google......
take care, peace.
lost_bro
Hello All.
OK, Jean...
"One question is: does Nature care (NRG-wise) whether or not a tank circuit is made of "normal"
caps and coils or tricky (non-conventional) ones? Seems not."
"Another question is: does the AC Source will figure out (and react (Lenz law)) if
we try to get some power from a non-conventional tank circuit?"
I believe the answer lies in the Phase relationship ( I&E) in relation to the Primary and Secondary coils.
I will quote 'Salty Citrus' from China, the developer of the Tang-Don Smith replication last year in China.
" "salty citrus" Q & A:
Question : About the 1/4 wavelength relationship, your opinion on L1 and L2; do they have a 1/4 wavelength relationship?
Answer: I feel the 1/4 wavelength relationship is not great, the only relatively important point is the phase relationship. "
This was translated from Chinese to English, but the important part is that his group spent over two years working up to this final version and his opinion is that the Phase relationship between L1 & L2 is the hot item.
Unfortunately, he never stated what exactly was the phase relationship but I believe this information is a step in the right direction.
Take care, peace
lost_bro
Great analogy with the beer Lost_bro, i love it O0
Ok, made a rough setup with the components i had.
Still some more needs to be collected/sorted like the leds but this will be my main layout.
Regards Itsu
Quote from: Itsu on 2014.03.22, 11:31:43
Ok, made a rough setup with the components i had.
Still some more needs to be collected/sorted like the leds but this will be my main layout.
Very nice Itsu.
I think there is a wager going around that you will have this working before any of us even get circuit boards fabricated. If you do, I'll happily send you some boards--my compliments.
Thanks Matt Watts,
but i doubt i will be ready soon, i just have 1 week left before i am off for my work for 2 weeks.
You guys can catch up on me then ;D
Regards Itsu
Dear all.
MenofFather posted this today on OU.C.
Quote"He also say "Транс мотайте по картинке", that means, transformer need wound like in picture, so I understand, primary wound on one side of core, secondary on another side of core. :)
Maybe then not need screens or only one screen?
(https://dfsuknfbz46oq.cloudfront.net/p/icons/kolourpaint.png)" unquote.
This was my reply.
Quote " From that statement "transformer need wound like in picture" . I deduce that we still have T-1000's original coil layout as attached below.
We have been given a turns count L1 = 15 Turns. L2 = 45 turns. But the photo clearly shows us, that externally we have the same or similar height to the two coils. Suggesting to me that there are at least 30 turns underlying both halves !! I am basing this thought on that L2 would be twice a high as L1 !!
I have also attached a picture of my Mk2 rewound transformer" unquote.
Thoughts ??
Cheers Grum.
Let's way what Ruslan would say about it. He promised to post new video with better resolution and details.
IMHO : 1/4 wavelength - it is important in HV and overall to maximize output voltage from flyback for example. Very hard to do something useful about it except RF and HV applications.
Hi grumage, its anybody's guess, i really don't know.
I had problems in viewing the linked video from the post from Les banki over on overunity.com which was pasted to this thread
in post #159 by grumage.
This link does not work for me:
http://www.youtube.com/watch feature=player_embedded&v=EwCS15pRtH0&list=PLC7684829E98CAD74
After fiddling around i came finally to this video which shows indeed 3 circuits of the akula 30W version:
https://www.youtube.com/watch?v=EwCS15pRtH0&list=PLC7684829E98CAD74
Most difference is in how to drive the MOSFET, perhaps there is more interesting info in the audio, so for those
who understand please take a look/listen.
Thanks, regards itsu
Those 3 circuits are interesting, they seem to me to be a progressive circuit development, looks to me he has been trying to find the best way to switch that fet faster, which would mean his latest circuit uses a fet driver.
Quote from: Peterae on 2014.03.22, 20:59:00
Those 3 circuits are interesting, they seem to me to be a progressive circuit development, looks to me he has been trying to find the best way to switch that fet faster, which would mean his latest circuit uses a fet driver.
If my instincts are correct on this, it is the impulse that initiates the ringing between the two halves of ferrite. So starting with the smallest duty cycle that will work is best. Keep the on-time intact and decrease the off-time as much as possible, which will at the same time increase frequency. My experience with pulsed coils has proved the TC4422 to be a formidable mechanism in pushing a MOSFET or IGBT as hard as it can go. I suspect building flexibility and hefty overkill into a transformer test bed will allow us to hit the sweet spot and possibly variant transformer designs that can go beyond the 30 watt Akula unit. Myself, I have my eye on a large split toroid transformer design, but will focus on baby steps until we get a handle on this thing.
Indeed
It's easy to start planning a large unit but definitely we need to get the baby going so we can analyse the circumstances creating the effect, who knows once the mechanism is understood then there's no reason not to be able to design a 3-4kw off grid house unit.
Quote from: Peterae on 2014.03.22, 21:29:04
Indeed
It's easy to start planning a large unit but definitely we need to get the baby going so we can analyse the circumstances creating the effect, who knows once the mechanism is understood then there's no reason not to be able to design a 3-4kw off grid house unit.
;) :) O0
@new members
Welcome to the forum! O0
@Itsu
Nice work on your build! 8)
I pulled the original MOSFET version of the 30W schematic from the video.
This is a quick graphic of what I hear some are saying that the transformer primary is under the secondary. This might be, but it doesn't seem to match the picture.
Success? Post from OU.c
Quote
The schematic that TOPruslan (successful replicator) uses, has 2000μF capacitors not 2200μF.
See: http://imageshack.com/a/img819/1125/v836.jpg
Changes that TOPruslan made to the schematic:
1) Pin 16 must be connected to the ground.
2) C11 must be connected to the input Pin of VR-7812 and not after the inductor.
3) The wire for both windings must be 1mm.
See: http://www.freenergy.com.ua/topic/89-razborki/?p=11778
Found these on Russian forum:
http://www.freenergy.com.ua/topic/132-ustanovka-akula0083/?p=11438 (http://www.freenergy.com.ua/topic/132-ustanovka-akula0083/?p=11438)
Check out this video. This is the simplest one I've seen yet.
http://www.youtube.com/watch?v=YPiUSCQIKC4
@All,
I have designed a PCB based on the original version of the 30W schematic.
I have used component values that are as close to the original values as I could
find. I also did pay attention to the RCL part on the right side of the original drawing,
and did design this part to be tunable by the use of a variable resistor and variable
capacitor. I did design the PCB to fit standard components for easy soldering.
Attached is the the original version of the 30W schematic drawing,
my design circuit drawing,
my PCB layout image
and all the Eagle design files if anybody wants to make a PCB for them self.
Added: There will be four (4) professional made PCBs that I can give away for free!
If you want a PCB then PM me your snail mail address and country.
It will take a little over a week before I have the PCBs.
I will snail mail the PCBs to the first four PMs I get...................
Added: All four boards are now claimed!
Added: I did choose not to put the component numbers on the PCB to keep the
layout more "clean", so you must use the free EAGLE CAD program and
load the design files to see where the components must be soldered to the PCB.
The free EAGLE CAD program can be found here: http://www.cadsoftusa.com/download-eagle/
Added: I did add the data sheet for the L1 coil, and corrected a minor issue on the PCB.
Regards,
GL.
GL
Great work on the pcb O0
So how many PCB's have you ordered and are you going to give 4 away and sell the rest or did you get 5 made in total.
Quote from: Peterae on 2014.03.23, 08:16:37
GL
Great work on the pcb O0
So how many PCB's have you ordered and are you going to give 4 away and sell the rest or did you get 5 made in total.
Peter,
Thanks :-) (Did you see the nice shark on the circuit drawing?)
Did order 5 PCBs. Can't afford any higher count. I need one, and four PCBs I will give away for free.
GL.
Quote from: 4Tesla on 2014.03.23, 07:00:15
Check out this video. This is the simplest one I've seen yet.
http://www.youtube.com/watch?v=YPiUSCQIKC4
4Tesla,
Looks like a standard Joule Thief running on big electrolytic capacitors.
GL.
How much would 10 or 20 pcb's cost i will help with money, tell me how much and i will paypal it over to you. O0 if not too late that is.
Quote from: Peterae on 2014.03.23, 08:31:06
How much would 10 or 20 pcb's cost i will help with money, tell me how much and i will paypal it over to you. O0 if not too late that is.
Peter,
Sorry too late, my 5 PCBs is already ordered. My cost for 5 PCBs = 3155 NOK. The parts add 700 NOK to this.
Norway is a high cost country and that is way I have posted the design files so that people in lower cost countries
can order the PCB and parts them self. But my offer for 4 free PCBs still stands.
GL.
3155 NOK = £315.83p surely something wrong here, very expensive?
Quote from: Peterae on 2014.03.23, 08:39:30
3155 NOK = £315.83p surely something wrong here, very expensive?
Peter,
Nothing wrong, the PCB company takes a fixed starting cost, and then the PCB number of boards cost add to this.
If I order a million PCBs then the cost for each PCB will only be a very small sum. But when you order just a few PCBs
then the total cost of each PCB is high because of the fixed starting cost.
GL.
Wow man, i will enquire with china and get 20 boards made up. O0
http://www.ebay.co.uk/itm/2-4-layer-PCB-Manufacture-Prototype-Etching-L-10cm-or-Width-10cm-or-Qty-10-/170920142583?pt=LH_DefaultDomain_2&hash=item27cba266f7
Quote from: Peterae on 2014.03.23, 08:46:04
Wow man, i will enquire with china and get 20 boards made up. O0
http://www.ebay.co.uk/itm/2-4-layer-PCB-Manufacture-Prototype-Etching-L-10cm-or-Width-10cm-or-Qty-10-/170920142583?pt=LH_DefaultDomain_2&hash=item27cba266f7
Peter,
Yes, that is one way to do it. That is why I posted the design files. :-)
GL.
Cheers GL thanks for your time on the PCB's
Let's try and get an idea of how many we want, are you able to cancel your pcb's, i can send you some also.
I can send a bulk package to the US and someone can distribute from there, i can do UK and maybe Europe.
PCB is free for OUR members O0
Place your order on this thread O0
Quote from: Peterae on 2014.03.23, 08:57:22
Cheers GL thanks for your time on the PCB's
Let's try and get an idea of how many we want, are you able to cancel your pcb's, i can send you some also.
I can send a bulk package to the US and someone can distribute from there, i can do UK and maybe Europe.
PCB is free for OUR members O0
Place your order on this thread O0
Peter,
I can't cancel my order, to late for that. I have a program on my computer that automatic sends the PCB
design files to the factory server. When I press BUY then it is to late to change my mind. But I salute you
for your generous offer to make PCBs for forum members so that we all can replicate the circuit.
GL.
Thankyou for your kind offer also.
So if you keep your boards for people over in your country, i will do US & UK, if you want more boards say, i dont mind getting as many as we need, it wont cost much more for 40 or so boards.
Quote from: Peterae on 2014.03.23, 09:12:05
Thankyou for your kind offer also.
So if you keep your boards for people over in your country, i will do US & UK, if you want more boards say, i dont mind getting as many as we need, it wont cost much more for 40 or so boards.
Peter,
I don't mind snail mailing my 4 free boards to people anywhere in the world.
I will have the boards around the first week of April. Another note, in
the design files there is a complete order list for ELFA DISTRELEC
(covers all EU): https://www1.elfa.se/elfa3/StartPage.do
for easy order of parts. The order list can be imported into you web browser.
GL.
Thanks guys! I've been reading on the Russian forums via Google Translate. It seems the ones that do get self runners, run at most for a few days. But hey that is a start.
Dear Groundloop.
I wish there were more people like you living on planet Earth !! What a wonderful world it would be !! Your selfless attitude, I'm afraid I don't have the words to fully explain myself.
That said, and in simple terms, a MASSIVE thank you for all your time and effort, you truly are a very good man !! O0
Sincerest wishes to you. Grum.
Quote from: Grumage on 2014.03.23, 12:52:18
Dear Groundloop.
I wish there were more people like you living on planet Earth !! What a wonderful world it would be !! Your selfless attitude, I'm afraid I don't have the words to fully explain myself.
That said, and in simple terms, a MASSIVE thank you for all your time and effort, you truly are a very good man !! O0
Sincerest wishes to you. Grum.
Grumage,
Thank you for the nice words. :-)
I did set myself some goals when designing this new PCB. First was not to deviate to much
from the original circuit drawing posted by Akyla0083. The next was that all parts was of
the "old" type (no SMD) components so that everybody could solder this unit. My third goal
was to make the feed back loop tunable by adding a variable resistor and variable capacitor.
My forth goal was that all components (except transformer windings) was buy-able from a
EU located electronics firm. The final goal was to make the PCB look great. :-)
To clarify, the transformer Ferrite core etc. is buy-able but the builder wind his own wire onto the
coil bobbin.
Do you want one of the free PCBs?
GL.
Quote from: 4Tesla on 2014.03.23, 09:24:45
Thanks guys! I've been reading on the Russian forums via Google Translate. It seems the ones that do get self runners, run at most for a few days. But hay that is a start.
A very good start!
Thanks for the work Grumage and Groundloop and others... I'm climbing the learning curve on this approach... fascinating...
Quote from: Groundloop on 2014.03.23, 08:28:36
4Tesla,
Looks like a standard Joule Thief running on big electrolytic capacitors.
GL.
He said he based on Akula's circuits. Commenters say it runs a few days. Maybe Akula/Kapa circuits are high wattage Joule Thief circuits. As they do quit working after a few days.
Quote from: 4Tesla on 2014.03.23, 15:25:53
He said he based on Akula's circuits. Commenters say it runs a few days. Maybe Akula/Kapa circuits are high wattage Joule Thief circuits. As they do quit working after a few days.
Could be very close to the truth!
Quote from: Hoppy on 2014.03.23, 15:33:45
Could be very close to the truth!
I think we should still work on getting a good self runner. Then phase two.. make it run longer than a few days.
Quote from: 4Tesla on 2014.03.23, 06:57:22
Found these on Russian forum:
http://www.freenergy.com.ua/topic/132-ustanovka-akula0083/?p=11438 (http://www.freenergy.com.ua/topic/132-ustanovka-akula0083/?p=11438)
Good Day 4 Tesla;
Those are very interesting schematics that you posted. I will study them when I find time.
Hello All.
Short of sounding like a broken record, I post my thoughts on the drive portion of the 'Final' Akula TPU circuit.
First I will say that it is very encouraging to see such enthusiasm with any project, let alone one as complicated as this one.
I am NOT trying to be critical, and only offering suggestions based on my observations and experience with the TL494 chip.I am NOT diss-ing GroundLoop for his work and time donated to the cause................................
My question is: Has anyone proto-typed (breadboard) the proposed drive circuit and traced the signals with an o-scope?
I have NOT had time to do so myself, so my comments are only based on experience and not signal tracing the proposed circuit.
take care, peace.
lost_bro
Quote from: lost_bro on 2014.03.23, 16:19:33
My question is: Has anyone proto-typed (breadboard) the proposed drive circuit and traced the signals with a o-scope?
I have NOT had time to do so myself, so my comments are only based on experience and not signal tracing the proposed circuit.
take care, peace.
lost_bro
On the Russian forums, yes.. here, I don't think so. I myself like to breadboard first, but everyone has their own preferences.
Russian forums:
http://realstrannik.ru
http://www.freenergy.com.ua
Quote from: 4Tesla on 2014.03.23, 16:35:33
On the Russian forums, yes.. here, I don't think so. I myself like to breadboard first, but everyone has there own preferences.
Good Day 4 Tesla
Yes, I have learned the hard way about prototyping first: highly recommended....... and to each their own.
I know that it sucks to waste time and components...
take care, peace.
lost_bro
I've learned a lot by surfing the Russian forums. If anyone here using Firefox, I highly recommend this translator add-on:
S3.Google Translator
https://addons.mozilla.org/en-US/firefox/addon/s3google-translator/
It automatically translates the pages using Google Translate. Not perfect, but helpful. Also, I think Chrome might have this feature built in.
4Tesla,
With the reading you have been doing, are any of those guys revealing what component fails when the unit quits running? We heard about the one where the core crumbled, any other symptoms you have spotted like burnt wiring, blown semiconductors...? When the unit quits are some able to restart it with no changes?
MOSFETS/Transistors. I think it is because for this to work the core may have to be saturated which creates a short.. from my understanding.
I don't know about restarting.. if components have to be replaced or just started again with battery.
Quote from: Groundloop on 2014.03.23, 07:41:14
I have designed a PCB based on the original version of the 30W schematic...
This is excellent work GL. It has my vote as "The Reference Design" for testing this device. Also a very kind offer putting four of them out there for grabs, which by now all should have been claimed.
Exciting times ahead.
I'm also looking at placing an order for 20-30 pcb's and am taking orders, the pcb's are free, I'm in the UK and will post a number to someone in the US who can then distribute them, maybe Chet.
So please post here if you want one so i can get an idea how many to get made up, at the moment i don't think anyone has claimed any of Groundloops and so far no one has asked for one from me.
Quote from: Peterae on 2014.03.23, 17:44:20
I'm also looking at placing an order for 20-30 pcb's and am taking orders, the pcb's are free, I'm in the UK and will post a number to someone in the US who can then distribute them, maybe Chet.
So please post here if you want one so i can get an idea how many to get made up, at the moment i don't think anyone has claimed any of Groundloops and so far no one has asked for one from me.
+1 I'm in US. GL in UK? I'll let locals get GL's
Thanks you!
Quote from: lost_bro on 2014.03.23, 16:19:33
Good Day 4 Tesla;
Those are very interesting schematics that you posted. I will study them when I find time.
Hello All.
Short of sounding like a broken record, I post my thoughts on the drive portion of the 'Final' Akula TPU circuit.
First I will say that it is very encouraging to see such enthusiasm with any project, let alone one as complicated as this one.
I am NOT trying to be critical, and only offering suggestions based on my observations and experience with the TL494 chip.
I am NOT diss-ing GroundLoop for his work and time donated to the cause................................
My question is: Has anyone proto-typed (breadboard) the proposed drive circuit and traced the signals with an o-scope?
I have NOT had time to do so myself, so my comments are only based on experience and not signal tracing the proposed circuit.
take care, peace.
lost_bro
lost_bro,
You comment on my drawing is not correct. The mosfet is on (conducting current) when the C2/E2 transistor is off.
When the C2/E2 transistor is on again then the mosfet gate is discharged to ground, switching off the mosfet. The
output of the C2/E2 transistor can only sink current.
You don't need a pull down resistor at the mosfet gate because the C2/E2 transistor in the TL494 is pulling
down. The transistor never pulls up. This because R3 in the circuit drawing is the pull up. The BD139
transistor will carry most of the current to charge the mosfet gate when the TL494 transistor output
is not active, so I think it is needed.
The 1/2 frequency output is on purpose, I guess that was what Akyla0083 wanted. Also I wanted
to keep my design as close to the Akyla0083 posted design as possible.
GL.
Dear lost_bro and all.
Quote.
First I will say that it is very encouraging to see such enthusiasm with any project, let alone one as complicated as this one.
Unquote.
Sometimes it is enthusiasm that provides the catalyst for great things to happen !! The rallying cry, so to speak. I may be nearing 58 years but I have always been young at heart !! Taken the bull by the horns, jumped in with both feet, you know where I am coming from !!
Groundloops efforts, I feel, will soon be rewarded particularly with the team we have now assembled here at OUR. Away from those people that try to dissuade and dampen the enthusiasm that is really needed IMO to see something through.
So "Cry 'Havoc!', and let slip the dogs of war".
Lets build a self runner. 3 days will be fine for me !! ;D
Cheers Grum.
Quote from: Peterae on 2014.03.23, 17:44:20
I'm also looking at placing an order for 20-30 pcb's and am taking orders, the pcb's are free, I'm in the UK and will post a number to someone in the US who can then distribute them, maybe Chet.
So please post here if you want one so i can get an idea how many to get made up, at the moment i don't think anyone has claimed any of Groundloops and so far no one has asked for one from me.
None has claimed any PCBs from me yet, so I have still 4 to give away.
GL.
@GL
Is it possible to add jumpers so we can configure either scheme? Transistor or MOSFET.
Quote from: 4Tesla on 2014.03.23, 18:12:18
@GL
Is it possible to add jumpers so we can configure either scheme? Transistor or MOSFET.
4Tesla,
No change on my boards because I have already ordered them.
But anything is possible. If Peter has not ordered yet and want changes done
to the board then OK by me.
GL.
Quote from: Groundloop on 2014.03.23, 17:53:56
lost_bro,
You comment on my drawing is not correct. The mosfet is on (conducting current) when the C2/E2 transistor is off.
When the C2/E2 transistor is on again then the mosfet gate is discharged to ground, switching off the mosfet. The
output of the C2/E2 transistor can only sink current.
You don't need a pull down resistor at the mosfet gate because the C2/E2 transistor in the TL494 is pulling
down. The transistor never pulls up. This because R3 in the circuit drawing is the pull up. The BD139
transistor will carry most of the current to charge the mosfet gate when the TL494 transistor output
is not active, so I think it is needed.
The 1/2 frequency output is on purpose, I guess that was what Akyla0083 wanted. Also I wanted
to keep my design as close to the Akyla0083 posted design as possible.
GL.
Hello Ground Loop
Thanks for the comments:
Actually my comments are CORRECT: Please reference attached copy of original posted pic.
Ground Loop: "The mosfet is on (conducting current) when the C2/E2 transistor is off. When the C2/E2 transistor is on again then the mosfet gate is discharged to ground, switching off the mosfet.
The output of the C2/E2 transistor can only sink current."
Yes, I agree as stated in the pic. that I posted:
"MOSFET is ON when TL494 Output LOW" and "TL494 is SINKING CURRENT thru RED trace" ( Red trace is drawn thru C2 & E2) so
We are in agreement----You can place a SPST between Pin #13 & Pin #14 to choose at random the output mode : push pull or single pulse = 1/2 or complete Frequency of PWM.
Ground Loop: "You don't need a pull down resistor at the mosfet gate because the C2/E2 transistor in the TL494 is pulling
down. The transistor never pulls up."
I am NOT discussing the action of the BD139, I am referrencing a possibility of a High Noise Floor generated by any possible cause including but not limited to: parasitic inductance of traces on PCB or feedback from the L1 or L2. Depending on the Gate threshold values of the MOSFET in question it is ALL the more important.
Please see excerpt below from Tamids Blog: http://tahmidmc.blogspot.mx/2012/10/magic-of-knowledge.html
Importance of the gate-to-source resistor:
"It prevents accidental turn on of the MOSFET by external noise usually at startup when the gate is floating. The MOSFET may sometimes turn on with a floating gate because of the internal drain to gate "Miller" capacitance. A gate to source resistor acts as a pull-down to ensure a low level for the MOSFET. I have had MOSFETs blowing up in high voltage circuits, without the resistor in place. In most of the commercial power supplies / inverters I have seen, there is a 1k resistor used."
Sorry if I offended you, not my intention....
take care, peace.
lost_bro
Quote from: lost_bro on 2014.03.23, 18:21:15
Hello Ground Loop
Thanks for the comments:
Actually my comments are CORRECT:
Ground Loop: "The mosfet is on (conducting current) when the C2/E2 transistor is off. When the C2/E2 transistor is on again then the mosfet gate is discharged to ground, switching off the mosfet. The
output of the C2/E2 transistor can only sink current."
Yes, I agree as stated in the pic. that I posted:
"MOSFET is ON when TL494 Output LOW" and "TL494 is SINKING CURRENT thru RED trace" ( Red trace is drawn thru C2 & E2) so We are in agreement
----You can place a SPST between Pin #13 & Pin #14 to choose at random the output mode : push pull or single pulse = 1/2 or complete Frequency of PWM.
Ground Loop: "You don't need a pull down resistor at the mosfet gate because the C2/E2 transistor in the TL494 is pulling
down. The transistor never pulls up."
I am NOT discussing the action of the BD139, I am referrencing a possibility of a High Noise Floor generated by any possible cause including but not limited to: parasitic inductance of traces on PCB or feedback from the L1 or L2. Depending on the Gate threshold values of the MOSFET in question it is ALL the more important.
Please see excerpt below from Tamids Blog: http://tahmidmc.blogspot.mx/2012/10/magic-of-knowledge.html
Importance of the gate-to-source resistor:
"It prevents accidental turn on of the MOSFET by external noise usually at startup when the gate is floating. The MOSFET may sometimes turn on with a floating gate because of the internal drain to gate "Miller" capacitance. A gate to source resistor acts as a pull-down to ensure a low level for the MOSFET. I have had MOSFETs blowing up in high voltage circuits, without the resistor in place. In most of the commercial power supplies / inverters I have seen, there is a 1k resistor used."
Sorry if I offended you, not my intention....
take care, peace.
lost_bro
lost_bro,
I'm not offended at all. :-) Use to be fisherman in my young days LOL.
I think we are talking about the same issues. I also think the 1K resistor and 100nF (C1 and R4) snubber
circuit will take care of any transients. Also the anti series zener diodes will prevent some of the "spikes".
Lets build it first and test it. It is possible to change things around by adding components and cutting in
the PCB afterwards. One thing I think needs changing is the 1N4148 diode. We will probably need a Schottky
diode here.
GL.
Hello Ground Loop:
Yes, you are correct.... until it is built it is ALL TALK and NO WALK ......................
When I have time I will prototype it also.....
I am inclined to believe along the lines suggested by Matt the other day: That the use of a Discrete MOSFET Driver chip would actually be the best solution if a super fast Rise and Fall time is required for the pulse signal.
I have heard talk on the Russian Forums of 'Nano Pulsing'...... If that is the case then there are other 'PWM' chips and discrete MOSFET driver options that will make the 'Nano Pulsing' a reality.
Take care, peace
lost_bro
Quote from: lost_bro on 2014.03.23, 18:43:14
Hello Ground Loop:
Yes, you are correct.... until it is built it is ALL TALK and NO WALK ......................
When I have time I will prototype it also.....
I am inclined to believe along the lines suggested by Matt the other day: That the use of a Discrete MOSFET Driver chip would actually be the best solution if a super fast Rise and Fall time is required for the pulse signal.
I have heard talk on the Russian Forums of 'Nano Pulsing'...... If that is the case then there are other 'PWM' chips and discrete MOSFET driver options that will make the 'Nano Pulsing' a reality.
Take care, peace
lost_bro
lost_bro,
I have read about that too. Maybe the forward diode is for that purpose?
So we probably need a more powerful diode than the 1N4148.
BTW: I did check the data sheet for the TL494 and the RC oscillator values
that Akyla0083 did use puts the frequency off the factory scale for low Ohm
values. For highest Ohm value the frequency is just inside the window.
GL.
Hi all,
Not good with graphics, but this is what I was thinking for PCB. If others like the idea... please let us know so GL could draw up.. if he is cool with the idea.
Thanks!
Quote from: 4Tesla on 2014.03.23, 19:06:09
Hi all,
Not good with graphics, but this is what I was thinking for PCB. If others like the idea... please let us know so GL could draw up.. if he is cool with the idea.
Thanks!
4Tesla,
It is doable. It depend on what Peter says? If he want me to redesign
the PCB for such a breadboard layout then I will do it. But no time until
next weekend for this.
Is there any other people here that are familiar with the EAGLE CAD design program?
GL.
Quote from: Groundloop on 2014.03.23, 19:14:18
Is there any other people here that are familiar with the EAGLE CAD design program?
I can muddle through it GL, though I'm more comfortable with DipTrace.
Quote from: Groundloop on 2014.03.23, 19:02:33
lost_bro,
I have read about that too. Maybe the forward diode is for that purpose?
So we probably need a more powerful diode than the 1N4148.
BTW: I did check the data sheet for the TL494 and the RC oscillator values
that Akyla0083 did use puts the frequency off the factory scale for low Ohm
values. For highest Ohm value the frequency is just inside the window.
GL.
Hello Groundloop
Yes, the spec sheet list 300Khz as the MAX freq. for the TL494......
So. Akula wanted this beast to run in the upper spectrum of the chips capacity.
About the Nano Pulsing: First thing I would do is the Bypass the integral Body diode in the MOSFET using an External Super Fast Diode..... Place series Schottky at Drain ( cathode to Drain on MOSFET) and then run UF or SiC Schottky diode( in Paralled with MOSFET) from upper side of schottky (which is series with the Drain) to the Source of the MOSFET. Anode of parallel SiC Schottky connects to Source of MOSFET.
This effectively bypasses the internal body diode (Now parallel external Schottky or UF will respond in place of body diode) and will clean up the recovery time for the MOSFET by maybe an order of magnitude .
This will also have the effect to REDUCE the heat generated by the MOSFET because now the EXTERNAL UF or SiC Schottky is bypassing and not the internal body diode......... You can even heatsink the SiC Schottky separately for better dissipation.
This is a trick used in SS Tesla coil driver circuits and works really well.
Take care, peace.
lost_bro
Quote from: lost_bro on 2014.03.23, 19:27:44
Hello Groundloop
Yes, the spec sheet list 300Khz as the MAX freq. for the TL494......
So. Akula wanted this beast to run in the upper spectrum of the chips capacity.
About the Nano Pulsing: First thing I would do is the Bypass the integral Body diode in the MOSFET using an External Super Fast Diode..... Place series Schottky at Drain ( cathode to Drain on MOSFET) and then run UF or SiC Schottky diode( in Paralled with MOSFET) from upper side of schottky (which is series with the Drain) to the Source of the MOSFET. Anode of parallel SiC Schottky connects to Source of MOSFET.
This effectively bypasses the internal body diode (Now parallel external Schottky or UF will respond in place of body diode) and will clean up the recovery time for the MOSFET by maybe an order of magnitude .
This will also have the effect to REDUCE the heat generated by the MOSFET because now the EXTERNAL UF or SiC Schottky is bypassing and not the internal body diode......... You can even heatsink the SiC Schottky separately for better dissipation.
This is a trick used in SS Tesla coil driver circuits and works really well.
Take care, peace.
lost_bro
Sounds good! Can you draw out schematic? This is why I think a little bread on the PCB would be cool, because so many mods in that area of the circuit.
Quote from: 4Tesla on 2014.03.23, 19:45:46
Sounds good! Can you draw out schematic? This is why I think a little bread on the PCB would be cool, because so many mods in that area of the circuit.
Hello 4Tesla
As requested;
see below.
take care, peace
lost_bro
Hi People,
Thanks to everybody. Thanks to Lost_Bro, for his comments.
For the moment, I'm interested in very basic "theoretical" stuffs.
I just made a new quick experiment:
(http://www.freenrg.info/4OUR_F/My_N_Exp5.jpg)
--------------------------
It works = you can get voltage out of an air core coil wrapped with aluminium foil (as to build a kinda "capacitor")
even if this coil is just 'excited' by this fictitious displacement current. :)
You have to find the right ("resonant") freq.
Grounding/Eathing one particular end of the coil strongly increase the output voltage and change the "resonant" freq.
First results. No load. No "OU" claimed.
The "capacitor" : From 260 pF to 410 pF according to 2 different f' cheap meters.
The coil: Air core coil. From 19 to 22 Milli Henrie according to 2 different cheap bl' meters.
The Freq. Gen is given for: Max freq 3 MHZ. (Mine can goes a little bit further (3.15)).
Max output 10 volts ???
Actually, according to my scope:
At 3 MZH the Freq. Gen. output voltage (with not load) is only about 5 volts PK-to-PK.
At freqs less than about 500 KHZ this voltage is about 20 Volts PK-to-PK.
---------
Sine wave as CCT input. CCT Output volts-wise only. With no load.
Not grounded/'earthed'
Feq. Gen. output (= CCT input)about 20 volts PK-to-PK.
CCT output about 2 volts PK-to-PK at about 150 Khz.
Grounded/'earthed':
Feq. Gen. output (= CCT input) about 5 volts PK-to-PK.
CCT output about 20 volts PK-to-PK at about 3,07 Mhz.
--------------------
Something seems to be going on, here. No.?
These are just first results. The numbers have to be confirmed. The circuit (CCT) have to be loaded.
----------
Very very easy to reproduce. This experiment is for the slackers like me. :P
Take any off-the-shelf enameled copper wire spool.
It has to have his 2 ends available, of course. :)
Wrap it with some kitchen aluminium foil around it.
You have made a "capacitor".
Use a crocodile clip for connecting one "plate" of you cap.
Follow the picture and utilize your favorite Freq. Gen. and Oscilloscope.
Cheers,
Jean
Quote from: lost_bro on 2014.03.23, 19:54:08
Hello 4Tesla
As requested;
see below.
take care, peace
lost_bro
I have seen that used.
Thanks!
Is this the same idea, but transistor, not MOSFET?
Guys i sent the files this afternoon for Groundloops boards, so there will be no mods, but we can cut traces and glue on vero nods, until we reach a state where we could get some modded boards done.
Quote from: Matt Watts on 2014.03.23, 19:26:46
I can muddle through it GL, though I'm more comfortable with DipTrace.
Even though there are none on order.. please do draw out.
I have attached some possible test points.. feel free to add more.. also a couple more leads to VR1.
Thanks!
Quote from: 4Tesla on 2014.03.23, 20:52:05
Even though there are none on order.. please do draw out.
I have attached some possible test points.. feel free to add more.. also a couple more leads to VR1.
Thanks!
4Tesla,
I think it will be easier for you to just use a Veroboard. You are
defeating the purpose of a factory made board with so many
changes and loose ends. Just saying..............
GL.
@All,
4 boards claimed, 0 to go....................
GL.
Quote from: Groundloop on 2014.03.23, 21:13:42
4Tesla,
I think it will be easier for you to just use a Veroboard. You are
defeating the purpose of a factory made board with so many
changes and loose ends. Just saying..............
GL.
Well just the small section of the circuit.. but if no one likes the idea of this that is fine. I usually do test whole circuit on breadboard, but thought this was a compromise until we for sure know what is the best schematic in that area of the circuit. I just thinking if you're going to spend money on PCB, then at least be able to customize the section in question. The idea wasn't for me, but for those who like the PCB method.
Hi all,
I just subscribed to OUR. Glad to meet you all.
On Akula 30w, I'm making a breadboard first, but before testing, I'm making a simulation on Pspice, not to find out if the circuit is over unity, but to see what to expect in terms of currents and voltages. Some areas of the simulation are a little gray, mainly the transformer, but, anyway, I expect to gain some insight, that I will gladly post, as soon as it is ready.
Cheers,
Black Bird
Quote from: Black_Bird on 2014.03.23, 23:10:26
Hi all,
I just subscribed to OUR. Glad to meet you all.
On Akula 30w, I'm making a breadboard first, but before testing, I'm making a simulation on Pspice, not to find out if the circuit is over unity, but to see what to expect in terms of currents and voltages. Some areas of the simulation are a little gray, mainly the transformer, but, anyway, I expect to gain some insight, that I will gladly post, as soon as it is ready.
Cheers,
Black Bird
Welcome, BlackBird. Looking forward to results from your sim. Also, can you share the circuit schematic AS used for your sim?
Steve
Quote from: Black_Bird on 2014.03.23, 23:10:26
I just subscribed to OUR. Glad to meet you all.
On Akula 30w, I'm making a breadboard first, but before testing, I'm making a simulation on Pspice, not to find out if the circuit is over unity, but to see what to expect in terms of currents and voltages. Some areas of the simulation are a little gray, mainly the transformer, but, anyway, I expect to gain some insight, that I will gladly post, as soon as it is ready.
Welcome aboard.
Something worth taking a look at is a means to capacitively couple each winding of the outer loops to ground in an array type structure. Based on what Nerz has posted, this shielding like construct may instead be acting like small capacitors connected to each outer winding. You also have similar coupling between each winding, so mimicking in a simulation the physical device will likely be a little tricky. Certainly nothing wrong with coming at this phenomena from both sides--mathematically as well as actual prototypes. The truth is in there someplace.
Welcome to the forum Black Bird! O0
Quote from: PhysicsProf on 2014.03.23, 23:47:27
Welcome, BlackBird. Looking forward to results from your sim. Also, can you share the circuit schematic AS used for your sim?
Steve
Thanks, PhysicsProf. I can certainly share the schematics. Please just give me sometime to finish the simulation.
Cheers
Black Bird
Quote from: Matt Watts on 2014.03.24, 00:25:30
Welcome aboard.
Something worth taking a look at is a means to capacitively couple each winding of the outer loops to ground in an array type structure. Based on what Nerz has posted, this shielding like construct may instead be acting like small capacitors connected to each outer winding. You also have similar coupling between each winding, so mimicking in a simulation the physical device will likely be a little tricky. Certainly nothing wrong with coming at this phenomena from both sides--mathematically as well as actual prototypes. The truth is in there someplace.
Thank you, Matt.
I value simulations as a way to gain an insight on the circuit, but I do not expect to be able to simulate all aspects of the circuit, particularly those related to over unity, even because simulators work under "conventional physics" which might fail here, if it is the case. Also, I don't think my knowledge of Pspice is enough to simulate what you suggest, based on Nerz post. Anyway, let me advance on the simpler case and maybe later we can develop what you suggest.
Cheers,
Black Bird
Quote from: 4Tesla on 2014.03.24, 00:45:11
Welcome to the forum Black Bird! O0
Thank you 4Tesla.
Wow :o :) O0 Thank u very much dear Groundloop for making that PCB! No SMD is great.
@Peterae:
If its possible to ship to germany, I would take 2 Boards (+ 2 transformers if available). If prices are not like 100€ per piece I can also pay for 1-2 more boards+shipping for people here that maybe not can afford to buy one.
Im very happy and thankful btw I was allowed to join here with all u brilliant minds and tinkerers. Conversation quality and –climate is really high here. ;)
I still not started experimenting with this akula setup, but am busy with a similar device that seems promising so far. I also have a feeling that resonance (maybe different harmonics for the coils), capacitive coupling and phase shift between coils, like u guys already mentioned, are critical for the operation of this type of setup. And thinking of akula video where he shows the 3 one layer coils on top of each other and their scopeshots (harmonics). He deleted that already and i dont have link of copy now or file, but there is a stivep/Wesley copy with english translation on youtube. Will add later when i find. Here it is: https://www.youtube.com/watch?v=PPHipGkfSAY
Concerning simulations: Does anybody maybe have Ansoft Maxwell 3D (electromagnetic simulator) or similar program and knows how to use it? Could be of help...
kind regards
@All,
All four of my free boards is now claimed. Further board requests must go to Peter
here at this forum.
GL.
OK chaps
I had the quote come through for the pcb's
QuoteSpecs -2-Layer FR4 1.6mm 1oz HAL Tin-lead Green Soldermask White Silkscreen
dimensions - 83.8x106.4mm
20pcs - 76usd
30pcs - 93usd
PCB lead time is 5 workdays.
So all i need to know if wether i need 20 or 30 boards
@Other Planet
The boards are free, so if i send you 5 boards is that enough, I dont have any transformer's though, it's just PCB's from me O0
Thanks for the nice comments ;)
So any more orders for PCB's anyone ?
Quote from: Peterae on 2014.03.24, 07:19:57
OK chaps
I had the quote come through for the pcb's
So all i need to know if wether i need 20 or 30 boards
@Other Planet
The boards are free, so if i send you 5 boards is that enough, I dont have any transformer's though, it's just PCB's from me O0
Thanks for the nice comments ;)
So any more orders for PCB's anyone ?
You two guys are tops.
There are simply no excuses left for anyone still on the fence with this device. Even if it doesn't work, one can look at it daily and think about what to try next. That in and of itself is priceless. And if it does work, well... You saw it happen here first.
OK the boards are ordered, GL £76.00 for 30 boards including soldermask & Silkscreen O0
Quote from: Peterae on 2014.03.24, 08:33:28
OK the boards are ordered, GL £76.00 for 30 boards including soldermask & Silkscreen O0
Peter,
Thank you!
That was a nice low price. :-)
GL.
@All,
4Tesla requested a experiment prototype board. Attached is a design for
a single sided (only copper on the underside) board. This board is for
the experimenter and the builder must himself keep track on where the
wires should go. My main idea was that the user solder insulated wires
between the different areas. This board is for the experienced builder
that want full freedom to change the circuit and add parts at will.
I hope this is what you wanted.
Added: Updated the zip file. Corrected small error on PCB.
Added: Added missing regulator + input pads. Files updated.
Added: Corrected error on PCB, files updated.
Added: I do not recommend making this PCB until the error issue discovered by Itsu is solved!
Added: I have changed the design so that the user can decide on how to use the comparators.
This is the final update on this PCB.
Added: The above was not quite correct. I did realize that it would be handy to have some
areas on the PCB for MOSFET drivers etc. So I did split some tracks for that purpose.
Also the user have full control of the 494 pins now.
GL.
I will send you some boards as well GL O0
Let me know how many.
Quote from: Peterae on 2014.03.24, 11:57:35
I will send you some boards as well GL O0
Let me know how many.
Peter,
Thank you. I will need only one so that I can compare the quality
between the two boards. Do you still have my address?
GL.
Dear Black Bird.
Welcome to OUR :)
What a lot of ground covered whilst I was sleeping !!
Dear Groundloop, I like the way you have created the ultimate breadboard, possibilities, endless !!
Best wishes, cheers Grum.
Quote from: Grumage on 2014.03.24, 13:47:23
Dear Black Bird.
Welcome to OUR :)
What a lot of ground covered whilst I was sleeping !!
Dear Groundloop, I like the way you have created the ultimate breadboard, possibilities, endless !!
Best wishes, cheers Grum.
Grumage,
There is still room for improvement, I think. I will happily add changes to the layout if
someone has suggestions. But we have to wait until the overseas members do wake
up. Day for us but night for them. :-)
Thanks,
GL.
I have finished the control part (TL494IN) of the AKULA 30W circuit.
Frequency can be adjusted between 16 and 116KHZ at 12V with the 10K Ohm potmeter (too high?).
But the 5K potmeter does not seem to change the pulse width, allthough the voltage on pin 1 changes from 1V to 4V.
I seem to have a problem with the 12V voltage regulator as sometimes the voltage output on it drops from 12 to 10V
Most parts are as mentioned on the diagram, however a few resistors are the first close in the range (like 6.8 Ohm for the 7 Ohm etc)
I show the input voltage from a battery on a DMM and the output voltage of the 12V regulator also on a DMM.
Scope probes are used to measure:
# the output from pin 11 (after the pull up resistor) (yellow probe)
# the signal on pin 1 (used for the Pulse width) (blue probe)
# the signals on all the 16 pins of the TL494 one after the other (purple probe)
Video here: https://www.youtube.com/watch?v=R2r0en-kC3M&feature=youtu.be
Comments are welcome especially why the pulse width is not changing (is it because the pinned to ground via R7/R5 of pin 15?)
Regards itsu
Quote from: Itsu on 2014.03.24, 15:26:28
I have finished the control part (TL494IN) of the AKULA 30W circuit.
Frequency can be adjusted between 16 and 116KHZ at 12V with the 10K Ohm potmeter (too high?).
But the 5K potmeter does not seem to change the pulse width, allthough the voltage on pin 1 changes from 1V to 4V.
I seem to have a problem with the 12V voltage regulator as sometimes the voltage output on it drops from 12 to 10V
Most parts are as mentioned on the diagram, however a few resistors are the first close in the range (like 6.8 Ohm for the 7 Ohm etc)
I show the input voltage from a battery on a DMM and the output voltage of the 12V regulator also on a DMM.
Scope probes are used to measure:
# the output from pin 11 (after the pull up resistor) (yellow probe)
# the signal on pin 1 (used for the Pulse width) (blue probe)
# the signals on all the 16 pins of the TL494 one after the other (purple probe)
Video here: https://www.youtube.com/watch?v=R2r0en-kC3M&feature=youtu.be
Comments are welcome especially why the pulse width is not changing (is it because the pinned to ground via R7/R5 of pin 15?)
Regards itsu
Itsu,
I did misread the circuit drawing The two 4K are in series. So what you are seeing on pin 2 is VREF.
How can the comparator work if this voltage always is VREF?
GL.
Quote from: Groundloop on 2014.03.24, 16:12:55
Itsu,
I did misread the circuit drawing The two 4K are in series. So what you are seeing on pin 2 is VREF.
How can the comparator work if this voltage always is VREF?
GL.
Hello All:
I agree with Groundloop:
The pulse width modulator comparator provides a means
for the error amplifiers to adjust the output pulse width from
the maximum percent on−time, established by the deadtime
control input, down to zero, as the voltage at the feedback
pin varies from 0.5 V to 3.5 V. Both error amplifiers have a
common mode input range from −0.3 V to (VCC − 2V), and
may be used to sense power−supply output voltage and
current. The error−amplifier outputs are active high and are
ORed together at the noninverting input of the pulse−width
modulator comparator
. With this configuration, the
amplifier that demands minimum output on time, dominates
control of the loop.This is from ON-semi spec. sheet TL494.
Take care, peace
lost_bro
Quote from: lost_bro on 2014.03.24, 16:32:22
Hello All:
I agree with Groundloop:
The pulse width modulator comparator provides a means
for the error amplifiers to adjust the output pulse width from
the maximum percent on−time, established by the deadtime
control input, down to zero, as the voltage at the feedback
pin varies from 0.5 V to 3.5 V. Both error amplifiers have a
common mode input range from −0.3 V to (VCC − 2V), and
may be used to sense power−supply output voltage and
current. The error−amplifier outputs are active high and are
ORed together at the noninverting input of the pulse−width
modulator comparator. With this configuration, the
amplifier that demands minimum output on time, dominates
control of the loop.
This is from ON-semi spec. sheet TL494.
Take care, peace
lost_bro
lost_bro,
How can we fix this? I think there is a fault in the circuit drawing. My best guess is that
the R6 and R8 should be a voltage divider (1/2) for the VREF so that pin 2 on the IC
will have 1/2 the VREF. Then we can adjust the pot-meter R11 and change the duty cycle.
OR is the other comparator overriding the first one because of no positive feedback via R7?
GL.
Quote from: Groundloop on 2014.03.24, 16:39:23
lost_bro,
How can we fix this? I think there is a fault in the circuit drawing. My best guess is that
the R6 and R8 should be a voltage divider (1/2) for the VREF so that pin 2 on the IC
will have 1/2 the VREF. Then we can adjust the pot-meter R11 and change the duty cycle.
GL.
Hello Groundloop
This circuits works:
take care, peace
lost_bro
Itsu,
Can you try this:
Connect a pot-meter (10K) between VREF and GND. Feed the
center tap of the pot-meter to R7. (Ground to R7 removed).
Then try your duty cycle pot-meter.
This test pot-meter will simulate the output current that
flows through the R5 thus resulting av positive voltage
over the C5 electrolytic capacitor.
If my theory is correct then the duty cycle ON time should be
smaller when the voltage input to R7 is going UP.
GL.
Quote from: Groundloop on 2014.03.24, 16:39:23
lost_bro,
How can we fix this? I think there is a fault in the circuit drawing. My best guess is that
the R6 and R8 should be a voltage divider (1/2) for the VREF so that pin 2 on the IC
will have 1/2 the VREF. Then we can adjust the pot-meter R11 and change the duty cycle.
OR is the other comparator overriding the first one because of no positive feedback via R7?
GL.
Yes, I believe that is correct, as when I have used the TL494 I connect pin #14 directly to pin #15 as in the pic that I attached.
Does NOT make sense to have R6 between #14 and #15.
So R6 and R8 can be configured as a Voltage divider for Pin #2....
take care, peace.
lost_bro
Must be a drawing oversight on the schematic.......
Quote from: lost_bro on 2014.03.24, 16:56:17
Yes, I believe that is correct, as when I have used the TL494 I connect pin #14 directly to pin #15 as in the pic that I attached.
Does NOT make sense to have R6 between #14 and #15.
So R6 and R8 can be configured as a Voltage divider for Pin #2....
take care, peace.
lost_bro
Must be a drawing oversight on the schematic.......
lost_bro,
I would like to see Itsu test the "load simulator" first before I conclude about the error.
Itsu, du you have time to test a 10K pot-meter as I described above?
GL.
Quote from: Groundloop on 2014.03.24, 17:11:32
lost_bro,
I would like to see Itsu test the "load simulator" first before I conclude about the error.
Itsu, du you have time to test a 10K pot-meter as I described above?
GL.
Hi GL, yes i can do that.
I will report back what i found, but it takes some time (hours)
Regards Itsu
Quote from: Groundloop on 2014.03.24, 16:39:23
lost_bro,
How can we fix this? I think there is a fault in the circuit drawing. My best guess is that
the R6 and R8 should be a voltage divider (1/2) for the VREF so that pin 2 on the IC
will have 1/2 the VREF. Then we can adjust the pot-meter R11 and change the duty cycle.
OR is the other comparator overriding the first one because of no positive feedback via R7?
GL.
Hi GL,
My simulation (still an ongoing process) showed exactly that. Pin 2 on tl494 has to have an additional 4.7 k resistor to ground for the pwm to work.
Cheers,
Black Bird
Quote from: Itsu on 2014.03.24, 18:25:05
Hi GL, yes i can do that.
I will report back what i found, but it takes some time (hours)
Regards Itsu
Itsu,
Thanks. Looking forward to your test.
GL.
Quote from: Black_Bird on 2014.03.24, 19:18:52
Hi GL,
My simulation (still an ongoing process) showed exactly that. Pin 2 on tl494 has to have an additional 4.7 k resistor to ground for the pwm to work.
Cheers,
Black Bird
Black Bird,
Thanks for doing simulations.
The extra resistor can easily be added to the PCB by soldering it on the underside of the PCB.
GL.
Ok, first tried Black_Birds solution by simple adding a 3.9K resistor from pin 2 to ground.
Indeed the voltage on pin 2 now halfs to 2.5V, but the 5K potmeter only reacts in the middle and is very touchy.
Duty cycle switches from 55% in a fraction of a turn to no signal (flat +12V) at all.
No way to set it inbetween (55 -> 1).
Then i tried GL's solution, by disconnecting R7 from the link to R5 (ground) and attaching there the middle tap of a (in my case ) 20K pot-meter.
The both ends of this pot-meter where connected to pin 14 (REF) and pin 7 (ground).
I see the voltage on this center tap (and thus on pin 15) vary between 5V and ground, but no change in the duty cycle. :(
Regards itsu
Quote from: Itsu on 2014.03.24, 21:28:58
Ok, first tried Black_Birds solution by simple adding a 3.9K resistor from pin 2 to ground.
Indeed the voltage on pin 2 now halfs to 2.5V, but the 5K potmeter only reacts in the middle and is very touchy.
Duty cycle switches from 55% in a fraction of a turn to no signal (flat +12V) at all.
No way to set it inbetween (55 -> 1).
Then i tried GL's solution, by disconnecting R7 from the link to R5 (ground) and attaching there the middle tap of a (in my case ) 20K pot-meter.
The both ends of this pot-meter where connected to pin 14 (REF) and pin 7 (ground).
I see the voltage on this center tap (and thus on pin 15) vary between 5V and ground, but no change in the duty cycle. :(
Regards itsu
Itsu,
Thank you for testing. Keep your 20K "current simulator" connected and try both, add the voltage divide
resistor and then try out different settings for the 20K and the duty cycle pot-meter.
GL.
Quote from: Groundloop on 2014.03.24, 21:35:55
Itsu,
Thank you for testing. Keep your 20K "current simulator" connected and try both, add the voltage divide
resistor and then try out different settings for the 20K and the duty cycle pot-meter.
GL.
No problem, ok, both solutions together, 2.5V on pin 2 and variable voltage (5 - 0v) on pin 15, but it does not matter how i set this pin 15 voltage, the 5K pot-meter still shows this sudden switch from 55% dc to flat 12V.
Added, this "switch" happens with exactly 2.5V on pin 1
Regards Itsu
Quote from: Itsu on 2014.03.24, 21:28:58
Ok, first tried Black_Birds solution by simple adding a 3.9K resistor from pin 2 to ground.
Indeed the voltage on pin 2 now halfs to 2.5V, but the 5K potmeter only reacts in the middle and is very touchy.
Duty cycle switches from 55% in a fraction of a turn to no signal (flat +12V) at all.
No way to set it inbetween (55 -> 1).
Then i tried GL's solution, by disconnecting R7 from the link to R5 (ground) and attaching there the middle tap of a (in my case ) 20K pot-meter.
The both ends of this pot-meter where connected to pin 14 (REF) and pin 7 (ground).
I see the voltage on this center tap (and thus on pin 15) vary between 5V and ground, but no change in the duty cycle. :(
Regards itsu
Hi Itsu,
Try to ground pin 15 directly, without the 270 ohm resistor. I'm having problems in the simulation when the 270 ohm is in place.
Cheers,
Black Bird.
Black_Bird, ok only your solution, but then with pin 15 grounded directly.
Still the same, at 2.5V on pin 1 it flips over from 55% dc to flat +12V
When i remove both solutions and short pins 2 and 3, then i can vary the dc between 65% and 95%
Ok enough for today, thanks for all your suggestions.
Regards Itsu
Quote from: Itsu on 2014.03.24, 22:03:33
No problem, ok, both solutions together, 2.5V on pin 2 and variable voltage (5 - 0v) on pin 15, but it does not matter how i set this pin 15 voltage, the 5K pot-meter still shows this sudden switch from 55% dc to flat 12V.
Added, this "switch" happens with exactly 2.5V on pin 1
Regards Itsu
Itsu,
OK, thanks for testing. We need at some point to sort out what changes to make
to get a correct duty cycle regulation both when tuning R11 and on the feed back
line through R7.
GL.
Welcome also, FromOther Planet!
@all - good comments.
PS - Jean -- keep up those studies.
@All,
I think this is the solution to the duty cycle problem:
Do not use C2 and C3.
Do not use R5, R8 and R9.
Solder wire at C3 and R8.
Solder R15 standing vertically on the PCB connected to R13.
Solder wire from to of R15 to VREF on pin 14 on TL494.
Add 100nF capacitor from pin 1 to ground.
I referencing to the component labels on MY posted design, the one we are making PCB to.
After the modification the circuit will look like the attached drawing.
I have changed the layout for the experiment board posted one page back.
Comments.................???
GL.
Are you sure you want duty cycle variable ? can you explain why ? I think it is not needed, just frequency regulation is enough.
Quote from: forest on 2014.03.25, 07:07:24
Are you sure you want duty cycle variable ? can you explain why ? I think it is not needed, just frequency regulation is enough.
Hi Forest,
It is not about what I want, but about what Akyla0083 wanted when he designed his circuit.
So my first attempt to replicate this circuit will be using the same parts (or a close as possible)
on the PCB and see what happens. Later on we can change tings around.
GL.
Do you have Multisim 12 or do you able to get it to test simulation ? C.C
I like what user MenofFather says on overunity.com here:
http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg394343/#msg394343
He points out that this R5 (and C5) in the original diagram could be an automatic duty cycle adjuster depending on the load, he says
as a reaction on this post from T1000:
T1000
QuoteP.S> The function of C5 + R5 is still not clear to me in original circuit. Perhaps the intention was to manipulate duty cycle with this feedback path?
MenofFather:
QuoteHow I understand, this function change duty cycle depending on load. Wihout load 0 precents duty cycle, with small load let say 20 precents, with big lload let say 60 precents.
So a changing voltage on pin 15 via R7 could/should vary the d.c.
And its an essential part of the circuit as it regulates depending on the current drawn.
Any thoughts?
Regards Itsu
Quote from: Itsu on 2014.03.25, 09:28:47
I like what user MenofFather says on overunity.com here:
http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg394343/#msg394343
He points out that this R5 (and C5) in the original diagram could be an automatic duty cycle adjuster depending on the load, he says
as a reaction on this post from T1000:
T1000
MenofFather:
So a changing voltage on pin 15 via R7 could/should vary the d.c.
Any thoughts?
Regards Itsu
Itsu,
Yes, that is also my understanding of the R5/C5. The voltage over C5 will be more positive
as the current through R5 increases. This positive voltage is feed back via R7 to the negative
input of the error amplifier 2. The positive pin of the error amplifier 2 is grounded.
Does your 20K pot-meter has any impact on the duty cycle if you wire up your circuit as
shown in my posted image above?
GL.
Quote from: Groundloop on 2014.03.25, 09:39:55
Does your 20K pot-meter has any impact on the duty cycle if you wire up your circuit as
shown in my posted image above?
Well, that's not that easily done, as it needs substantial (re)moving parts/connections on my tightly soldered print :(
Perhaps its faster done in a simulator.
Would this feedback link to pin 15 not be more appropriate to be connected to pin 3 (feedback)?
Anyway, it will have to wait till tonight as i am at work right now.
Regards Itsu
Quote from: Itsu on 2014.03.25, 09:48:49
Well, that's not that easily done, as it needs substantial (re)moving parts/connections on my tightly soldered print :(
Perhaps its faster done in a simulator.
Would this feedback link to pin 15 not be more appropriate to be connected to pin 3 (feedback)?
Anyway, it will have to wait till tonight as i am at work right now.
Regards Itsu
Itsu,
OK, I understand.
>>Would this feedback link to pin 15 not be more appropriate to be connected to pin 3 (feedback)?
I do not know. I hope those simulator guys can give it a try.
GL.
Quote from: Itsu on 2014.03.25, 09:28:47
I like what user MenofFather says on overunity.com here:
http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg394343/#msg394343
He points out that this R5 (and C5) in the original diagram could be an automatic duty cycle adjuster depending on the load, he says
as a reaction on this post from T1000:
T1000
MenofFather:
So a changing voltage on pin 15 via R7 could/should vary the d.c.
Any thoughts?
Regards Itsu
Hi Itsu,
If we look into the TL494 specs, we see that each amplifier has a 90dB+ gain at DC. So, for DC, the amplifier at pins 1 and 2 is operating at full gain, because of capacitor C7. The same applies to amplifier at pins 15 and 16. This explains the jump in the output, because the gain is so high.
There are some possibilities here:
1) a resistor is missing between pins 2 and 3, so that the DC gain will be lower at DC. In this case, R12+C7 will further limit the gain at AC, starting at around 270 Hz (=1/(2*PI*R12*C7) ).
2) R12 should be connected directly from pin3 to pin 2 and capacitor C7 to from pin 2 to ground.
I'm running simulations on these two situations and let you know what I found.
Cheers,
Black Bird
Sounds good Black Bird, let me know the outcome, so i can put some changes in tonight.
Regards Itsu
You guys are sure to nail this down soon, but I have to ask, are you sure it wouldn't make sense to just utilize a pulse generator like this one to drive the looped portion of the circuit:
(http://www.bkprecision.com/photos/?src=images/products/photos/large/4030_front_lrg.jpg)
https://www.bkprecision.com/downloads/datasheets/4030_datasheet.pdf
You would have full control of frequency, duty cycle and could probably get the coil to respond once it's wound correctly.
I'm probably off my rocker. Just seems like this is becoming a hair more difficult than it needs to be. I've always tried to take what I know works and use it to figure out the things I don't know. If we can get something dialed-in, we can always go back and add the components that mimic the pulse generator.
Please take no offense. I'm just curious if I'm the only one that has a similar idea.
I must agree with Matt, using bench equipment to nail down the operating parameters of the coil seems a very flexible approach for those who have the equipment. Understanding the "magic of the coil" if any exists seems the best route, lest we become a "cargo cult".
Both routes are fine, but understanding is key.
Dear ION.
Welcome aboard !! :)
Your extensive knowledge and expertise will be a great help to this thread.
Best wishes, Grum.
@all
I asked Wesley if he can talk or email Akula for the winds.
While you guys ponder the circuit (I told you guys not to make it yet....but boys will be boys - hehehe), I am preparing a Standard Procedure for OU Coil Winding.
Point is forget what Akula says about 15/45, it means nothing but a range of turns, so you don't make a coil with 100/400. The 15/45 is a 1:3 ratio. The only real fixed component is the E core. The primary and secondary must be wound as per the characteristics of both the wires used and the number of turns required to produce not more and not less of an impulse. My idea is if everyone followed a set procedure to make the primary and secondary winding, your coil will be well balanced regardless of the turns or core type since it will match your core limitations.
15 turns primary. So what if with only 11 turns you get the same or better output at 36% less impress cost plus 36% less mosfet killing flyback plus better core to wind sensitivity. What if 15 turns is saturating your core at only a half impulse and with that other half excess being over the core limit seizes the reactivity of the core. Some are saying their cores have cracked already. Wondering why????
I think it's about time that we started to work smarter on both sides of the pond. With such an OU based standard, our Russian friends or any builder can simply say "This transformer was wound using XXXXX standard except step 3 where I started at 2 turns". Then provide some core, wire specs and possible resonance frequencies as a guide to others. This way anyone in the world will be able to come very close to the same transformer build regardless of number of turns since the main builder used the same standard to wind his coils with which he obtained the results that others are looking to now replicate.
Many here have been around the block a few hundred times already and can attest to the fact that although all come on board with the best intentions, that is not enough. Good intentions are great but making sure your efforts are not unknowingly thwarted by actions based on a lack of foresight is worth gold to any such enterprise. What we want is not winding up again at page 345 by playing musical chairs.
One observation like;
1) @Grumage, your coil wire type seems to have its plastic outer coating to be very soft and in a transformer dishing out 30 watts, it will risk shorting in the center winds possible then making your drive circuit overheat and blow. So make sure the wire you are using has a good coating on it that will withstand some heat.
About the standards, if we can have a good 4-5 standards that cover a good range of building methods, then the actual information resulting from these should help everyone stay on the same page.
I am presently working on the LTPU/FTPU and will have some great discoveries to announce soon enough. But I will place the procedure in this thread and at my Understanding Overunity thread since this applies to everyone.
All the best.
wattsup
Dear Wattsup.
I have the greatest respect for your input and agree with your sentiments. YES we could really do with the winding schedule it would make our quest so much easier !!
I also agree with your sentiments as regards my transformer windings. However, for the "Looking for anomalies" phase they are probably adequate to perform that function.
I do really feel that the preliminary work being done by Itsu and the guys will hold us in great stead when the time comes to hook up the two parts together.
In the mean time I would just simply like to say that you're all doing a great job, many thanks. O0
Cheers Grum.
@ all
If and when we receive some firm information on the coil winding, I shall initially breadboard the TL494, Vreg and mosfet circuitry. I'm tempted to put everything on strip-board, as I'm experienced in using this medium. However, its not an advisable method for novices in electronics and as Wattsup suggests, its sensible that we all use the same build method. If we were to all initially use a breadboard approach, then a Janet & John row and column listing could, if deemed necessary, be drawn up for the necessary component placing and inter-connections. Those that will be receiving a PCB (kindly free issued by Groundloop and Peterae) may be best advised to populate this only when some initial groundwork has been carried out on the full build, to verify that there is either no necessity to deviate from Akula's design, or that any necessary variation(s) can be accomodated easily on the PCB. Just my thoughts.
Hoppy
Grum, Peterae, u got my PM?
regards
Hello Gentlemen,
In the Akula's circuit, this aluminium foil wraped (but not short circuited) around the trafo is strange.
Is it not? So, why not testing it? Just 5 pictures. No comments:
(http://www.freenrg.info/4OUR_F/Akula_Detail2.jpg)
(http://www.freenrg.info/4OUR_F/Another_Exp.jpg)
(http://www.freenrg.info/4OUR_F/Coils_Cap_1.jpg)
(http://www.freenrg.info/4OUR_F/Coils_Cap_2SMod.jpg)
(http://www.freenrg.info/4OUR_F/Coils_Cap_3M.jpg)
Le bonjour vous sied,
Jean
Quote from: Itsu on 2014.03.24, 15:26:28
But the 5K potmeter does not seem to change the pulse width, although the voltage on pin 1 changes from 1V to 4V.
Oh, C'mon! You've connected the TL494 to adjust its duty cycle with a pot, before. See
your video (http://youtu.be/ACNzggg1QC0?t=9m41s) and the schematic
here (http://www.overunity.com/12736/kapanadze-cousin-dally-free-energy/msg340106/#msg340106).
Quote from: Itsu on 2014.03.24, 15:26:28
I seem to have a problem with the 12V voltage regulator as sometimes the voltage output on it drops from 12 to 10V
Of course you do, the minimum difference in voltage between the input and output of the linear voltage regulator (V
DO) should be respected as stated in its datasheet.
This circuit has a high yawn factor. The mystery is in the E-core inductor/transformer.
Quote from: verpies on 2014.03.25, 20:47:03
Oh, C'mon! You've connected the TL494 before to adjust its duty cycle with a pot. See your video (http://youtu.be/ACNzggg1QC0?t=9m41s) and the schematic here (http://www.overunity.com/12736/kapanadze-cousin-dally-free-energy/msg340106/#msg340106).
Correct, but that was with a different diagram. We are now trying to get this specific AKULA 30W diagram to work without changing to much, else it would not be called a replication.
QuoteOf course you do, the minimum difference in voltage between the input and output of the linear voltage regulator (VDO) should be respected as stated in its datasheet.
This circuit has a high yawn factor. The mystery is in the E-core inductor/transformer.
Yep, you are right, when using my bench PS, the 12V regulator output sags from 12V to 10V when the input goes below 14V.
Regards Itsu
Quote from: Black_Bird on 2014.03.25, 11:39:47
Hi Itsu,
If we look into the TL494 specs, we see that each amplifier has a 90dB+ gain at DC. So, for DC, the amplifier at pins 1 and 2 is operating at full gain, because of capacitor C7. The same applies to amplifier at pins 15 and 16. This explains the jump in the output, because the gain is so high.
There are some possibilities here:
1) a resistor is missing between pins 2 and 3, so that the DC gain will be lower at DC. In this case, R12+C7 will further limit the gain at AC, starting at around 270 Hz (=1/(2*PI*R12*C7) ).
2) R12 should be connected directly from pin3 to pin 2 and capacitor C7 to from pin 2 to ground.
I'm running simulations on these two situations and let you know what I found.
Cheers,
Black Bird
Black Bird, i tried the first option and put a 20K pot-meter between pins 2 and 3.
Now i can manipulate the duty cycle between 56% and 100% by varying both this 20K pot and the Original 5K pot.
It never will go below 56%.
Regards Itsu
Quote from: Itsu on 2014.03.25, 20:57:37
Correct, but that was with a different diagram. We are now trying to get this specific AKULA 30W diagram to work without changing to much, else it would not be called a replication.
Yep, you are right, when using my bench PS, the 12V regulator output sags from 12V to 10V when the input goes below 14V.
Regards Itsu
So far, a couple of quite serious 'yawns' have been identified with Akula's circuit. However, you are right that Akula's circuit should be replicated as closely as possible and this means that full coil details are essential as its here as Verpies suggests that any magic is bottled-up.
Regards
Hoppy
Quote from: Itsu on 2014.03.25, 20:57:37
Correct, but that was with a different diagram.
Which diagram is the most up to date??
Oh there are many, too many.
I am using the diagram (don't know if its the latest) as can be found in this thread on page 9 in post #204 from Groundloop:
http://www.overunityresearch.com/index.php?topic=2358.msg36967#msg36967
We do work also on this magical transformer now and then, at least i do, and i know Grumage did some testing and he made his squeel, but i did not find any abnormalities (no static electricity either).
See short video here: https://www.youtube.com/watch?v=1mPy2rjDkhE&feature=youtu.be
See also a screenshot of the signal across the bulb while squeeling on 2.9KHz
Regards Itsu
Quote from: Itsu on 2014.03.25, 22:01:24
Oh there are many, too many.
I am using the diagram (don't know if its the latest) as can be found in this thread on page 9 in post #204 from Groundloop:
http://www.overunityresearch.com/index.php?topic=2358.msg36967#msg36967
We do work also on this magical transformer now and then, at least i do, and i know Grumage did some testing and he made his squeel, but i did not find any abnormalities (no static electricity either).
See short video here: https://www.youtube.com/watch?v=1mPy2rjDkhE&feature=youtu.be
See also a screenshot of the signal across the bulb while squeeling on 2.9KHz
Regards Itsu
Itsu,
I couldn't hear the 10KHz either. That says something about our age. LOL
@All,
I have updated the "Prototype Board" design here: http://www.overunityresearch.com/index.php?topic=2358.250 / Reply #270
This will be my final design. This board can probably be used to other projects also. I have also split some tracks so that it
it is possible to solder in 555 or MOSFET drivers (8 pin DIL) types. It is also possible to plug a 555 into the top part of the
socket because all 8 pins here is isolated from the rest of the tracks. The board is single sided so all copper tracks is on
the underside of the board. The prototype tracks are all marked on the top side of the board. It is my hope that this board will
inspire people to do hobby research on switching power circuits. Even if we do not find any OU, there will be hours of fun hobby
electronic work. But who knows, maybe we can find a circuit that define some of the man made laws regarding magnetic fields etc.
I will give away 8 boards of this design for free. I can snail mail everywhere in the world. When I said free, I mean free, so I
will also pay for the postal charges. The boards will arrive in the week before Easter. The first
8 people that PM me a snail mail
address and country location will have a free board. If you want to stay anonymous then rent a P.O. box or something, but I
will need
a real postal address. Have fun.............. :-)
Added:
8 PCB claimed,
0 boards left!
GL.
Quote from: Itsu on 2014.03.25, 22:01:24
We do work also on this magical transformer now and then, at least i do, and i know Grumage did some testing and he made his squeel, but i did not find any abnormalities (no static electricity either).
You won't see any anomalies in this test case because in way it is made and connected. Three things should be there from my point of view:
1) The load must be LEDs, no any other resistive/inductive load;
2) The transformer core must have gap in all 3 sections, not only in 2;
3) The transformer is supposed to be connected like choke as per schematics and both windings should end up in series with capacitor between them when BEMF pulse goes on. Your LEDs should be attached like there too.
Hopefully that helps.
Cheers!
Quote from: T-1000 on 2014.03.26, 00:56:53
2) The transformer core must have gap in all 3 sections, not only in 2;
So that means, the core must be an E-Core with the center gapped as well as the two outer ends gapped. Is that correct? Would that mean to use a non-gapped core with some sort of spacer between all three surfaces?
Thanks much for this info T.
Quote from: 4Tesla on 2014.03.23, 21:21:33
Well just the small section of the circuit.. but if no one likes the idea of this that is fine. I usually do test whole circuit on breadboard, but thought this was a compromise until we for sure know what is the best schematic in that area of the circuit. I just thinking if you're going to spend money on PCB, then at least be able to customize the section in question. The idea wasn't for me, but for those who like the PCB method.
4Tesla,
OK, now that you got your whish come true, do you want a board?
GL.
Quote from: Itsu on 2014.03.25, 21:01:30
Black Bird, i tried the first option and put a 20K pot-meter between pins 2 and 3.
Now i can manipulate the duty cycle between 56% and 100% by varying both this 20K pot and the Original 5K pot.
It never will go below 56%.
Regards Itsu
Hi Itsu,
Have you also added the 4.7k resistor from pin 2 to ground?
As I said before, I was having problems with the simulation until I also grounded pin 15. When I did this, I could see the circuit working and I found a very interesting fact, which is that the circuits shows two frequencies: the first one is the switching frequency and another one, at much lower frequency, which is directly related to the value of the output filter capacitor C3 and the inductor at the input of the 7812 regulator. For C3=2200uF and the inductor = 100uH, this frequency is about 340Hz, and it appears in the transformer current as kind of a sawtooth. If the inductor is increased to 500uH, this frequency lowers to 115Hz, with the switching frequency unchanged.
Many of the alleged working OU circuits mention using two frequencies mixed at the output transformer, and it is interesting that this circuit apparently does the same. This triggers a thought that it might be that the "magical" effects on the transformer will only appear when it is connected to the real circuit.
Concerning the load, I ran simulations with a resistor and with LEDs and did not see any apparent difference, so far, but let's not forget this is only a simulation, not the real thing.
Cheers
Black Bird
PS: I modified the position of C3 according to Menofather information to the junction of the diode that goes to the battery and the inductor. keeping it in the original position give rise to currents of more than 60 amps, and changing the position as described keeps currents below 25 amps.
Quote from: T-1000 on 2014.03.26, 00:56:53
You won't see any anomalies in this test case because in way it is made and connected. Three things should be there from my point of view:
1) The load must be LEDs, no any other resistive/inductive load;
2) The transformer core must have gap in all 3 sections, not only in 2;
3) The transformer is supposed to be connected like choke as per schematics and both windings should end up in series with capacitor between them when BEMF pulse goes on. Your LEDs should be attached like there too.
Hopefully that helps.
Cheers!
thanks for the info T1000,
it was just a quick setup to show what this transformer can do when hard driven by a 12V switching MOSFET.
I know others are also experimenting with the x-former and wanted to encourage them to show what they are doing.
Good to know that only LEDs must be the load, i have ordered some.
As the two outer cores of my e-core had a taped gap, this automatically means the inner core had a similar air gap.
Thanks, regards Itsu
Quote from: Black_Bird on 2014.03.26, 10:04:24
Hi Itsu,
Have you also added the 4.7k resistor from pin 2 to ground?
As I said before, I was having problems with the simulation until I also grounded pin 15. When I did this, I could see the circuit working and I found a very interesting fact, which is that the circuits shows two frequencies: the first one is the switching frequency and another one, at much lower frequency, which is directly related to the value of the output filter capacitor C3 and the inductor at the input of the 7812 regulator. For C3=2200uF and the inductor = 100uH, this frequency is about 340Hz, and it appears in the transformer current as kind of a sawtooth. If the inductor is increased to 500uH, this frequency lowers to 115Hz, with the switching frequency unchanged.
Many of the alleged working OU circuits mention using two frequencies mixed at the output transformer, and it is interesting that this circuit apparently does the same. This triggers a thought that it might be that the "magical" effects on the transformer will only appear when it is connected to the real circuit.
Concerning the load, I ran simulations with a resistor and with LEDs and did not see any apparent difference, so far, but let's not forget this is only a simulation, not the real thing.
Cheers
Black Bird
PS: I modified the position of C3 according to Menofather information to the junction of the diode that goes to the battery and the inductor. keeping it in the original position give rise to currents of more than 60 amps, and changing the position as described keeps currents below 25 amps.
Hi Black Bird, no i did not add the 4.7k resistor from pin 2 to ground. Where did that come from?
For the rest of your comments i can not respond to as i do not have those components installed, so i do not see any "lower frequency".
But i agree that probably we need to setup the WHOLE circuit to be able to test the proper functioning of the control section (TL494).
Thanks, regards Itsu
Quote from: Itsu on 2014.03.26, 10:33:37
Hi Black Bird, no i did not add the 4.7k resistor from pin 2 to ground. Where did that come from?
For the rest of your comments i can not respond to as i do not have those components installed, so i do not see any "lower frequency".
But i agree that probably we need to setup the WHOLE circuit to be able to test the proper functioning of the control section (TL494).
Thanks, regards Itsu
The 4.7k resistor from pin 2 to ground is to keep the voltage of pin 2 at about 2.5V, below the threshold voltage of the sawtooth that appears at pin 5.
Cheers,
Black Bird
Dear All.
I have just come on line to find my dream has been interpreted by T-1000 !! :)
I saw this device as not being a transformer but as two separate chokes sharing a common core. Our load consisting of LED's is teed off L1 creating a partial saturation. L2 is our recovery coil.
I have a simple question. When I look at LED's that are fitted to all sorts of appliances around my home, if I look at them through the corner of my eye they appear to be flashing very rapidly !! Is this the case, do LED's flicker at high frequency as a matter of normal operation ? Or do they flicker because the drive to them is pulsed ?
If the answer is yes to the former then maybe there is an interaction between the LED's and the drive pulse ??
Comments/thoughts welcome.
Cheers Grum.
Quote from: Black_Bird on 2014.03.26, 10:37:25
The 4.7k resistor from pin 2 to ground is to keep the voltage of pin 2 at about 2.5V, below the threshold voltage of the sawtooth that appears at pin 5.
Cheers,
Black Bird
Ok, got it, that came from your #284 post. It is in my case a 3.9K Ohm resistor.
Will try that tonight.
Regards Itsu
Quote from: Itsu on 2014.03.25, 22:01:24
I am using the diagram (don't know if its the latest) as can be found in this thread on page 9 in post #204 from Groundloop:
Then in
this diagram (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) the potentiometer R11 is not there to adjust the duty cycle directly but to control the integration threshold of the inverting integrator formed by the 1
st error amplifier (pins 1 & 2) and C7.
The 2
nd error amplifier (pins 15 & 16) is configured as a voltage comparator and will be activated only when the voltage on pin 15 is pulled
below ground by R7.
Such negative voltage is impossible with ideal diodes connected as a load in such direction and with a non-inductive CSR (R5).
Diodes can briefly conduct backwards due to parasitic capacitances and the
step recovery (http://en.wikipedia.org/wiki/Step_recovery_diode) effect (SRD and DSRD) ...a non-ideal diode behavior. In any case, the C5 will integrate out any such brief reverse current pulses.
Quote from: Itsu on 2014.03.25, 22:01:24
We do work also on this magical transformer now and then, at least i do, and i know Grumage did some testing and he made his squeal...
Squealing has the potential to modulate the permeability of the core via the Villari effect (http://en.wikipedia.org/wiki/Villari_effect), but for this to happen the acoustic wavelengths should approximate the size of this small ferrite E-core. Smaller wavelengths mean higher frequency.
Quote from: Grumage on 2014.03.26, 10:55:26
...do LED's flicker at high frequency as a matter of normal operation ?
LEDs do not flicker as a matter of normal operation on DC current.
Quote from: Grumage on 2014.03.26, 10:55:26
Or do they flicker because the drive to them is pulsed ?
Yes. Pulsed deliberately or inadvertently by supplying them with AC.
Quote from: verpies on 2014.03.26, 11:48:39
Then in this diagram (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) the potentiometer R11 is not there to adjust the duty cycle directly but to control the integration threshold of the inverting integrator formed by the 1st error amplifier (pins 1 & 2) and C7.
Ok, good to know.
QuoteThe 2nd error amplifier (pins 15 & 16) is configured as a voltage comparator and will be activated only when the voltage on pin 15 is pulled below ground by R7..
Ok, so i could simulate this by applying a negative voltage to R7
QuoteSuch negative voltage is impossible with ideal diodes connected as a load in such direction and with a non-inductive CSR (R5).
Diodes can briefly conduct backwards due to parasitic capacitances and the step recovery (http://en.wikipedia.org/wiki/Step_recovery_diode) effect (SRD and DSRD) ...a non-ideal diode behavior. In any case, the C5 will integrate out any such brief reverse current pulses.
Ok, but there are no such things as ideal diodes, right? So it could exploit that together with an inductive CSR (R5).
QuoteSquealing has the potential to modulate the permeability of the core via the Villari effect (http://en.wikipedia.org/wiki/Villari_effect), but for this to happen the acoustic wavelengths should approximate the size of this small ferrite E-core. Smaller wavelengths mean higher frequency.
Yes, i have to increase the frequency to check it out if something abnormal happens, but will be needed the leds first.
Many thanks, regards Itsu
Quote from: Itsu on 2014.03.26, 12:20:05
Ok, so i could simulate this by applying a negative voltage to R7
You could do that, if you want to verify that the 2
nd error amplifier/comparator functions properly.
If the test is successful, then the output duty cycle should decrease almost to 0% immediately.
To test of the operation of the 1
st error amplifier/integrator, a different test should be made.
I don't know if you want to know about it, so I will not spew another paragraph here.
Quote from: Itsu on 2014.03.26, 12:20:05
Ok, but there are no such things as ideal diodes, right?
Yes, but exploiting non-ideal characteristics of diodes is considered a poor design and hard to replicate with different diodes
Exception could be made for the brief reverse conduction during the DSR effect (as in your nanopulser) as well as the Zener effect and Avalanche reverse breakdown of some diodes in some cases.
Quote from: Itsu on 2014.03.26, 12:20:05
So it could exploit that together with an inductive CSR (R5).
An inductive CSR (R5) would form a parallel LC tank with the nearby capacitor (C5), which would add another variable to the tuning of this device.
Quote from: Itsu on 2014.03.26, 12:20:05
Yes, i have to increase the frequency to check it out if something abnormal happens, but will be needed the leds first.
You can estimate this frequency if you know the size of your core and speed of sound in it, e.g. 5km/s.
Quote from: verpies on 2014.03.26, 13:04:50
You could do that, if you want to verify that the 2nd error amplifier/comparator functions properly.
If the test is successful, then the output duty cycle should decrease almost to 0% immediately.
To test of the operation of the 1st error amplifier/integrator, a different test should be made.
I don't know if you want to know about it, so I will not spew another paragraph here.
Ok, great, let me first test this negative voltage on R7 to see what it does to the d.c.
Then we always can go into more depth with the test of the operation of the 1
st error amplifier/integrator.
Regards Itsu
May I suggest an extension or variation of ND's initial experiment (post 311) to attempt electrostatic excitation of a coil, however in this approach with electrostatic plates at the 1/4 wave points driven out of phase.
L2 can also be treated as a transmission line with appropriate terminations, resistive, capacitive, open or shorted line. We know that in a transmission line containing one complete wave, there are two nodes where voltage peaks of opposite polarity, we wish to drive the voltage nodes.
L2 can also be wound as a toroid instead of linear coil, SM style.
Quote from: Grumage on 2014.03.25, 15:59:53
Dear ION.
Welcome aboard !! :)
Your extensive knowledge and expertise will be a great help to this thread.
Best wishes, Grum.
Dear Grum
I have never been overboard but have been spending some time away building a new lab while I ponder this thread from time to time. There appear to be many new and knowledgeable individuals posting here, and that is refreshing.
Cheers ION
@All,
No more free prototype boards left.
Peter, I will send you one of the new prototype boards if you want one.
GL.
@Groundloop
I am hesitant in asking you for one because I feel I do not have the skills required to solder all of those fine points. Maybe I can convince my brother to do it for me, or, just hire someone to do it for me and save me my precious bench time. Time is what we all need more of. hehehe
@all
Before I publish a resonance procedure, I am working on mapping the Akula E-core image to demystify that question once and for all and hopefully this will free up that question if it conforms to your vision. One little confusion at the start can send you on side angles for months. Months that we need to save.
About Wesley, he indicated he has not communicated with Akula since 1.5 months and does not know when his next discussion with him will happen. Maybe a Russian friend can ask some questions on their forum to see if Akula will respond about the coil. I would not risk doing this myself with a translation program to avoid the risk of offending anyone there. Something like wanting to ask "Hey you guys, how are things moving along!" Instead the program could produce "Hey you guys. Shove it!" hahahahahahaaha
That would not be good for Russian/Canadian relations.
wattsup
Added: Be warned. My next post will be a long one. hahaha
Quote from: Black_Bird on 2014.03.26, 10:37:25
The 4.7k resistor from pin 2 to ground is to keep the voltage of pin 2 at about 2.5V, below the threshold voltage of the sawtooth that appears at pin 5.
Cheers,
Black Bird
Ok, i put a 3.9K resistor from pin 2 to ground so that i have 2.5V on pin 2.
Then i put again my 20K pot-meter center tap and a normal tap between pins 2 and 3, but same thing happens as without the 3.9K to ground.
I can vary the d.c. between 56% and 100%. by manipulating the 20K and the Original 5K pot-meters.
Never it goes below 56% d.c.
Regards Itsu
Quote from: verpies on 2014.03.26, 13:04:50
You could do that, if you want to verify that the 2nd error amplifier/comparator functions properly.
If the test is successful, then the output duty cycle should decrease almost to 0% immediately.
Ok, negative variable voltage on pin 15 via R7, but at the slightes negative value (milivolts) the d.c. flips to 100% (from the 56% it was).
Increasing the negative voltage to -0.8V does this previous 56% d.c. return, and increasing further (-5V), does not change it.
It seems that the signal is more unstable with this negative voltage applied.
Manipulating the Original 5K pot-meter does not increase the d.c. range, it stays variable between 56% minimum till 100%.
Adding the 3.9K from pin 2 to ground and additional adding the 20K pot-meter between pins 2 and 3 (Black Birds solution) does not make any changes.
So we still have at best a d.c. variation of 56% till 100%.
Regards Itsu
Hi GL
Thanks for the offer, but pass it onto someone who cant build it, i wont have much time for a number of weeks yet.
Peter
Quote from: Itsu on 2014.03.26, 18:32:44
Ok, negative variable voltage on pin 15 via R7, but at the slightes negative value (milivolts) the d.c. flips to 100% (from the 56% it was).
That's good behavior, but you have misread (inverted) the duty cycle on the output transistor of the TL494 (pins 10 & 11)
You should have written:
"... at the slightest negative value (mllivolts) the DU.C. flips to 0% (from the 44% it was)."
When pin 13 is connected to the reference line (pin 14), the output transistor (pins 10 & 11) cannot be ON for longer than 48% of the whole cycle.
That's stated on the first page of
TL494's application note (http://www.onsemi.com/pub_link/Collateral/AN983-D.PDF).
Quote from: Itsu on 2014.03.26, 18:32:44
Increasing the negative voltage to -0.8V does this previous 56% d.c. return, and increasing further (-5V), does not change it.
The TL494 datasheet clearly states that pins 1 & 2 & 15 & 16 cannot be brought more than 300mV below ground.
I am
not surprised that the chip malfunctioned below -300mV
Quote from: Itsu on 2014.03.26, 18:32:44
It seems that the signal is more unstable with this negative voltage applied.
I am surprised that it still works after its specification has been exceeded by so much (-5V vs. -0.3V) - most likely the R7 limited the negative current and saved it.
Quote from: Itsu on 2014.03.26, 18:32:44
Manipulating the Original 5K pot-meter does not increase the d.c. range, it stays variable between 56% minimum till 100%.
The TL494 output DU.C. range is limited to from 0% to 48% when pin 13 is connected to the reference line (pin 14)
The function of the R11 pot is not to vary the output DU.C. directly or widen its range.
The 1
st error amplifier of the TL494 (pins 1 & 2) is configured to react only to quickly falling edges at R5 (via R7), see the diagram below.
The R11 pot determines how steep these falling edges need to be in order to activate this error amplifier.
Activation (high level on pin 3) of either one of these error amplifiers will immediately decrease the duty cycle of TL494's outputs.
Quote from: Itsu on 2014.03.26, 18:32:44
Adding the 3.9K from pin 2 to ground and additional adding the 20K pot-meter between pins 2 and 3 (Black Birds solution) does not make any changes.
That interferes with the original configuration of the 1
st error amplifier (pins 1 & 2).
Quote from: Itsu on 2014.03.26, 18:32:44
So we still have at best a d.c. variation of 56% till 100%.
0% to 44% at the TL494's output.
Quote from: Itsu on 2014.03.26, 18:23:13
Ok, i put a 3.9K resistor from pin 2 to ground so that i have 2.5V on pin 2.
Then i put again my 20K pot-meter center tap and a normal tap between pins 2 and 3, but same thing happens as without the 3.9K to ground.
I can vary the d.c. between 56% and 100%. by manipulating the 20K and the Original 5K pot-meters.
Never it goes below 56% d.c.
Regards Itsu
Isn't the waveform inverted? I mean, if you are measuring it at the collector, this would really be (100-56)= 44%. On the other extreme, 96% or so would correspond to 4%.
Cheers
Black Bird :)
OK, but bottomline is that putting a negative steady voltage on R15 via R7 does not change the d.c.
It stays fixed at 56% (after the use of the pull up resistor R16 to invert it).
Indeed the not used pin 8 output has (much smaller output) pulses varying between 0% and 43.9% when using Back Birds solution.
So only with enough steep pulses (determined by the R11 pot) this d.c. can be changed, right?
So how can i test this? Can i put some pulses on pin 15 (via R7) which mimic these "quickly falling edges at R5"?
And, can i by removing the short between pins 13 and 14 achieve more then 48% d.c.?
Added for Black Birds new post: it does not matter where i measure, directly at pin 11 or at the junction of R17 and R16, the signal has a d.c. of 56% varying to 100% when using your solution.
Measuring pin 8 which does NOT have a pull up resistor shows 44% going to 0% with your solution active.
Thanks, regards Itsu
Quote from: Itsu on 2014.03.26, 20:15:24
OK, but bottom line is that putting a negative steady voltage on pin 15 via R7 does not change the du.c. It stays fixed at 56%...
But you just wrote that the DU.C. changes to 100% (although you should have written 0% when writing about TL494's outputs) when pin 15 is between 0V and -0.3V.
Any input below -0.3V or above V
CC-2V is out of legal bounds.
Quote from: Itsu on 2014.03.26, 18:32:44
negative variable voltage on pin 15 via R7, but at the slightest negative value (millivolts) the du.c. flips to 100%...
So to me, the bottom line is that the output DU.C. of the TL494 changes to 0% after its pin 15 is brought below ground (within legal limits of the chip).
Quote from: Itsu on 2014.03.26, 20:15:24
Indeed the not used pin 8 output has (much smaller output) pulses varying between 0% and 43.9% when using Back Birds solution.
The TL494 is considered ON when its output transistor is conducting. There are two such output transistors internally: between pins 10 & 11 and between pins 8 & 9.
What's connected to these internal output transistors is secondary...and so is the voltage at those pins.
Quote from: verpies on 2014.03.26, 20:28:23
But you just wrote that the DU.C. changes to 100% (although you should have written 0% when writing about TL494's outputs) when pin 15 is between 0V and -0.3V.
Any input below -0.3V or above VCC-2V is out of legal bounds.
So to me, the bottom line is that the output DU.C. of the TL494 changes to 0% after its pin 15 is brought below ground (within legal limits of the chip).
The TL494 is considered ON when its output transistor is conducting. There are two such output transistors internally: between pins 10 & 11 and between pins 8 & 9.
What's connected to these internal output transistors is secondary...and so is the voltage at those pins.
Verpies,
So if I understand you correctly, the second error amplifier is not in use because in the Akyla0083 circuit the plus input pin 16 is grounded
and the minus input pin can never go below zero volt because of the positive bias from the VREF voltage output.
And, the first error amplifier is not in use because there can be no pulses through the 22nF capacitor because the input to the 210 Ohm
resistor is DC because of the 10uF capacitor. The DC voltage over the 10uF capacitor will vary with the load current.
Is this correct?
GL.
Quote from: verpies on 2014.03.26, 20:28:23
But you just wrote that the DU.C. changes to 100% (although you should have written 0% when writing about TL494's outputs) when pin 15 is between 0V and -0.3V.
Any input below -0.3V or above VCC-2V is out of legal bounds.
:D right, i mean it does not regulate, but yes it suddenly flips from 56% (44%) to 100% (0%) which is not what i want.
I want to have a variable d.c. which cannot been created by the negative steady dc on pin 15.
Regards Itsu
Quote from: Itsu on 2014.03.26, 20:15:24
So only with enough steep pulses (determined by the R11 pot) this du.c. can be changed, right?
Yes and a change is a change even if it the du.c of TL494 outputs changes to 0%.
Quote from: Itsu on 2014.03.26, 20:15:24
And, can i by removing the short between pins 13 and 14 achieve more then 48% d.c.?
Yes, and by connecting pin 13 to pin 7 (GND).
Quote from: Itsu on 2014.03.26, 20:49:41
:D right, i mean it does not regulate, but yes it suddenly flips from 56% (44%) to 100% (0%) which is not what i want. w
I want to have a variable d.c. which cannot been created by the negative steady dc on pin 15.
Oh, so you do not want to merely change the DU.C to whatever. You want to obtain
intermediate values of the DU.C. between zero and max.
With pin 13 connected to pin 14 you can only vary it between 0% and 48%.
To vary it that way you can remove R8 and C15, short pins 2 & 3.
After these alterations, the TL494 will respond to the R11 pot with intermediate DU.C. values.
P.S.
Shorting R9 and R10 will give you more adjustment range on the R11 pot but can damage the chip.
The right way to increase the adjustment range is to remove R9, short R10 and connect the R11 pot to pin 12.
Hmmm, why do i feel we are going round in circles?
This was what lost_bro already proposed a few pages back if i remember correctly.
But this is not what this replication is all about, at least not for me.
Offcourse can we make this thing behave like we want (variable frequency / variable d.c), but we have to find out why
this circuit is as it as, with all the components as they are.
Why does the d.c. not vary? Can it not vary in this configuration? Does it not need to vary?
I will continue to finish the rest of the print and put it all together without any further alterations first, perhaps it will
lead to a different behaviour alltogether.
In the mean time all you other replicators can catch up and confirm or reject my findings.
Thanks for all your inputs.
Regards Itsu
Quote from: Itsu on 2014.03.26, 20:15:24
So how can i test this? Can i put some pulses on pin 15 (via R7) which mimic these "quickly falling edges at R5"?
Yes. Using the unmodified circuit from
this schematic (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307), remove one side of R7 and connect the Rigol SG to it. Set it to "Ramp" 500Hz, HighL=+5.0V, LowL=+0.5V and the Symmetry to 90%.
Remove the power MOSFET or IGBT for these tests.
Verify with the scope that the LowL setting really gives you +500mV because there is a bug in some firmwares about this setting. We do not want to go below 0V because the 2nd error amplifier will get activated, as you have found out.
Vary the Symmetry setting to adjust the steepness of the falling edge applied to R7.
Monitor pin 3 with a scope. Depending on the R11 setting and the steepness of the falling edge applied to R7, you will observe positive pulses on pin 3.
There will be some settings that will result in rapid oscillations on pin 3.
Anytime pin 3 goes high, the TL494 output DU.C. will decrease to 0%.
Quote from: Itsu on 2014.03.26, 21:26:19
Hmmm, why do i feel we are going round in circles?
This was what lost_bro already proposed a few pages back if i remember correctly.
Because this is the way to make the DU.C. respond linearly to the R11 pot.
You are asking the question and getting two similar answers.
Quote from: Itsu on 2014.03.26, 21:26:19
Off course we can make this thing behave like we want (variable frequency / variable du.c),
But this is not what this replication is all about, at least not for me.
You are correct.
I thought that you had your private reasons for wanting to linearly vary the DU.C. with R11. ...e.g. testing the high power components at low DU.C.
Quote from: Itsu on 2014.03.26, 21:26:19
but we have to find out why this circuit is as it as, with all the components as they are.
With all the components as they are, it is obvious that the 2nd error amplifiers kills the output pulse as soon as the signal at R5 goes below ground. This is known as "pulse by pulse limiting".
The 1st error amplifier behavior is more complex. It responds to fast
falling edges at R5.
Quote from: Itsu on 2014.03.26, 21:26:19
Why does the d.c. not vary?
It does vary, but not gradually with a static feedback loop. With a dynamic feedback loop, the du.c. can vary gradually on a pulse by pulse basis.
Quote from: Itsu on 2014.03.26, 21:26:19
Can it not vary in this configuration?
It can vary gradually when a feedback loop is closed. For example when pulse by pulse limiting occurs individual pulses may be shortened to intermediate widths.
When you do the experiment with the Rigol and variable ramp pulses, you will see when the pin 3 goes high. If this pin goes high when the TL494's output is ON, the output pulse gets shortened.
If this shortening happens in synchronicity with TL494 internal oscillator, then intermediate output pulse widths can be generated.
Quote from: Itsu on 2014.03.26, 21:26:19
Does it not need to vary?
I don't know because I don't know how that transformer is supposed to work.
Sometimes it is important to realize that I don't know.
Quote from: Itsu on 2014.03.26, 21:26:19
I will continue to finish the rest of the print and put it all together without any further alterations first,
That's fine with me. I wrote about those alterations only because they were needed to achieve the behavior you were asking for. (linear adjustment of the du.c with the R11 pot).
So the next test seems to be applying those variable ramp pulses to R7 without any circuit alterations...and monitoring pin 3.
Quote from: verpies on 2014.03.26, 21:32:42
Yes. Using the unmodified circuit from this schematic (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307), remove one side of R7 and connect the Rigol SG to it. Set it to "Ramp" 500Hz, HighL=+5.0V, LowL=+0.5V and the Symmetry to 90%.
Remove the power MOSFET or IGBT for these tests.
Verify with the scope that the LowL setting really gives you +500mV because there is a bug in some firmwares about this setting. We do not want to go below 0V because the 2nd error amplifier will get activated, as you have found out.
Vary the Symmetry setting to adjust the steepness of the falling edge applied to R7.
Monitor pin 3 with a scope. Depending on the R11 setting and the steepness of the falling edge applied to R7, you will observe positive pulses on pin 3.
There will be some settings that will result in rapid oscillations on pin 3.
Anytime pin 3 goes high, the TL494 output DU.C. will decrease to 0%.
Right, works like a charm, see screenshot, but is it usefull somehow in the circuit we are replicating?
By the way, i made a short video earlier showing what i mean with the 56% etc. d.c. values
The upper 2 traces (yellow pin 11, purple junction R16/R17) are between 56% and 100%
The blue trace is on the unused pin 8 and shows a d.c. between 44% and 0%
https://www.youtube.com/watch?v=4zOBVxgp5JI&feature=youtu.be
Regards Itsu
@verpies,
Did you see my question one page back at Reply #345?
GL.
Quote from: Itsu on 2014.03.26, 21:56:41
Right, works like a charm, see screenshot, but is it usefull somehow in the circuit we are replicating?
Hmm, what is the yellow trace and what is the blue trace? What signal is at R7 ?
The behavior of the
1st error amplifier (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12403) is important because it tells us what is expected from the current flowing through the load.
Quote from: Groundloop on 2014.03.26, 20:49:17
So if I understand you correctly, the second error amplifier is not in use because in the Akyla0083 circuit the plus input pin 16 is grounded
and the minus input pin can never go below zero volt because of the positive bias from the VREF voltage output.
That is almost correct. Pin 15 must go below 0V for the 2
nd error amplifier to get activated. (but not too low, as less than -0.3V is illegal and causes the chip to malfunction).
I write "almost" because it is possible for the R5/R7 junction to become negative when:
1) R5 is a wirewound inductive resistor.
2) The load LED's conduct in reverse because of:
a) LED junction capacitance
b)
SRD (http://en.wikipedia.org/wiki/Step_recovery_diode) effect.
c) Reverse junction breakdown and avalanche.
So I cannot state with absolute certainty that the 2
nd error amplifier is unused. Maybe it limits the output pulse width in case of some catastrophic event in order to protects the device ...or maybe it hunts for some
SRD (http://en.wikipedia.org/wiki/Step_recovery_diode) effect in the LEDs ...or core saturation.
Quote from: Groundloop on 2014.03.26, 20:49:17
And, the first error amplifier is not in use because there can be no pulses through the 22nF capacitor because the input to the 210 Ohm
resistor is DC because of the 10uF capacitor.
Again, maybe.
Indeed the 1
st error amplifier needs steep falling edges to activate and the 10μF seems to prevent that.
But if you do the math, a 1μH inductive resistor (R5) will form a parallel LC tank with the 10μF capacitor that will oscillate at 50kHz - at this frequency the falling edges will be sufficiently steep to activate the 1
st error amplifier.
Two unused error amplifiers completely deprive the TL494 of a feedback loop and make it into a fixed 555 oscillator. It does not compute.
It is much more likely that the polarity of load LEDs is mistakenly reversed on that schematic.
Quote from: Groundloop on 2014.03.26, 20:49:17
The DC voltage over the 10uF capacitor will vary with the load current.
Yes, that's why that voltage seems ideal to monitor and limit the load current.... but it's backwards.
Quote from: verpies on 2014.03.26, 22:30:25
That is almost correct. Pin 15 must go below 0V for the 2nd error amplifier to get activated. (but not too low, as less than -0.3V is illegal and causes the chip to malfunction).
I write "almost" because it is possible for the R5/R7 junction to become negative when:
1) R5 is a wirewound inductive resistor.
2) The load LED's conduct in reverse because of:
a) LED junction capacitance
b) SRD (http://en.wikipedia.org/wiki/Step_recovery_diode) effect.
c) Reverse junction breakdown and avalanche.
So I cannot state with absolute certainty that the 2nd error amplifier is unused. Maybe it limits the output pulse width in case of some catastrophic event in order to protects the device ...or maybe it hunts for some SRD (http://en.wikipedia.org/wiki/Step_recovery_diode) effect in the LEDs.
Again, maybe.
Indeed the 1st error amplifier needs steep falling edges to activate and the 10μF seems to prevent that.
But if you do the math, a 1μH inductive resistor (R5) will form a parallel LC tank with the 10μF capacitor that will oscillate at 50kHz - at this frequency the falling edges will be sufficiently steep to activate the 1st error amplifier.
Two unused error amplifiers completely deprive the TL494 of a feedback loop and make it into a fixed 555 oscillator. It does not compute.
It is much more likely that the polarity of load LEDs is reversed on that schematic.
Yes, that's why that voltage seems ideal to monitor and limit the load current.... but it's backwards.
Verpies,
Thank you for the answer. I'm doing some simulations (using http://falstad.com/circuit/ ) and I get some effect
on the error amplifier. But, I had to use a low frequency and a higher value feed back capacitor because of
limitations in the simulator. Here is my sim file:
$ 1 5.0E-6 21.593987231061412 60 5.0 50
a 368 192 480 192 0 15.0 -15.0 1000000.0
c 480 128 368 128 0 2.2000000000000002E-8 -7.37297407604918
d 480 192 544 192 1 0.805904783
w 368 128 368 176 0
r 256 64 256 128 0 12000.0
174 256 128 256 288 0 4700.0 0.005 Resistance
r 256 288 256 352 0 2000.0
w 272 208 368 208 0
v 192 112 192 64 0 0 40.0 12.0 0.0 0.0 0.5
g 192 112 192 144 0
v 320 112 320 64 0 0 40.0 5.0 0.0 0.0 0.5
w 320 64 368 64 0
g 320 112 320 144 0
w 192 64 256 64 0
g 256 352 256 384 0
r 368 64 368 128 0 4000.0
c 336 240 336 288 0 4.7E-6 4.531756659791694
r 304 288 304 240 0 4000.0
w 336 288 304 288 0
w 304 240 304 176 0
w 304 176 288 176 0
w 288 176 288 64 0
w 288 64 320 64 0
w 336 240 336 176 0
w 336 176 368 176 0
r 336 288 336 336 0 210.0
c 384 336 384 368 0 1.0E-5 3.0
w 384 336 336 336 0
g 384 368 384 384 0
v 336 336 336 400 0 2 100.0 3.0 0.0 0.0 0.5
g 336 400 336 432 0
r 544 192 544 256 0 10000.0
g 544 256 544 288 0
x 124 76 164 82 0 24 +12
x 294 43 321 49 0 24 +5
r 448 336 448 400 0 1.0
g 448 400 448 432 0
w 448 336 384 336 0
r 544 192 544 128 0 27000.0
w 544 128 480 128 0
o 7 64 0 35 5.0 9.765625E-5 0 -1
o 31 64 0 35 20.0 0.0015625 1 -1
Added: And I probably did not get it right. :-)
GL.
Quote from: Groundloop on 2014.03.26, 23:15:30
Thank you for the answer. I'm doing some simulations (using http://falstad.com/circuit/ ) and I get some effect
on the error amplifier. But, I had to use a low frequency and a higher value feed back capacitor because of
limitations in the simulator. Here is my sim file:
I don't have a problem with a higher frequency and real component values. See
here (http://falstad.com/circuit/#%24+1+5.0E-6+2.275989509352673+66+5.0+43%0Ac+368+176+416+176+0+2.2000000000000002E-8+4.396612886951252%0Ar+416+176+464+176+0+27000.0%0Aw+464+176+480+176+0%0Aw+368+176+368+224+0%0Aw+368+224+336+224+0%0Ar+336+224+336+128+0+4000.0%0AR+336+128+336+112+0+0+40.0+5.0+0.0+0.0+0.5%0Ac+336+224+288+224+0+2.2000000000000002E-8+0.035583968692974643%0A174+336+272+336+336+0+47000.0+0.6782+Resistance%0Aw+368+304+352+304+0%0AR+336+272+336+256+0+0+40.0+12.0+0.0+0.0+0.5%0Aw+368+256+368+304+0%0Ar+288+224+224+224+0+210.0%0Ag+336+336+336+352+0%0Ar+288+224+288+128+0+4000.0%0AR+288+128+288+112+0+0+40.0+5.0+0.0+0.0+0.5%0A170+224+224+192+224+0+100.0+1400.0+5.0+0.02%0Aa+368+240+448+240+0+12.0+0.7000000000000001+1000000.0%0Ad+448+240+480+240+1+0.805904783%0Aw+480+176+480+240+0%0Ai+480+240+480+336+0+3.0E-4%0Ag+480+336+480+352+0%0Aw+480+240+528+240+0%0AO+528+240+544+240+0%0Ao+16+4+0+34+9.353610478917778+0.001461501637330903+0+-1%0Ao+22+4+0+290+18.707220957835556+9.765625E-55+0+-1%0A).
It is educational to delete the 27k resistor and observe what changes...
I did not simulate R5 and C5 because I consider those components part of the Signal Source.
Quote from: Itsu on 2014.03.26, 21:56:41
By the way, i made a short video earlier showing what i mean with the 56% etc. d.c. values
The upper 2 traces (yellow pin 11, purple junction R16/R17) are between 56% and 100%
The blue trace is on the unused pin 8 and shows a d.c. between 44% and 0%
https://www.youtube.com/watch?v=4zOBVxgp5JI&feature=youtu.be
That's a silly problem of terminology - not technology.
Formally, the duty cycle of the TL494 is specified as the percentage of the cycle when the output is ACTIVE, not when it is high (or low). "Active" merely means that the output transistor is conducting.
There are two ways TL494 output transistors can be connected. In the first method, the output is HIGH when active and in the second method, the output is LOW when active.
You have the output transistor configured generate a LOW state when active (as in Fig.2). See the diagrams below:
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12419) (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12421)
When measuring the duty cycle from outputs configured as LOW=ACTIVE, the scope channel should be
inverted.
Quote from: verpies on 2014.03.26, 22:05:40
Hmm, what is the yellow trace and what is the blue trace? What signal is at R7 ?
The behavior of the 1st error amplifier (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12403) is important because it tells us what is expected from the current flowing through the load.
Sorry i was not being clear, but the yellow trace is, as it always was in my videos on this circuit, the output of pin 11 measured at the junction of R16/R17.
The blue trace is the signal as measured at pin 3.
The signal on R7 is the signal as setup by my Rigol as described by you earlier.
The result (see above screenshot) is as you already predicted: "Anytime pin 3 goes high, the TL494 output DU.C. will decrease to 0%.". (in my terminology goes to 100% ;) )
Regards Itsu
Quote from: verpies on 2014.03.27, 07:31:08
That's a silly problem of terminology - not technology.
Formally, the duty cycle of the TL494 is specified as the percentage of the cycle when the output is ACTIVE, not when it is high (or low). "Active" merely means that the output transistor is conducting.
Yes, you are right, formally.
But i was describing to real visible output as shown by my scope to avoid confusion, however the opposite seems to be the case.
I did in an earlier video mention that the circuit was using the common Emitter configuration by means of the 740 Ohm pull up resistor R16.
QuoteWhen measuring the duty cycle from outputs configured as LOW=ACTIVE, the scope channel should be inverted.
Good idea, i will do that from now on.
Regards Itsu
Hi
I have no much time so shortly. I checked this circuit , no OU for me. :-[ Interesting is that when I disconnect the secondry the light is brighter, and including secondary has an opposite effect to what expected - light is dimmer and consumption on source is higher. Maybe I should reverse secondary connection.
Anyway I could post simulation on Multisim but only normal DC boost part , very interesting indeed, and I believe some expert can make it self-running in simulation too! The only request from me is if I post simulation, then every change should be posted here freely (like GPL ;D) I really really think it can be simulated.
Although you have to be very patient and have big free time, I'm occupied now with other works. Tell me what you think
Quote from: Itsu on 2014.03.27, 09:12:25
The blue trace is the signal as measured at pin 3.
The signal on R7 is the signal as setup by my Rigol as described by you earlier.
V: "Anytime pin 3 goes high, the TL494 output DU.C. will decrease to 0%.".
But what SigGen setting makes the pin 3 go high, and what setting keeps it at the low level?
If you make a video about behavior of pins 1,2,3 in the unaltered citcuit (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307), try to show other members what happens to pin 3 as you adjust the ramp "Symmetry" on the SigGen.
They should see that when the falling edge gets steep enough then pin 3 changes, while the steepness of the rising edge does not matter at all.
Formally, the answer should be expresses like this: "When the rate of change of voltage exceeds e.g. -1V/1µs at the input (R7), then the output (pin 3) gets activated."
This discovery is significant because it tells us what a boring circuit expects of the load and the transformer - the mysterious component.
This analysis already starts to bear fruit:
You have two signal feedback paths that "want" to function to shorten the widths of TL494's output pulses* and they react to:
1) the
rate of change of current through the load.
2) the
level of current through the load.
Ask yourself a question: What common phenomenon causes transformer currents to change too quickly and reach dangerous levels ?
* shortening output pulses on-the-fly could have been designed for protection of the device and/or for hunting of some effect ...or operating at a threshold of that effect.
Quote from: forest on 2014.03.27, 11:10:59
I have no much time so shortly. I checked this circuit , no OU for me. :-[
That circuit is not mysterious. The transformer is. ...its connections to the circuit, too.
Quote from: forest on 2014.03.27, 11:10:59
Interesting is that when I disconnect the secondary the light is brighter, and including secondary has an opposite effect to what expected - light is dimmer and consumption on source is higher.
You just described the behavior of a conventional transformer, where a load on the secondary winding increases the current in the primary winding.
Quote from: forest on 2014.03.27, 11:10:59
Anyway I could post simulation on Multisim but only normal DC boost part
What is a normal DC boost part? Doesn't the MultiSim simulate the TL492 chip?
If you replace transformer with a coil and remove second diode and all related parts it is just DC-DC boost converter,right? TL494 is perfectly simulated in Multisim. I have simulated the modified circuit with totem pole driver and it works, maybe the original too but I was impatient and didn't know about the soft-start implemented. That circuit is very interesting but you have to make coffe and wait a few minutes before simulation goes to starting point. Ok, back to Kicad works
Quote from: verpies on 2014.03.27, 12:15:55
But what SigGen setting makes the pin 3 go high, and what setting keeps it at the low level?
If you make a video about behavior of pins 1,2,3 in the unaltered citcuit (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307), try to show other members what happens to pin 3 as you adjust the ramp "Symmetry" on the SigGen.
They should see that when the falling edge gets steep enough then pin 3 changes, while the steepness of the rising edge does not matter at all.
Formally, the answer should be expresses like this: "When the rate of change of voltage exceeds e.g. -1V/1µs at the input (R7), then the output (pin 3) gets activated."
This discovery is significant because it tells us what a boring circuit expects of the load and the transformer - the mysterious component.
This analysis already starts to bear fruit:
You have two signal feedback paths that "want" to function to shorten the widths of TL494's output pulses* and they react to:
1) the rate of change of current through the load.
2) the level of current through the load.
Ask yourself a question: What common phenomenon causes transformer currents to change too quickly and reach dangerous levels ?
* shortening output pulses on-the-fly could have been designed for protection of the device and/or for hunting of some effect ...or operating at a threshold of that effect.
OK, i have set up the scope to monitor the pins 1, 2 and 3 signals together with the pin 11 (output) signal while inputting this ramp signal on R5/R7 as mentioned earlier.
At the end i substitute the pin 3 signal for the FG signal which seems to be a poor choice, so in a followup video (now uploading) i will substitute the pin 1 signal (a steady dc) for the FG signal so we can compare the FG ramp signal to this pin 3 signal (and the output signal)
Be aware that channel 4 (green) is inverted. I have corrected that in the follow up video.
Concerning your phenomenon, i guess you are referring to saturation which can cause currents to peak
Video here: https://www.youtube.com/watch?v=i_b18Q3Hueg&feature=youtu.be
WARNING! i seem to have screwed up the setting on the FG, so i have corrected that in the next video, sorry about that.
Regards Itsu
Quote from: forest on 2014.03.27, 14:03:49
If you replace transformer with a coil and remove second diode and all related parts it is just DC-DC boost converter, right?
Yes, it works as a boost converter when the unusual feedback loops are ignored and the L2 circuit is disconnected.
The load LEDs need more than 15V to light and that higher voltage is provided by L1 when VT1 stops conducting (opens)....or during the startup transient.
This simulation (http://falstad.com/circuit/#%24+1+4.9999999999999996E-6+1.2182493960703473+41+12.0+43%0A162+496+160+496+192+1+3.0+1.0+1.0+1.0%0A162+496+192+496+224+1+3.0+1.0+1.0+1.0%0A162+496+224+496+256+1+3.0+1.0+1.0+1.0%0A162+496+256+496+288+1+3.0+1.0+1.0+1.0%0A162+496+288+496+320+1+3.0+1.0+1.0+1.0%0Av+288+224+288+160+0+0+40.0+10.0+0.0+0.0+0.5%0AT+528+80+576+80+0+0.01+3.0+0.21785124776704234+-6.830473686658678E-18+0.999%0Ad+528+112+528+160+1+0.805904783%0Aw+496+160+528+160+0%0A159+400+224+400+256+0+20.0+1.0E10%0Aw+288+32+288+80+0%0Aw+288+80+288+160+0%0AR+352+240+320+240+0+2+500.0+5.0+0.0+0.0+0.3%0Ar+496+320+496+368+0+1.0%0Aw+288+224+288+368+0%0Ac+528+160+528+368+0+0.002+13.784003347977666%0Aw+496+368+528+368+0%0Ac+448+320+448+368+0+1.0E-5+0.0639247473696134%0Aw+496+368+448+368+0%0Aw+448+368+400+368+0%0Aw+288+368+400+368+0%0Aw+448+320+496+320+0%0Aw+528+112+400+112+0%0Aw+400+112+400+224+0%0Aw+400+256+400+368+0%0Ad+560+160+624+160+1+0.805904783%0Ar+624+192+560+192+0+420000.0%0Ac+624+224+560+224+0+1.0E-10+3.286134775073611E-13%0Aw+560+160+560+192+0%0Aw+560+192+560+224+0%0Aw+624+160+624+192+0%0Aw+624+192+624+224+0%0Aw+528+160+560+160+0%0Aw+608+112+624+112+0%0Aw+608+80+624+80+0%0Aw+624+112+624+160+0%0Aw+576+80+608+80+0%0Aw+576+112+608+112+0%0Ar+288+32+400+32+0+1.0%0Ar+288+80+400+80+0+0.22%0Aw+400+80+528+80+0%0As+624+32+624+80+0+1+false%0Aw+400+32+624+32+0%0As+352+240+384+240+0+0+false%0Ao+15+16+0+34+20.0+0.2+0+-1%0A) illustrates this.
Quote from: Itsu on 2014.03.27, 15:17:13
Concerning your phenomenon, i guess you are referring to saturation which can cause currents to peak
Yes, this phenomenon is shown in the video quoted in
this message (http://www.overunity.com/12736/kapanadze-cousin-dally-free-energy/msg341029/#msg341029).
If I wanted to make a driver that would automatically operate a transformer at or near the saturation point, I would construct a feedback loop that would react to quickly changing current and a critical level of current.
This is what Acula's feedback loops look for ...BUT their polarity is backwards and not sensing the currents in the transformer windings but in the load bypassed with 10µF and 2000µF caps which effectively attenuate any high frequency signals there. So that makes no conventional sense.
Quote from: Itsu on 2014.03.27, 15:17:13
... while inputting this ramp signal on R5/R7 as mentioned earlier.
It is better to disconnect on side of R7 from R5 and apply the ramp signal to the disconnected side of R7.
In the circuit, the R5 is part of the signal source that you are trying to simulate with the SigGen, so there is no need to have that duplication and a heavy 1Ω load for the weak SigGen's output.
In the video the ramp signal from the SigGen is getting negative (below 0V) and that triggers the
2nd error amplifier (pins 15 & 16) which is set up to detect any signal below zero.
The
2nd error amplifier also affects pin 3 and is interfering with the picture of the
1st amplifier's operation.
Quote from: Itsu on 2014.03.27, 15:17:13
WARNING! i seem to have screwed up the setting on the FG, so i have corrected that in the next video, sorry about that.
Do you mean this setting?
Quote from: verpies on 2014.03.27, 15:52:09
It is better to disconnect on side of R7 from R5 and apply the ramp signal to the disconnected side of R7.
In the circuit, the R5 is part of the signal source that you are trying to simulate with the SigGen, so there is no need to have that duplication and a 1Ω load for the weak SigGen's output.
Do you mean this setting?
Yes.
R7 and R5 are in series only in my setup, R5 point at ground side is disconnected there and this is where the FG signal is put in.
OK, second attempt, pins 2, 3, 11 and the FG output monitored.
(Pin 1 has a steady DC which does not change during the video, so i changed it to show the FG output.)
FG set up properly now.
Video shows that the steepness of the ramp signal translate to a signal on pin 3 (feedback) which kind of mutes (set to 0% d.c.) the output signal on pin 11
Video here: https://www.youtube.com/watch?v=exIFDDgSrng&feature=youtu.be
Regards Itsu
Quote from: Itsu on 2014.03.27, 15:58:38
R7 and R5 are in series only in my setup, R5 point at ground side is disconnected there and this is where the FG signal is put in.
What about C5 and C3? They seem to be disconnected as well because they don't affect the SigGen's output but I have to ask.
Quote from: Itsu on 2014.03.27, 15:58:38
OK, second attempt, pins 2, 3, 11 and the FG output monitored.
FG set up properly now.
Video shows that the steepness of the ramp signal translate to a signal on pin 3 (feedback) which kind of mutes (set to 0% d.c.) the output signal on pin 11
This video illustrates the function of the
1st error amplifier well.
So how steep does the falling edge need to be in order to translate to a signal on pin 3 ? ...in negative Volts/µs ;)
Remember, that this steepness sensitivity can be adjusted by the R11 pot. (an ordinary voltmeter is good for monitoring the wiper of R11)
C5 and C3 are not installed in my setup.
basically all right from C13 and R16 is not installed with the exception of R5 (in series with R7) which is installed (but ground point lifted for FG input) and R19 which is NOT installed.
Looking at my scope,
at 85% symmetry there is NO signal on pin 3, and the steepness of the ramp then is 5.4V over 300us = 1V/55us
at 90% symmetry there is a signal on pin 3, and the steepness of the ramp then is 5.4V over 200us = 1V/37us
Varying R11 moves this No signal point till minimum 57% symmetry with a steepness of the ramp of 820us at 5.4V = 1v/151us
If thats what you mean :)
Regards itsu
Quote from: Itsu on 2014.03.27, 17:49:59
If thats what you mean :)
Yes :)
Since it is customary to keep the denominator at unity, I'd rearrange the units to state that the input slew rate needs to be faster than -6.6V/ms to -18V/ms, depending on the setting of R11, in order to activate the 1
st error amplifier and limit the output pulse width of the TL494.
What good is that information?
Well for starters it lets us know that if the peak current though those load LED's is e.g. 3A, then the switching frequency of the VT1 transistor has to be
Quote from: verpies on 2014.03.27, 18:38:44
Well for starters it lets us know that if the peak current though those load LED's is e.g. 3A, then the switching frequency of the VT1 transistor has to be
Is that a question? (did you forgot the ?)
Or do you want to build up the tension ;D
I have 3W leds on order, they pull 700mA at 3.5V. With 10 Leds (5 in series, 2 rows of 5 parallel), it should pull about 1.4A at 15V
Regards Itsu
Dear All.
I have been a bit busy this last couple of days and have now managed to catch up with what has been going on here.
I felt that I needed to extend a note of gratitude and thanks to both Itsu and Verpies for what appears to me, to be an amazing amount of effort and time spent unravelling the control circuit.
It has been wonderful to see members coming together, providing their particular expertise toward a common goal. Heart warming indeed !!
Thank you all. O0
Cheers Grum.
Salve a Tutti,
Sorry for interfering.
Quote from: verpies on 2014.03.27, 12:49:16
................
That circuit is not mysterious. The transformer is. ...its connections to the circuit, too.
................
Precisely, about transformers. To me, the following is new:
I did not know that you could "feed" the primary of a transformer with some sinus or square
voltage
via a single wire and get any 'volts out' in a (
not loaded, indeed) secondary.
See, for example, case #4 (picture number 3) with square waves.
The basic (but nicely easily operated, no?)
Falstad simulator gave me a big nothing. :-X
Please, do not believe that these are just useless volts junk.
Actually, I just figured out -with another experiment- that I was able to light
a 220 Vols/1 watts Led bulb out of a single wired (not grounded) single (cored) coil.
Yes, this bulb is not fully lighted.
My question could be:
Does a single wire (= not a "'real" 'circuit'"???) consume amps? According to a cheap meter, it does.
Ex: without the Leds: (about) 8 ma. With the Leds: 4 ma. This behavior is not new for me.
Actually, with a
very very simple CCT, I'm able (as anybody(?)) to light this kinda (1W to 4W) leds bulbs
with 1.5 volts and 10-80 ma.
Now? Are 20 volts AC-sinus Pk-to PK and 4 ma too much to get about the same result than
with 1.5 Vols-DC/10-20ma???
I do not thrust this cheap meter. Is it able -In the AC-Amp range - to measure things
at about 20 Khz? I have to test this with my (40 Y.old) scope and a one ohm resistor...
Anyway, here are my 3 pictures:
(http://www.freenrg.info/4OUR_F/isol_trafo_1.jpg)
(http://www.freenrg.info/4OUR_F/isol_trafo_2.jpg)
(http://www.freenrg.info/4OUR_F/Nexp-with-Isol-Trafo.jpg)
Avec mes salutations les plus cordiales,
Jean
==============================
"Spell Check" out of order ???
Quote from: verpies on 2014.03.27, 18:38:44
Well for starters it lets us know that if the peak current though those load LED's is e.g. 3A, then the switching frequency of the VT1 transistor has to be
This is the conclusion I was hoping you would find. Does in fact the control circuitry compensate faster than the main signal? Because if it does, that changes the whole ball game--one cannot simply inject a signal into the MOSFET driver and expect the proper response.
Quote from: verpies on 2014.03.26, 13:04:50
Exception could be made for the brief reverse conduction during the DSR effect (as in your nanopulser) as well as the Zener effect and Avalanche reverse breakdown of some diodes in some cases.
Yes and in there lies the Out of the Box jewel as far as I am concerned. Said this pages ago.
@Grumage (@All)
While you guys work out more of that circuit, I took some time and measured some diameter ratios of the E-cores since they are farthest away from the actual bobbin and windings in the Akula photo. With these ratios, I extrapolated their relative diameters to overlay the bobbin coil that are closer in the photo. As you can see, on the second image, the center core of the Ecore takes up a good amount of diameter under the winds. Add the thickness of the plastic bobbin and you will realize that Akula never had enough room in there for 15/45, far from it. Especially since he used thicker wire on the primary side. That also explains why the wire going to the core side first had a very weak angle of entry. Yes, the second image is not a perfect portrayal since the coil bobbin has a depth of vision but keeping the ratio the same gives you a good enough confirmation of how much space the Center Ecore is taking in that bobbin. I could be wrong, but I doubt it even if slightly off it stilldoe snot figure all those winds.
You usually make the coil first with the FG resonance method and identify your targeted drive frequency, then the circuit to drive it. Think of Akulas' steps. He made a coil and started to drive it. He probably blew a mosfets after each time he hit the sweet spot before he got sick and tired and decided to start working to tame the flyback beast, hence he comes up with the following circuit which is great because it incorporates a standard self-running circuit as well. A very impressive mind indeed. We'll see if the IC side survives.
I had prepared a long explanation on why your wind method will be too difficult to run but decided not over post. Maybe when it's more the appropriate time.
So for me, the standard wind method with a copper strip under that last layer is what I see. Keep the strip a good 1/8th open. And this falls perfectly with what is drawn in his diagram. It's just that as in the past, Akula does not give precise wind numbers and from my previous post, you can realize that those numbers are pointless if you do not use the exact same wires and core you will have to tune your coil anyways. Maybe this is Akulas way of saying "You want it, then break your ass and figure it out".
In my book, this should be the first method for trials in order to both not risk blowing the circuit or the core and at least having a base measure for performance. Let's say the standard method lights your bulb blinding white and the circuit directs the self-running automatically, then we can all yell out Yes Yes Praise Be. Now you try your method and see it lights the bulb 1/2 bright or 1/4 bright and 2 times more amperage is consumed. At least you know that the core can dish out the juice in standard mode and in your mode you can then try other configurations to try and go above the standard method output. When in doubt start from where you know.
Think of this simple fact. The less secondary winds, the higher the amperage. If this ever is not OU, there are ways to add another secondary on that coil to draw some more so it still remains a good base of research and I commend anyone taking it.
My full respect and God Speed.
wattsup
PS: Hmmmmmm, just a thought, probably crazy but I wonder if anyone has ever tried to use four LEDS to make a bridge rectifier, what will happen? If LEDS are diodes, they should light up as energy is transferred between the LEDs unidirectional rectifier functions so all the output can be used. If you run it at AC 10Hz with a good enough impulse you should see the interactions of the lights, yes or no? Purely theoretical.
Here is Multisim 12 simulation, please help me complete it. Something interesting is happening after 0.038 s, you have to be patient or have faster computer then mine.
@Grumage
This is the other part of my post above that I decided to put down today since I will be less on the forum during the weekend but wanted you to get some other insights as an OPINION only. Take what yo think is OK.
OK, about your winding method, I know it looks really exotic with a secondary wound 2 layers over a gap and your primary wound at one side of the gap over the secondary were the first layer turns to the second layer. So the Primary is over the secondary that has 15 turns under the primary. The Sec also has let's say 7 winds on the other gap side first layer before it goes under the primary so that makes a total of 22 turns out of 45 that are before or under the primary.
Now what happens when the primary pulse in on. This impress pushes on both the sec under the primary as well as the core. What happens in the sec. Nothing because since it is going in both directions under the primary, they cancel out as the pulse from the primary increases in its cycle. Now the pulse on the primary is off, by that time you get a reflection hitting the other side of the gap there you have sec winds 1-7 on a first layer and winds 23 to 45 above it. So what happens. Nothing because the sec winds under the primary are producing a node cancellation (bottleneck) while they try to realign. But it is R&D 100% so technically it deserves trails but please read on.
If I took an educated guess and the object was to keep the secondary as per your vision over both gaps, this is what I would do.
Wind the secondary only one layer starting at left and passing over the gap till the end of the first layer to the right side of the gap as a first layer, then wind the rest of the sec over the right hand side of the gap. Then wind 2 layers of the primary on the left side.
You see the object of any energy system serving a load is to provide the greatest potential differential possible to the ends of the loaded coil. The way above has a better chance because only one layer starts under the primary or high potential impress and the end furthest away on the secondary right side will be your low potential impress.
So the primary pulses and the sec can now push in one direction without cancellation from a second layer caught under the primary. Now the reflection hits the other side of the gap to turns 8-16 of the secondary and can also push in the same direction as the sec winds 1-7 thus keeping the highest impress on that side of the sec coil. You will always have to experiment by trying the pulse on one side and compare when running on the other side.
To ensure that the outer most wind layers are at their lowest impress possible, you also have to consider that in a coil wire you have energy in the wire but also energy as skin to skin effect between the wires and this always reduces the total differential from one end of the coil to the other. So what Akula did is add a copper strip before both coils had their last layer wound. (See photos in previous post). Now it could be copper or even maybe home aluminum siding obviously either painted white or originally white but I would bank on a grounded copper strip. Also the strip does not cover the coil turns completely and I would say you have a good 1/8th of the strip circumference open. This strip will dampen any skin to skin transfer to the last layer to ensure that the differential of both sec ends is the maximum possible. Plus the energy dampened by the strip will provide a greater negative peak to the ground side thus providing an even greater differential. You work the top side, you work the bottom side and then hope for the best.
But in both scenarios, either your way (double sec layer under the primary) or the way I just defined above (single sec layer under the primary), the frequency that will make it all work will be in a very thin spread and the slightest frequency fluctuation during operation will most likely kill the effect so it is not easy to push and maintain this effect because you are creating various conditions of primary pulse, secondary gets that pulse, core gets that pulse, core gaps provide reflection, primary provides flyback, etc. All these movements are very difficult to predetermine in order to design an optimal winding method using such a gap overlap scheme. I don't think the Akula circuit itself is designed to take hold of such events since the circuit shows more of a standard drive that has some looping lines to capture flyback. Plus you will need an extremely fast rise time on the pulse to not over-saturate the core with so much movement and the frequency required for such a method risks to be much higher then the circuit design.
When you test something and see it does not work, it only means at the available frequency range (and also often the diode used) it does not work. It's like diving 20 feet in the middle of the ocean, seeing no fish and thinking this ocean is fishless. So really, it does not mean it will not work at a higher frequency because in reality there is a frequency for everything in existence. You can take a toroid with 10 different coil winds and eventually you will find that one frequency that will bring them all to life. hehehe
The above is my vulgarized technical vision of the problem.
wattsup
Dear Wattsup.
Thank you for your theory regarding the transformer.
I wonder whether we can call it such ?? I have attached a drawing of the schematic with a RED line drawn through it. This RED line is depicting a constant DC current flowing through the primary winding ( L 1 ) to the LED load. This current will provide a steady state of magnetic flux within the core. We now also have a MOSFET that effectively grounds L1 at a frequency yet to be determined, at an even lower resistance than the load!!
Are we in fact looking at a "Nifty" way of creating an NMR response from the core material ?? My cores are Zink / Manganese, highly suited for this purpose !! Or, if I understood Verpies correctly, a Ferro resonance response ??
I agree with your sentiments from your earlier post, we will have to work hard to find out what Akula was actually doing.
Back to the bench !!
Cheers Grum.
@Grumage
This is exactly what I was reffering to in a previous post explaining that this one fact makes this circuit Out of the Box. For me I would call this a Damped Inductive Charge and Dischage System that is typically based on the Tesla Ozone Patent (TOP).
L1 goes to load but you have C5 there. How does C5 charge? It charges to the inductive capacity of the L1 when mosfet is open charge is held by the diode between the mosfet and C5 and discharges into the LED load when mosfet is closed because voltage across L1 drops to zero.
I have tried working TOP pulsing for years, on and off always ending up with blown mosfets then reverting to slow mechanical or relay methods and sometimes guys would get together to try and work out an EE solution. So Akula, with this circuit found a way to do it and make it last. Right there, if that mosfet lasts more then a few minutes, believe me you don't need any fancy fancy coil.
The output will be somewhat reactive and some amperage via the L2 so I doubt it can handle a large inductive load like a motor, but lighting LEDS, no problem.
wattsup
Hi all,
Great work on troubleshooting the control circuit! I only get a few days on the PC each week.. the rest of the time I can catch a little on my phone, but not easy. A lot happening!
@watttsup
Nice work on the transformer!
Quote from: Grumage on 2014.03.28, 13:59:47
Dear Wattsup.
Thank you for your theory regarding the transformer.
I wonder whether we can call it such ?? I have attached a drawing of the schematic with a RED line drawn through it. This RED line is depicting a constant DC current flowing through the primary winding ( L 1 ) to the LED load. This current will provide a steady state of magnetic flux within the core. We now also have a MOSFET that effectively grounds L1 at a frequency yet to be determined, at an even lower resistance than the load!!
Are we in fact looking at a "Nifty" way of creating an NMR response from the core material ?? My cores are Zink / Manganese, highly suited for this purpose !! Or, if I understood Verpies correctly, a Ferro resonance response ??
I agree with your sentiments from your earlier post, we will have to work hard to find out what Akula was actually doing.
Back to the bench !!
Cheers Grum.
I was taught electron flow, but I get conventional flow. ;)
Quote from: wattsup on 2014.03.28, 03:33:46
PS: Hmmmmmm, just a thought, probably crazy but I wonder if anyone has ever tried to use four LEDS to make a bridge rectifier, what will happen?
I've never tried.. would be interesting! Reminds me of AV plugs.
I know main focus here is 30w version, but I'm going to build this micro version this weekend since I already have the components. Although I don't know correct values.
http://www.overunityresearch.com/index.php?topic=2358.msg36966#msg36966 (http://www.overunityresearch.com/index.php?topic=2358.msg36966#msg36966)
Edit: Sorry.. made some errors on schematic. I think this is correct now.
@All,
There is 0 free board (the breadboard type) left.
See here: http://www.overunityresearch.com/index.php?topic=2358.300 Reply #318 on: 2014-03-26
Files for this board is found here: http://www.overunityresearch.com/index.php?topic=2358.250 Reply #270 on: 2014-03-24
Peter probably have more of the Akyla0083 replica boards left, you have to ask him.
GL.
@GL
I am tempted but know that I will only make a mess of it. My talents in small circuit works is not good enough. Are there companies out there that can make the complete circuit for a reasonable price? Then I would be interested in a few. I am sure anyone who made these in quantity will sell them all and more. I'd take four of them to stack the output. (hint-hint)
@all (I posted this at OU so I am putting it here as well - first unrelated paragraphed removed)
About the circuit, the stop switch is to kill energy return from the secondary to the source. This is normal if you want an easy manual method of stopping the return in case R3 or C11 started heating up. He probably burnt a few of them while fine tuning his system, before putting the stop switch in there.
Always remember this is a self-running circuit. All the features of pulsing, secondary and return are in the circuit. During self-running you only have C11 and C5 to hold enough energy to run the circuit for a second at most.
Like I said at OUR, the L1/VT1/C3 combo is an analog or I would say a very smart variation of the Tesla Ozone Patent (TOP). In that patent Tesla explains that you can produce a damped impulse through a primary where the energy used can return to source. But here Akula is using the L1 for two usages, the first being an inductive source as in the motor coil of the TOP and second as the working transformer to impart to the secondary L2.
So the difference here compared to a TOP is that, instead of using the L1 pulse to then drive a primary to secondary coil to produce a discharge arc thus producing ozone, Akula is using it to directly drive the load (working primary without a secondary) where at the end of the load he put R5/C5 to reduce the bleed or load consumption to keep the system primed at all times and where he checks the system status. That is because the system cannot produce enough to run the load with straight ground hence full draw on the source will kill the looping effect.
The secondary L2 can then be freed 100% to simply concentrate its effort to produce enough output for the positive return line to C11.
So the L1 has two usages. It feeds the load directly via a TOP discharge and it energizes the secondary to recycle the energy. That is so smart that most of our brains will need an upgrade to understand it. hahaha
By contrast, our standard method of undamped pulse the primary and draw off the secondary where the energy used can never return to the source. Our whole EE system is based on undamped pulse.
The above also confirms to me that you don't need any special transformer winding other then what we use normally. That is, when primary is over the gap, the secondary is over another gap, or, when primary is on one side of gap, the secondary is on the other side of the same gap. Anything more and you are just creating unusable havoc and energy cancellation.
To test this is simple. Take any coil primary and tank it. Then take any battery and short it across the primary very quickly. Now look at the voltage on your tank. That's what is running the load. Then the secondary is free to return to source. Guys should be clamoring over their work benches by now.
wattsup
Dear All.
It has been pointed out to me privately that my "RED LINE" analogy, whilst Ok for a purely resistive load, is incorrect for an LED load. The fact is we have LED's in a series/parallel configuration. If we apply our 12 Vdc without the Mosfet being triggered then zero current will flow through the diodes simply because the forward voltage conduction threshold has not been reached!! Only when we start the flyback cycle will the voltage reach a high enough value to light the load.
Now the thread has reached a more pedestrian pace I would like to present a simulation for perusal by our more learned members of OUR, I would be grateful for your feedback.
http://falstad.com/circuit/#%24+1+4.9999999999999996E-6+1.2182493960703473+41+12.0+43%0A162+352+176+352+208+1+3.0+0.0+0.0+1.0%0A162+352+208+352+240+1+3.0+0.0+0.0+1.0%0A162+352+240+352+272+1+3.0+0.0+0.0+1.0%0A162+352+272+352+304+1+3.0+0.0+0.0+1.0%0A162+352+304+352+336+1+3.0+0.0+0.0+1.0%0Av+144+240+144+176+0+0+40.0+12.0+0.0+0.0+0.5%0AT+384+96+432+96+0+0.01+3.0+0.14293321860627017+3.0691946109173E-13+0.999%0Ad+384+128+384+176+1+0.805904783%0Aw+352+176+384+176+0%0A159+256+240+256+272+0+20.0+1.0E10%0Aw+144+48+144+96+0%0Aw+144+96+144+176+0%0AR+208+256+176+256+0+2+500.0+5.0+0.0+0.0+0.3%0Ar+352+336+352+384+0+1.0%0Aw+144+240+144+384+0%0Ac+384+176+384+384+0+0.002+14.339777088833753%0Aw+352+384+384+384+0%0Ac+304+336+304+384+0+1.0E-5+0.16850144714049262%0Aw+352+384+304+384+0%0Aw+304+384+256+384+0%0Aw+144+384+256+384+0%0Aw+304+336+352+336+0%0Aw+384+128+256+128+0%0Aw+256+128+256+240+0%0Aw+256+272+256+384+0%0Ad+416+176+480+176+1+0.805904783%0Ar+480+208+416+208+0+420000.0%0Ac+480+240+416+240+0+1.0E-10+2.518181219102189E-10%0Aw+416+176+416+208+0%0Aw+416+208+416+240+0%0Aw+480+176+480+208+0%0Aw+480+208+480+240+0%0Aw+384+176+416+176+0%0Aw+464+128+480+128+0%0Aw+464+96+480+96+0%0Aw+480+128+480+176+0%0Ar+144+48+256+48+0+1.0%0Ar+144+96+256+96+0+0.22%0Aw+256+96+384+96+0%0As+480+48+480+96+0+1+false%0Aw+256+48+480+48+0%0As+208+256+240+256+0+0+false%0Ag+144+384+144+400+0%0Aw+432+96+464+128+0%0Aw+464+96+432+128+0%0Ax+372+118+390+122+0+16+L1%0Ax+426+117+440+120+0+12+L2%0Ax+266+367+287+371+0+16+C5%0Ax+274+260+303+264+0+16+VT1%0Ax+391+303+412+307+0+16+C3%0Ax+113+199+142+203+0+16+12V%0Ao+15+16+0+34+20.0+3.2+0+-1%0A
Cheers Grum.
Quote from: wattsup on 2014.03.29, 12:45:12
@GL
I am tempted but know that I will only make a mess of it. My talents in small circuit works is not good enough. Are there companies out there that can make the complete circuit for a reasonable price? Then I would be interested in a few. I am sure anyone who made these in quantity will sell them all and more. I'd take four of them to stack the output. (hint-hint)
wattsup
Hi Wattsup,
Yes there exists firms that can produce soldered boards but most of them is for large quantity production.
But I think we should keep everything on a small scale basic right now at this point in time. We need to
know if it this circuit really do work. If it does, then perhaps we can think about large scale production.
That said, both the boards I have made is easy to solder with a standard 40 Watt soldering iron. The
board has a solder masks at each solder point that prevent any solder to "float" over to a nearby solder
pad. So it should be easy to solder using good light and a steady hand. If you are old, like me, then
there is also magnifying glasses sold that you can wear on your head, thus keeping both hand free
to do the job. See here: http://www.ebay.com/sch/sis.html?_nkw=Light%20Head%20Magnifying%20Glass%20Magnifier%20jewelry%20repair%20Hobby%201%205%203%208%205%2010%20x%20Lenses&_itemId=390560658089
GL.
@Groundloop
Thanks for your post.
Another @member just offered to have one populated by one of his friends and I agreed if he can do one for me. I won't say his name because it was offered via a PM unless he says its OK to mention his name here.
I'm 56 (not that old - hehehe), still in good shape. The thing is I would prefer to have it done by a seasoned professional that knows not to put too much heat, placement, etc. The time I will spend on that would be better spent on present works and I am not worried to pay for the time and components.
So for sure I would like a circuit to do these tests wit the others when it is available.
wattsup
Quote from: wattsup on 2014.03.29, 16:22:15
@Groundloop
Thanks for your post.
Another @member just offered to have one populated by one of his friends and I agreed if he can do one for me. I won't say his name because it was offered via a PM unless he says its OK to mention his name here.
I'm 56 (not that old - hehehe), still in good shape. The thing is I would prefer to have it done by a seasoned professional that knows not to put too much heat, placement, etc. The time I will spend on that would be better spent on present works and I am not worried to pay for the time and components.
So for sure I would like a circuit to do these tests wit the others when it is available.
wattsup
Wattsup,
Sure I can send you the "bread board type" board and you can ask Peter to send you the Akyla0083 replica board.
Then you can choose if you want to build one, or the other, or both of the boards. Just PM me a snail mail address.
GL.
Good day Grum:
OK, checked out your new sim:
I changed the upper top left hand resistor from 1 ohm (original value) to a 50 ohm and now the LEDs will stay lit when the upper right hand switch is closed.
With the smaller value resistors (<50ohms), the LED are effectively bypassed from rest of circuit when switch is closed.
Voltage @C3 will maintain 12.79v w/ 50ohm resistor. Sim shows <1/2 watt power dissipation (50ohm resistor); maybe can use a potentiometer in place of this resistor.
Varying the resistance at this resistor changes the voltage across the LEDs.
Let me know what you think.
take care, peace
lost_bro
I finished placing the components on the print i made.
Only thing left to do is to hook up the leds (10 x 3W Leds), but they have not arrived.
First thing to notice is the difference in the number of components compared to the picture of the original circuit.
Anyway, i shortly fired up the thing without the load (leds) and made some scope shots of the gate voltage (yellow), drain voltage (blue) and the drain current (green, has to be taken * 2!!)
They probably are completely useless this way, but its all i can show right now.
I will be away from the bench the next 2 weeks due to work and have limited access to the forums/mail.
Hope you guys have catched up by then and can show a working device.
Video here: https://www.youtube.com/watch?v=05jIODPdp2U&feature=youtu.be
Regards itsu.
Quote from: Itsu on 2014.03.29, 21:05:14
I finished placing the components on the print i made.
Only thing left to do is to hook up the leds (10 x 3W Leds), but they have not arrived.
First thing to notice is the difference in the number of components compared to the picture of the original circuit.
Anyway, i shortly fired up the thing without the load (leds) and made some scope shots of the gate voltage (yellow), drain voltage (blue) and the drain current (green, has to be taken * 2!!)
They probably are completely useless this way, but its all i can show right now.
I will be away from the bench the next 2 weeks due to work and have limited access to the forums/mail.
Hope you guys have catched up by then and can show a working device.
Video here: https://www.youtube.com/watch?v=05jIODPdp2U&feature=youtu.be
Regards itsu.
Hi Itsu,
Great work! My simulations show drain currents in the range of what you found, around 20 amps, with a 10 ohm load plus some low current LEDS (3 sets of 5 LEDs in series, paralleld to the resistor)
Cheers,
Black Bird
Quote from: Black_Bird on 2014.03.29, 22:28:34
around 20 amps
Wow. We'll be seeing some smoke! :)
Quote from: 4Tesla on 2014.03.30, 00:27:52
Wow. We'll be seeing some smoke! :)
A fuse is likely to be the most important component in this build!! I have started to build this circuit, so by the time Itsu returns to his bench to complete his build, I may have something to report.
Hoppy
Hello All:
I posted a comment the other day regarding the 'New' sim by Grum.
No-one commented, so I am taking the liberty to post the Modified 'New' sim.
$ 1 4.9999999999999996E-6 2.4737822143832555 46 12.0 43
162 384 176 384 208 1 3.0 0.0 0.0 1.0
162 384 208 384 240 1 3.0 0.0 0.0 1.0
162 384 240 384 272 1 3.0 0.0 0.0 1.0
162 384 272 384 304 1 3.0 0.0 0.0 1.0
162 384 304 384 336 1 3.0 0.0 0.0 1.0
v 176 240 176 176 0 0 40.0 12.0 0.0 0.0 0.5
T 416 96 464 96 0 0.01 3.0 0.013916074305551474 0.08171659486717558 0.999
d 416 128 416 176 1 0.805904783
w 384 176 416 176 0
159 288 240 288 272 0 20.0 1.0E10
w 176 48 176 96 0
w 176 96 176 176 0
R 240 256 208 256 0 2 500.0 5.0 0.0 0.0 0.3
r 384 336 384 384 0 1.0
w 176 240 176 384 0
c 416 176 416 384 0 0.002 13.028650819569869
w 384 384 416 384 0
c 336 336 336 384 0 1.0E-5 0.014039012853552583
w 384 384 336 384 0
w 336 384 288 384 0
w 176 384 288 384 0
w 336 336 384 336 0
w 416 128 288 128 0
w 288 128 288 240 0
w 288 272 288 384 0
d 448 176 512 176 1 0.805904783
r 512 208 448 208 0 420000.0
c 512 240 448 240 0 1.0E-10 -0.7431471995083534
w 448 176 448 208 0
w 448 208 448 240 0
w 512 176 512 208 0
w 512 208 512 240 0
w 416 176 448 176 0
w 496 128 512 128 0
w 496 96 512 96 0
w 512 128 512 176 0
r 176 48 288 48 0 67.0
r 176 96 288 96 0 0.22
w 288 96 416 96 0
s 512 48 512 96 0 0 false
w 288 48 512 48 0
s 240 256 272 256 0 0 false
g 176 384 176 400 0
w 464 96 496 128 0
w 496 96 464 128 0
x 404 118 422 122 0 16 L1
x 458 117 472 120 0 12 L2
x 298 367 319 371 0 16 C5
x 306 260 335 264 0 16 VT1
x 423 303 444 307 0 16 C3
x 145 199 174 203 0 16 12V
o 40 64 0 35 20.0 0.2 0 -1
o 15 64 0 35 20.0 0.4 1 -1
I changed the value of the upper left hand resistor to 67ohms (originally 1 ohm). With the original value of 1 ohm, the LEDs self extinguish when the stop switch is shorted: Now (with the 67ohm resistor) when the stop switched is 'Closed' or shorted, the LEDs will stay lit to a certain degree and there is a partial charge returning the the 'Battery' (according to the sim.).
I believe that Wesley is correct to comment that Akula has posted these schematics & videos to gain notoriety and not necessarily to help the people replicate his devices.
For me it is then a given that there will be anomalies so to speak in the wiring and or actual components which were listed on the 'Many' different schematics posted all over the web.
No-one, anywhere up to now has just 'given' away a 'correct & working' schematic to an O.U. device. I do not believe this case to be any different with Akula.
Anyhow, I find it interesting that the sim will show a 'working' feedback loop with the LEDs lit-up.
I realize it is just a sim, but it is promising........
take care, peace.
lost_bro
Dear Peterae.
RMC PWM pdf file attached.
See page 5 please. I might be near an outcome !! O0
Cheers Grum.
OK
Basically this board seems to use a FET that is switched, so you could connect this board in place of the FET on the test circuit.
but without a beefy flyback diode which Akule does not use then you may find the PWM board starts tripping on voltage and/or current.
It's not easy driving the gate of the Akula FET from your PWM board because the PWM pulls the L- to gnd, and to turn on the Akula fet we need to pull high.
Right the only way i can think of driving the Akula fet gate, is to connect your PWM L- to the gate and connect a pull up resistor from the gate to V+
This way the pull up resistor will turn the fet on, and when the PWM L- get's pulled to gnd then it will also pull the Akula fet gate to gnd, maybe use a 33 ohm in series from the PWM to the gate.
I am not sure what value of pull up resistor to use, maybe try 1K and if the turn off is slow and horrible then decrease the value.
Maybe someone more capable would throw a couple of values.
One thing to note if you do this, you must make sure your PWM is functioning before power up the Akula board, because if the PWM is not on then your Akula fet will be constantly turned on and sparks will fly very fast.
Hello Everybody,
A single picture:
(http://freenrg.info/4OUR_F/Akula_CCT_Principle.jpg)
J'ai bon? Я хорошо?
Cheers,
Jean
Jean,
Your schematic is the premise I started from also. I have to assume the RC network paralleled with D2 is significant. My hunch is this causes a delay that prompts the core to begin ringing. In a typical transformer, the ringing initiated by the MOSFET pulse would quickly dampen out, but with the 95% wrapped foil/shield in there, something "magical" happens. Somehow energy that would normally be trapped and dissipated off gets reversed. For grins, lets say it opens a portal; it captures from within the coils, energy that would otherwise get away from us. What I am pretty certain of is that you must continue to pulse the MOSFET, how often and at what duty cycle I am not clear about.
Sorry, I wish I could find better terminology to make sense of this phenomena. I think our only hope is to actively build this device and manipulate it until one of us hits the magic combination, then we all attempt to duplicate that. Eventually I think we will come to understand the fundamental principal at work here, or at least the components that seem to be most critical to its operation. What I'm afraid of are components like those diodes, they may have to have very particular characteristics in order to have a working device.
Hi EMdevices,
"...another joule thief"? Yes, IMHO, an Xtended one. For those who have the ability to synthesize. ;)
-----------------
Hi Matt Watts,
Thanks++ for your feedback.
Quote from: Matt Watts on 2014.03.30, 22:31:14
.........................................
In a typical transformer, the ringing initiated by the MOSFET pulse would quickly dampen out, but with
the 95% wrapped foil/shield in there, something "magical" happens. Somehow energy that would normally
be trapped and dissipated off gets reversed. For grins, lets say it opens a portal; it captures from within
the coils, energy that would otherwise get away from us.
..............................................
This is the kind of (Right Brain) "explanation" that I can "understand"/feel. O0
Thanks again!
Yes, IMHO, this "95% wrapped foil" is very important but I also beg to think that the "bucking coils" stuff is
also a big clue.
----------------------------------------------
About "bucking coils", you can consult:
"Bifilar Coils VS Bucking Coils"
http://jnaudin.free.fr/html/bifvsbuk.htm (http://jnaudin.free.fr/html/bifvsbuk.htm)
And also: http://freenrg.info/Some_Experiments/Bucking_Coils/ (http://freenrg.info/Some_Experiments/Bucking_Coils/)
I'm not a Scientist. Anyway, as far as I can catch it: a coil is a sort of spring or can be 'modeled' as such. No?
So, you need NRG to compress it. OK? In compensation, it gives you back his NRG, minus some losses. OK?
In the "Bucking Coil" configuration (2 springs that are kinda canceling themselves), it seems that you only need few NRG to "compress" them but, in return, these "springs" will give you back the NRG (minus some losses) that
would had been necessary to compress them should they were not canceling themselves.
Hence, some kinda "OU" behavio(u)r? Or watts?
--------------
Now, please, have a look at this picture:
(http://freenrg.info/4OUR_F/Self%20Running8.jpg)
It is not exactly the same thing that the (already quoted)
JLN's "Bifilar Coils VS Bucking Coils" but, for my part, I can perceive some similarities.
------------------------
A very bad quote from a person that I do not remember whom:
"You only need to understand the motive to write the symphony"...
Meilleures Salutations,
Jean
Quote from: EMdevices on 2014.03.30, 22:24:48
Hi Guys, I've been quite busy and haven't had much time to check the forums. I notice this topic is receiving a lot of attention, so is it worth my time to persue, or is it another joule thief (time thief)?
EM
Like JT, but high wattage and self-running for a few days (as told by the those on the Russian forums). There aren't too many with circuits built, but I'm sure there will be quite a few in the coming weeks.
Im pretty sure this thing is not a Joule Thief, at least not an ordinary one.
The caps not have enough capacity to supply alone the 494 and Mosfet driver for 1-2 days or more. Maybe the thing gets "out of tune" with time.
I believe there is a reason why akula implemented this super complicated automatic tuning system in his big device: http://s6.uploads.ru/2DLk4.jpg
@ALL
GEEZ. I made a mistake and posted before my post was ready. I will post it soon. Sorry.
wattsup
@all
While the circuit boards are in transit maybe I have some side question that will help explain the circuit much better then what we think right now because I was mentally analyzing in a micro second step by step of what is happening in the circuit in order to predict the waveform that one should see when putting a scope probe at certain points in the circuit. This way when the circuit is eventually operational, the waveforms will confirm or cancel what I think is happening. This enables me to grade my own predictions before the effects are on the table, a game I like to play to sharpen my mental equity to work out problems in the mind before doing them on the bench, but either way, questions like WHY will be abundant. Something like, I predict the scope at the L2 will show two sharp rises and one big dip in the first 1/2 cycle of the waveform. If we see this, I know my theory is consistent. hehehe
But while doing this, I realized something that just does not click.
More often then not, it is when you consider side-examples that may help explain what is happening or what is not known.
So, a small mental exercise...... if you take two transformers each on their own core, put the primaries in series, then add a dc bulb (not an LED) in series to the second primary (forget the secondaries) your are left with two open ends. With a battery you put the positive to the open end of the first primary and the negative to other end of the dc bulb. Now looking at this very rudimentary scene, we already know that to light up the bulb, it has to have a positive and a negative signal reaching both of its respective connection points otherwise no light. Right or wrong. Has to be right. Right?
So the question is this. We know the bulb is getting the negative side because it is connected directly to it. But on the other side, we also know the positive has to pass the two primary coils to reach the bulb for it to light up. Does this mean both coils are fully immersed in the positive side of the energy source? hehehe
Yes we already know the bulb will not be as bright going through both primaries because of the internal resistance of the windings but regardless of that lower output, is the positive signal in all the primaries is the question?
What test or measurement process can be done to show this answer without fuzzing up the logic that will irrevocably prove if the coils are fully positive or not? Do you see the quandary. Yes very elementary indeed but it is still a very real problem to consider. Are our instruments so limited that we could never prove this or disprove this.
I think this is the grey area we see being capitalized in the Akula circuit where if when the battery is first connected, the positive passes through the L1 while going to the load. But then, if there is positive after L1 to reach the load, how can the mosfet switch, where I figure the mosfet is an NPN? Should it not be a PNP instead?
If the out of the L1 is positive, then the mosfet is creating a straight short between the L1 output and the load. If that short on the L1 out is positive hitting the ground via the mosfet, then this has to be Tesla Ozone based, and the inductance of L1 will load the polarized C3 after the diode clips off any negative spikes resulting from the short. This means the L1 primary is fully positive biased, whereas if the coil was connected directly to the battery, half the coil would be positive biased and the other half negative biased. It has to be.
But how can that be that an NPN mosfet is passing a positive signal to ground? Or is it that when the mosfet is closed, the negative signal is reversing through the mosfet diode and entering to take up its half of the L1 coil. If that is the case then this shows that both the positive and negative signals have a forward momentum (I'm using this term for simplicity because spin conveyance explains this so well but it is not yet in the discussion books - hehehe).
It seems to me that there is something missing in how we see energy transferring in the most fundamental effect that has cloaked our eyes to the real effects. It's hard to explain but the Akula circuit is playing with these effects and if we have no vocabulary or insight to explain it, we may be stuck in the present limbo we find ourselves in with standard EE logic. So for me to understand this device will require some new brain cells being activated with a new type of logic. Hmmmm.
Imagine that I have not even integrated the effect to the L2 yet and already there are hard questions to answer.
Everything is based on how the positive and negative signal will interplay in the circuit and how this shifts the north/south polarities of the fields inside the coils and over the core.
Am I ready for the asylum because I am the only one that sees this or does this make any sense to you guys?
I'll be in the office all day so keep on truck'in.
wattsup
OMG Wattsup.
What you wrote is sooooo confusing.
Can you rewrite it so every time you write about electric potentials you specify two points between which these potentials are measured?
e.g. when you write that something is "positive", please also write the reference that this something is positive to, because the same point that is positive in respect to one reference can also be negative to another reference, at the same time.
P.S.
When fully turned on, MOSFETs conduct in both directions between source and drain, regardless whether they are P-Channel or N-Channel devices.
Hi From Other Planet,
Thanks for your picture commented in Russian.
http://s6.uploads.ru/2DLk4.jpg (http://s6.uploads.ru/2DLk4.jpg)
According to Google:
"С учетам моментов времени в процессе работы которых создается ЯМР!
means: Taking into account the time points in the process which creates NMR!
NMR?? Nuclear Magnetic resonance :o
and
"Испытания проводились в лаборатории при институте по изучению ядерной энергетики"
means: Tests were conducted in the laboratory at the Institute for the Study of Nuclear Energy.
and:
"по адресу касахстан г курчатов парк ядерных технологий"
means: At (=care of) kasahstan g Kurchatov nuclear park technology
According to google, too:
http://cryptome.org/eyeball/kurchatov/kurchatov.htm (http://cryptome.org/eyeball/kurchatov/kurchatov.htm)
Kurchatov, Kazakhstan Nuclear Research Facilities
-------------
Yeah, if wee need some "Nuclear Research Facilities" we are not out of the woods. R we? :P
-------
Anyway, a little sim:
(http://www.freenrg.info/4OUR_F/One-Small-Sim.jpg)
The code is in the attached file.
Bien le bonjour,
Jean
Dear Verpies.
Please find attached scope shot of current from L2 using an unknown output non calibrated CT on Ch1. Voltage spikes on Ch2 (5 v/cm X 100 ) Ch3 (0.5 v/cm X 100 ) Signal as seen on L1 with respect to ground
Cheers Grum.
Quote from: NerzhDishual on 2014.04.01, 18:31:45
Yeah, if wee need some "Nuclear Research Facilities" we are not out of the woods. R we? :P
We don't need a big institute. NMR can be had on a table top.
"Nuclear" is just a scare word in the context of this circuit.
Quote from: NerzhDishual on 2014.04.01, 18:31:45
Anyway, a little sim:
That sim has several qualitative differences from the Akula circuit.
The largest one being that the LED can be lit directly from the power supply, but in the Akula circuit the load LEDs cannot be lit directly from the power supply (C11).
The next difference is the ground connection to L2, that completely bypasses the load.
Why did you create the abridged and inaccurate sim if a more complete one was already posted in this thread
here (http://falstad.com/circuit/#%24+1+4.9999999999999996E-6+1.9265835257097934+34+12.0+43%0A162+416+176+416+208+1+3.0+0.0+0.0+1.0%0A162+416+208+416+240+1+3.0+0.0+0.0+1.0%0A162+416+240+416+272+1+3.0+0.0+0.0+1.0%0A162+416+272+416+304+1+3.0+0.0+0.0+1.0%0A162+416+304+416+336+1+3.0+0.0+0.0+1.0%0Av+96+352+96+288+0+0+40.0+12.0+0.0+0.0+0.5%0AT+448+96+496+96+0+0.01+3.0+2.6990593974348505+5.600381269843524E-14+0.999%0Ad+448+128+448+176+1+0.805904783%0Aw+416+176+448+176+0%0A159+320+240+320+272+0+0.1+1.0E10%0Aw+208+48+208+96+0%0AR+272+256+240+256+0+2+500.0+5.0+0.0+0.0+0.3%0Ar+416+336+416+384+0+1.0%0Ac+448+176+448+384+0+0.002+0.9797183112074054%0Aw+416+384+448+384+0%0Ac+368+336+368+384+0+1.0E-5+7.016609515630988E-14%0Aw+416+384+368+384+0%0Aw+368+384+320+384+0%0Aw+208+384+320+384+0%0Aw+368+336+416+336+0%0Aw+448+128+320+128+0%0Aw+320+128+320+240+0%0Aw+320+272+320+384+0%0Ad+480+176+544+176+1+0.805904783%0Ar+544+208+480+208+0+420000.0%0Ac+544+240+480+240+0+1.0E-10+-1.2672862759188774E-11%0Aw+480+176+480+208+0%0Aw+480+208+480+240+0%0Aw+544+176+544+208+0%0Aw+544+208+544+240+0%0Aw+448+176+480+176+0%0Aw+528+128+544+128+0%0Aw+528+96+544+96+0%0Aw+544+128+544+176+0%0Ar+208+48+320+48+0+1.0%0Ar+208+96+320+96+0+0.22%0Aw+320+96+448+96+0%0As+544+48+544+96+0+1+false%0Aw+320+48+544+48+0%0As+272+256+304+256+0+0+false%0Ag+208+384+208+400+0%0Aw+496+96+528+128+0%0Aw+528+96+496+128+0%0Ax+436+118+454+122+0+16+L1%0Ax+490+117+504+120+0+12+L2%0Ax+330+367+351+371+0+16+C5%0Ax+338+260+367+264+0+16+VT1%0Ax+455+303+476+307+0+16+C3%0Ax+65+310+94+314+0+16+12V%0Ax+415+158+436+162+0+16+D1%0Ax+502+162+523+166+0+16+D2%0Ax+562+79+610+82+0+12+Click+me%0Ax+325+319+403+322+0+11+TL494+feedback%0Ax+334+329+398+332+0+11+take-off+point%0Ax+216+307+246+311+0+16+C11%0Ac+208+384+208+256+0+0.002+-11.750888351635904%0Aw+208+256+208+96+0%0Aw+208+96+160+96+0%0Aw+96+96+128+96+0%0Aw+96+352+96+384+0%0Aw+208+384+96+384+0%0Ar+96+96+96+288+0+0.1%0As+128+96+160+96+0+0+false%0Ax+117+116+171+120+0+16+Pwr+ON%0Ao+13+16+0+34+1.25+1.6+0+-1%0Ao+55+16+0+289+10.0+204.8+0+-1%0A) ?
Quote"С учетам моментов времени в процессе работы которых создается ЯМР!
means: Taking into account the time points in the process which creates NMR!
I knew it was only time ^-^
Quote from: Grumage on 2014.04.01, 19:49:09
Please find attached scope shot of current from L2
If I had a scopeshot like that from a working device, I would be able to replicate it in one day with the same core ;(
Quote from: Peterae on 2014.04.01, 20:22:50
I knew it was only time ^-^
The two conditions of TL494's negative feedback loop ( dv/dt<x && v<0 ) are consistent with beta current pulses, since the physics of such pulses makes their edges very steep and their polarity negative relative to the stimulating/confining pulses.
Also, the copper shield would come in handy in collecting the deccelerated fast-electrons, which would otherwise cause heavy negative charge buildup and associated HV insulation breakdown.
The usual electronic operational principles prevent the Akula's circuit and transformer from exceeding unity O/I power ratio, unless something unconventional happens in that transformer.
The conventional behavior of the transformer in this circuit is pretty accurately illustrated by the simulations posted in this thread.
Copper or Aluminium ?
you realize a MnZn fettire core has a real deep skin depth.
Quote from: Peterae on 2014.04.01, 20:48:16
Copper or Aluminium ?
For the shield?
Quote from: Peterae on 2014.04.01, 20:48:16
you realize a MnZn fettire core has a real deep skin depth.
Fuzzily.
Was MnZn the ferrite in reference to which you (or someone else) had corrected me about its resistivity?
QuoteFor the shield?
Yes Shield
Thats right O0 better memory than me :-[
If MnZn is the ferrite with higher resistivity, then its skin depth must be deeper as a consequence.
Quote from: Peterae on 2014.04.01, 21:00:54
Yes, shield
I think copper foil is more convenient as an electric shield.
My dear Verpies,
Quote from: verpies on 2014.04.01, 20:17:28
..................
That sim has several qualitative differences from the Akula circuit.
.......................
The largest one being that the LED can be lit directly from the power supply, but in the Akula circuit the load LEDs cannot be lit directly from the power supply (C11).
The next one is the ground connection to L2, that completely bypasses the load ?
.........
Why did you create the abridged and inaccurate sim if a more complete one was already posted in this thread.
........
Answers:
It is not the Akula CCT. I just titled it "One-Small-Sim". No?
Why? Just because:
1) My
abridged and inaccurate sim, indeed, lits the LEDS via the power supply,
but can show some "self sustaining" behavior during about one minute (new version).
2) The
more complete sim also lits the the LEDS via the power supply and can show no "self sustaining" behavior (not even one second). Did you test it?
Now, if you believe that "I Am One Too Many Here", please just ban me if you can or get me excluded...
Basta! and Va revr gant al laboused bihan...
Le bonjour chez vous,
Jean
QuoteIf MnZn is the ferrite with higher resistivity
I thought MnZn was low resistance, OK i will try to check.
QuoteI think copper foil is more convenient as an electric shield.
How about Beta Shield, Copper or Aluminium.?
Quote from: verpies on 2014.04.01, 21:17:30
If MnZn is the ferrite with higher resistivity, then its skin depth must be deeper as a consequence.
Dear Verpies.
I have just checked my core material resistance at approx. 10K ohms. Any help ??
Magnetic Characteristics:
uiac = 2100
AL Value = 4100
(+ 30% / -20%)
Cheers Grum.
NerzhDishual
We like you being here, you dont want out do you?
PS
NiZn ferrites exhibit higher resistivity than MnZn
Quote from: NerzhDishual on 2014.04.01, 21:22:40
It is not the Akula CCT. I just titled it "One-Small-Sim". No?
If your circuit is not relevant to Akula's circuit then my comparison with it was meaningless
Quote from: NerzhDishual on 2014.04.01, 21:22:40
2) The more complete sim also lits the the LEDS via the power supply
Only for half a second during the ON transient, due to L1 and C3 forming a series LCR circuit with R5, R1 and D1. When the VT1 switch is not operating, the LEDs will not be lit after 500ms. ...in your circuit they will be lit even if the switch is not operating.
Quote from: NerzhDishual on 2014.04.01, 21:22:40
...and can show no "self sustaining" behavior (not even one second). Did you test it?
Yes, the conventional operation of these circuits is not self-sustaining and its O/I energy ratio is under unity. I already wrote this explicitly.
Powering LEDs with capacitors does not count.
Quote from: NerzhDishual on 2014.04.01, 21:22:40
Now, if you believe that "I Am One Too Many Here", please just ban me if you can or get me excluded...
You will not be banned by opposing my arguments or disagreeing with me.
I think only Ad Hominem remarks, profanity or selling Viagra would ban you from here.
Here is the full specification for the cores I am using.
Cheers Grum.
Quote from: Grumage on 2014.04.01, 21:32:58
I have just checked my core material resistance at approx. 10K ohms. Any help ??
AL Value = 4100
It's the best you can do without a spectrometer.
The resistivity is normally measured across a 1cm
3 cube with 1cm
2 wet electrodes (the electrode area matters very much in this measurement).
Peterae seems to be looking at ferrite material characteristics now and he can probably identify its composition with these values.
Quote from: Grumage on 2014.04.01, 21:45:26
Here is the full specification for the cores I am using.
So it is a ~½T MnZn ferrite with 2100μ
r r.permeability and 8Ωm resistivity (800Ω/cm)
verpies,
In your estimate, how much leeway do you think we have in regards to the core specifications? Can you determine if there are certain characteristics that would clearly push a particular core outside of a tunable range via the electronic components?
@All,
The 4 Akyla boards was snail mailed to the forum members (that requested boards) yesterday by Air mail.
The 8 "Breadboard" type boards will be Air snail mailed to forum members (that requested boards) this week.
Please ask Peter from now on, for those of you that want free Akyla boards.
GL.
Quote from: Matt Watts on 2014.04.02, 04:16:35
In your estimate, how much leeway do you think we have in regards to the core specifications? Can you determine if there are certain characteristics that would clearly push a particular core outside of a tunable range via the electronic components?
I really do not know. I don't have sufficient information what is supposed to happen inside that Akula transformer.
Given a working transformer, I can tell you what's happening inside it, but
I cannot tell you what is supposed to happen unless I postulate some AdHoc operational principle ...or at least see a scopeshot of L2 current in a working device.
I don't want to postulate an operating principle without concrete data. All I can write to you is that Akula's circuit is not OU with a conventional "transformer" behaving in the usual manner.
Generally, an E-core can form a parallel magnetic circuit while an U-core can only form a series magnetic circuit, since the flux can take only one path through a U-core (besides leakage).
Thus those cores exhibit very different flux distributions and you cannot expect one shape to behave as the other,
Faraday's power transfer efficiency will be always lower in an E-core if only one leg has a secondary winding on it and another has a primary ...but a Faraday's power transfer in a transformer is a conventional effect ...not what we are after.
If you ever see any hints of that R5/R7 junction going below ground, even for a microsecond, that will be your sign of an anomaly.
The TL494 is wired to react to that, too...
Transformers are one of the most complex electronic components and worst of all, they are not standardized.
Anything can affect their operation, such as:
1) Core
a) AL value (material and gap dependent)
b) Shape and size
c) Conductivity
d) Air gap
e) Material in the air gap (e.g. hard mica or sound dampening paper)
f) Frequency response (mag.viscosity)
g) Coercivity
h) Hysteresis loss
i) Saturation value
j) Crossection
k) Permeability
l) Magnetostriction
m) Molecular configuration
n) Nuclear configuration
o) Villari permeabilty variability
p) Mechanical compression of the core (or mechanical freedom to vibrate)
q) Any embedded permanent magnets
2) Placement and orientation of the windings in reference to:
a) the core (bobbin specs)
b) the air gap
c) the shield
d) other windings
3) Winding technique
a) single layer
b) multiple odd layers
c) multiple even layers
d) winding direction
e) winding pitch & advancement direction (can vary between layers)
f) spacing between turns
g) turn count
h) wire composition (most are copper but not all)
i) wire Litzing
j) filarity (e.g. bifilarity, aiding, opposing)
k) winding tightness/looseness/binding/gluing/varnishing
l) size of the windings (localized/narrow vs. widely spanning)
4) Isolation material
a) of the wire
b) between windings
c) between windings and core
d) between windings and shield
d) between layers
e) between core halves
5) Shield
a) material
b) size
c) shape
d) placement
e) connection
6) Pinout sides
7) External magnetic fields
8) External electric fields (see the IEEE article by Konrad & Brudny)
9) External vibrations (see
this video (https://www.youtube.com/watch?v=qauZ4WBwAOM))
10) Core and Winding Temperature
So keep the list above in mind, next time somebody says or writes: "Just wind a transformer - all the turn counts are listed on the schematic..."
It would be irresponsible of me to guess and mislead you on any of these points listed above without being certain of the M.O.
Thank you kindly Verpies. This is what I was interested in, determining how many dependent variables we are faced with--your enumeration is about double of what I started with.
Absolute attention to detail is going to be critical in getting a replication should someone set the bar. We certainly do have OUR work cut out for us with this one. I do think we are going to need most if not all of those values in that list accounted for as well as several different probe traces. Two side-by-side replications are going to be needed, where the later one, the specifications are altered one-by-one to determine which elements are critical.
Let me take a deep breath...
Oh, I think you left off bobbin specs to make me feel better.
verpies
Let's hypothesize that this is in fact a McFreey based device or at least similar functioning.
We have the core MnZn which would be the fuel.
We need to drive the core at half the cores Acoustic resonant frequency, the core being equivalent to the copper ring in the McFreey device.
If this is the case, a core can me made to exhibit large amplitude acoustic resonance by not gluing the core halves and just resisting them face to face, i know this because as an ex TV repair man, a common problem with Lineoutput Transformers is the the resin that glues each half of the ferrite core can break, or sometimes they use a clip which looses it's springiness, when it does all the youngsters in the house can find the noise unbearable @ 17KHz in UK.
I have repaired quiet a few in the past exactly for this by re gluing them.
Now if we have acoustic vibration at the interface surface of each half of the core then we would also have mechanical abrasion, although Ferrite is hard so noise is favoured over wear.
When the fet switches on, a bias flux is inserted within the core for our T1 period, although the flux is large and intense, i assume that the T1 time reduces with a strong magnetizing force, if we can work out the flux strength within the core during magnetization and we can find out the T1 time for MnZn ferrite then we would have a good idea of how long the fet needs to be turned on for(Duty %)
Then all we need is an equivalent La&Lb+C , so we need a resonant system tuned to fr/2 for energy capture which it looks to me we may have a series resonant circuit built into the Akula device, once we know the fr of the core by using a microphone then we can work out if the Akula series tuned circuit is viable, this should prove weather we are barking up the wrong tree so to speak.
If this device does use NAR then we will have to tune +--5Hz to fr/2 like the McFreey device i presume, so many wont get a device running unless they listen to the core while tuning the freq of operation, also duty does not want to be too much longer than the calculated T1 time.
Perhaps a more "elementary" approach for starters as mentioned in post 333 ?
respectfully
Chet
Quote from: Chet K on 2014.04.02, 09:14:12
Perhaps a more "elementary" approach for starters as mentioned in post 333 ?
I like it because I have not tried that before. This approach is not "more elementary" - it is different.
I doubt that Akula's transformer is configured according to this principle because the wavelength of the stimulating pulse needs to be much shorter than the length of the coil in order to obtain standing waves in it ...and the TL494 gives up above 300kHz.
Quote from: Peterae on 2014.04.02, 09:07:17
Let's hypothesize that this is in fact a McFreey based device or at least similar functioning.
That's a large leap of faith ;) What if we are wrong and we mislead others into a blind alley?
Quote from: Peterae on 2014.04.02, 09:07:17
We have the core MnZn which would be the fuel.
Yes, both Manganese and Zinc can be "unwound" and used as fuel.
Quote from: Peterae on 2014.04.02, 09:07:17
We need to drive the core at half the cores Acoustic resonant frequency, the core being equivalent to the copper ring in the McFreey device.
And what is supposed to generate the acoustic vibrations in such setup? Magnetostriction, mechanical clapping of core halves...or something else?
Quote from: Peterae on 2014.04.02, 09:07:17
If this is the case, a core can me made to exhibit large amplitude acoustic resonance by not gluing the core halves and just resisting them face to face
I know this because as an ex TV repair man, a common problem with Lineoutput Transformers is the the resin that glues each half of the ferrite core can break, or sometimes they use a clip which looses it's springiness, when it does all the youngsters in the house can find the noise unbearable @ 17KHz in UK.
I have repaired quiet a few in the past exactly for this by re gluing them.
I am familiar with the problem. Just at the 15kHz horizontal frequency of the NTSC signal ;)
In order to maximize the clapping of core halves, they must repel and later attract at least once per cycle. That necessitates two windings (one on each core half)... I don't see how two cores halves can be made to repel with just one winding (pushing against remanent field, maybe?)
If the cores don't repel, then perhaps they can bounce back after they slam into each other? Seems silly, but if the ferrite can ring then it means that it can also bounce. Note, that vinyl electrical tape and paper would dampen the bouncing...
Quote from: Peterae on 2014.04.02, 09:07:17
Now if we have acoustic vibration at the interface surface of each half of the core then we would also have mechanical abrasion, although Ferrite is hard so noise is favoured over wear.
Note that the acoustic vibration at the interface of two core halves will occur at twice the frequency of the current flowing in the windings - not necessarily at the mechanical resonance frequency of the core (which would produce the highest acoustic amplitude).
The abrasion and/or clapping of core halves would be stressful to the core up to a point of breaking it ...or at least changing its mechanical resonance frequency with wear and time.
Incidentally did not one the Russian replicators report core cracking ?
Quote from: Peterae on 2014.04.02, 09:07:17
When the fet switches on, a bias flux is inserted within the core for our T1 period, although the flux is large and intense, i assume that the T1 time reduces with a strong magnetizing force,
I cannot find the T
1 time for the MnZn ferrites anywhere :(
Quote from: Peterae on 2014.04.02, 09:07:17
if we can work out the flux strength within the core during magnetization and we can find out the T1 time for MnZn ferrite then we would have a good idea of how long the fet needs to be turned on for(Duty %)
That's sound reasoning. I think 5*T
1 would be the upper time limit for spin pre-polarizing pulses.
Quote from: Peterae on 2014.04.02, 09:07:17
Then all we need is an equivalent La&Lb+C , so we need a resonant system tuned to fr/2 for energy capture which it looks to me we may have a series resonant circuit built into the Akula device,
Which components do you have in mind?
Quote from: Peterae on 2014.04.02, 09:07:17
once we know the fr of the core by using a microphone then we can work out if the Akula series tuned circuit is viable, this should prove weather we are barking up the wrong tree so to speak.
The f
r proved to be hard to measure with a microphone, because of the dependence of the sound's amplitude on the position of acoustic nodes and antinodes.
Also my microphones are always jammed by the EM soup near the windings and cores.
Quote from: Peterae on 2014.04.02, 09:07:17
If this device does use NAR then we will have to tune +--5Hz to fr/2 like the McFreey device i presume, so many wont get a device running unless they listen to the core while tuning the freq of operation,
Yes, if NAR or NMR is at play then the tuning of the magnetic flux density (B) and frequency will have to be razor sharp.
I was thinking that maybe that weird feedback loop to the TL494 accomplishes that tuning automatically by gradually increasing the current in L1 (and B in the core) and listening for a negative sharp beta pulse in the feedback path, occuring when the B & f conditions are satisfied.
QuoteThat's a large leap of faith Wink What if we are wrong and we mislead others into a blind alley?
Sure i understand that, but there is no difference, we still build as per the Akula device as everyone else is, the only difference is we listen with a microphone and look for acoustic resonance.
Quoteacoustic vibration at the interface of two core halves will occur at the frequency of the current flowing in the windings.
Yes that's interesting hence the LOPTX at 15 or 17KHz, so would that mean NAR will occur at any frequency just as long as the amplitude is high enough to cause grinding, maybe this is more like Meyer and his 142KHz which yfree seemed to indicate was still NAR but i would have thought more likely NMR
QuoteI cannot find the T1 time for the MnZn ferrites anywhere Sad
Ah yes sorry that's me being silly then, we just use Zn or Mn which ever takes the longest :-[ as they are considered individual constituents.
QuoteWhich components do you have in mind?
I'm not really sure yet until i get a bench setup to investigate, even the Alu foil should exhibit a capacitive effect, then there's a 100pf in series with the load.
QuoteThe fr proved to be hard to measure with a microphone, because of the dependence of the sound's amplitude on the position of acoustic nodes and antinodes.
Also my microphones are always jammed by the EM soup near the windings and cores.
Indeed, during my last tests, my La&Lb coils did indeed drown the ring noise while listening.
So maybe we need somewhere else maybe a sister thread to discuss how this might work using NAR.
Quote from: Peterae on 2014.04.02, 10:26:26
It's the interface of the ferrite that is the acoustic area, this is caused by the flux pulling and pushing
I don't see how one winding could generate a flux configuration that would "push". Pull - yes.
Please reread the 4
th sub-reply in my previous message.
OK we dont need push pulling of the core halves then all we need is a changing flux, again hence the LOPTX noise.
EDIT
Yes noise in the LOPTX must be due to the surfaces grinding due to flexing of the ferrite.
Quotenot necessarily at the mechanical resonance frequency of the core (which would produce the highest acoustic amplitude).
thank you yes but what if enough amplitude can be produced from the core flexing, i mean what is enough, we can go all the way from no core chatter to self destruction of the core due to flexing if the currents were high in the transformer, so the question has to be can we start entering an area where the chatter from the core is enough to cause small scale NAR or NMR, i mean it's only a 30 Watt device, we dont need all the core to transmute it one go, we just need a small low conversion rate and maybe this can be achieved without reaching the mechanical acoustic resonant frequency.
Dear Peterae and all.
If you have no objections, I would really like to keep this thread as a whole. All the aspects kept in one place, so to speak !!
Thoughts ??
Cheers Grum.
@All,
Started to solder my Akyla board. :-)
GL.
Quote from: Groundloop on 2014.04.02, 11:47:38
@All,
Started to solder my Akyla board. :-)
GL.
Dear Groundloop.
WOW that looks really nice, well done !! O0
Now for the serious work !! :)
Cheers Grum.
Sure Grum, i was worried about filling it with discussions about NMR & NAR but yes i think we have confirmation enough to know it may well play a part in this device. O0
verpies thanks, your input here is mostly our only hope and is much appreciated.
Quote"С учетам моментов времени в процессе работы которых создается ЯМР!
means: Taking into account the time points in the process which creates NMR!
Actually from the text conversion NerzhDishual did on that last diagram then we already know it's NMR not NAR DOH
i know this is off topic(Sorry) verpies you know my suspicion for the conversion of Nitrogen into Carbon we have previously spoken about RE welders being killed by Carbon Monoxide poisoning, any idea why the plasma would cause transmutation, is it NMR that is causing this, or is it a totally new form of induced Transmutation.?
PS Nice work GroundLoop, I am tracking my boards they are with a delivery courier as we speak. O0
Quote from: Peterae on 2014.04.02, 12:03:49
any idea why the plasma would cause transmutation,
Actually plasma is the only widely accepted method of artificially causing transmutation in hot fusion.
The atoms in plasma are bared and vulnerable.
Quote from: Peterae on 2014.04.02, 12:03:49
is it NMR that is causing this, or is it a totally new form of induced Transmutation.?
I don't think that's a major cause although NMR can play a minor role in plasma.
If you study the new nuclear model by M.Mathis, it becomes obvious that the atomic nucleus is not an aggregate of little balls as shown
here (http://upload.wikimedia.org/wikipedia/commons/thumb/f/f0/Nucleus_drawing.svg/200px-Nucleus_drawing.svg.png) but it appears like
this (http://milesmathis.com/argon.avi).
You'll notice immediately that it is not nice and symmetrically round but it has internal structure. This structure is more vulnerable to bombardment from some sides then others. Bare neutrons are so vulnerable outside of the nucleus (when protons don't protect them) that half of them cannot survive for longer than 15min. Chemical bonds, electric ionization and magnetism can orient the nuclei to vulnerable positions and change them in energetic collisions.
QuoteActually plasma is the only widely accepted method of artificially causing transmutation in hot fusion.
The atoms in plasma are bared and vulnerable.
Wow OK that explains a lot, so the boffins at the top are trying to maintain a plasma to maintain fusion using strong magnetic fields & billions of GBP, seems crazy expending all that energy to confine fusion, if it can be produced intermittently using an arc, and presumably a heavier atom of Nitrogen would release more energy anyway during transmutation into Carbon.
Anyway
Regarding the Meyer device below, we need 3 things
1) a bias flux, which we have when the fet is on and we set that once we know the flux strength with the fet on to 5*T1 this is set by our Duty.
2) NMR frequency, this will depend on the flux density within the core, our current is high and therefore our flux strength is high and our NMR for Mn will be lower( the lower the better to match our TL frequency range)
3) energy capture in the Meyer device this is seen as a 400Hz LF output
So the only thing i am not yet sure of is if we have a chance to run the Mn core at it's NMR and whether there's a low frequency capture path in the Akula device, we want to maybe see high frequency ringing for NMR excitation, and a low frequency component that can be filtered out for energy reclamation.
I would be most interested in any ringing that is seen when someone runs the circuit, we need a way of measuring our transformers peak flux so we can get an estimate for our target NMR frequency.
in the Meyer device why 400Hz what sets this frequency at 400Hz?
Is there anyway we can estimate the peak flux in our transformer, i think we can be sure we are seeing 10-20 Amps, if we can work the flux strength out for 10-20 amps, then hopefully we can work out our Mn NMR range, this will give us an idea what sort of frequency we are looking for, whether it's a HF ringing affect or within the TL setting range.?
Quote from: Peterae on 2014.04.02, 14:11:55
in the Meyer device why 400Hz what sets this frequency at 400Hz?
Most likely you already know the answer.
f=1/(n*T
1)
Quote from: Peterae on 2014.04.02, 11:02:44
OK we dont need push pulling of the core halves then all we need is a changing flux, again hence the LOPTX noise.
It could be due to core halves bouncing or magnetostriction. Nonetheless we know these vibrations are induced somehow from out TV repair days and videos like
this (http://youtu.be/il0X1sNO2PM?t=4m45s) or
this (http://youtu.be/gbeseiPPCeM?t=6m7s).
Quote from: Peterae on 2014.04.02, 11:02:44
what is enough, we can go all the way from no core chatter to self destruction of the core
AFAIR, the mainstream NAR experiments (Bolef, et al) used ultrasonic transducers around 50W range.
Quote from: Peterae on 2014.04.02, 11:02:44
can we start entering an area where the chatter from the core is enough to cause small scale NAR or NMR, i mean it's only a 30 Watt device, we dont need all the core to transmute it one goty
If you'd managed to do that in one go, you'd vaporize before a nerve impulse would get to your brain.
The limiting factor is not the amplitude but magnetic flux density homogeneity.
Amplitude has to be only high enough to overcome thermalization.
I have tested a number of ferrite cores on hand and can easily tell the low resistance MnZn from the rest of the lot.
e.g. with the meter on the 20 Megohm position, the MnZn cores read 000 while other cores read 1 to 15 Meg.
When I switch to k Ohm, I get anywhere from 0.4k to 5k for the MnZn.
Probes are spaced about 1/4 inch and test anywhere on the surface.
It appears I have a few each of ring cores, rods and pot cores that are very low resistance, probably MnZn.
All of the EMI suppression slip on or clamshell cores were in the very high meg ohms.
Various ferrite rod cores from old radios were very also high resistance.
So it appears I have some MnZn ferrites to play with.
Further testing with silver paste on 1 cm^2 opposing cube faces usually would be a few hundred to a few thousand ohms.
Anything in the megohm range is usually not MnZn.
Quotef=1/(n*T1)
Right indeed i do, thanks for the recap again i apologize for my dementia.
There seems to be a lot of Mn T1 times mentioned in biological studies, presumably these are valid for our T1 time, I'm still trying to find a solid value, again we may have been over this in the past and i apologize if we have, as soon as i track a figure i will post it in this thread, it's probably good you posted f=1/(n*T1) because not everyone would be aware of this in this thread.
QuoteAFAIR, the mainstream NAR experiments (Bolef, et al) used ultrasonic transducers around 50W range.
Depending on coupling, if they glued a piezo on to a surface i am wondering it that greatly reduces efficiency and therefore although there maybe 50 watts in the acoustic energy maybe a lot less due to the surface of the piezo being restrained from movement due to glue, i know the sound will travel due to mechanical coupling, but it just seems an inefficient coupling method to me, maybe i am wrong, hard grinding ferrite surfaces may well be quiet an efficient way to produce acoustic resonance, who knows it's certainly worth looking at.
We could estimate the acoustic power by listening to the core's noise and comparing to that of a glued piezo i suppose.
Some data on MN55
http://chem.ch.huji.ac.il/nmr/techniques/1d/row4/mn.html
QuoteProperty Value
Spin 5/2
Natural abundance 100%
Chemical shift range 3000 ppm, from -3000 to 0
Frequency ratio (Ξ) 24.789218%
Reference compound 0.82 m KMnO4 in D2O
Linewidth of reference 5.6 Hz
T1 of reference 0.072 s
Receptivity rel. to 1H at natural abundance 0.179
Receptivity rel. to 1H when enriched 0.179
Receptivity rel. to 13C at natural abundance 1050
Receptivity rel. to 13C when enriched 1050
Linewidth parameter 350 fm4
QuoteT1 of reference - Longitudinal relaxation time of the reference compound at room temperature in a 400 MHz spectrometer.
I cant see any mention at which Flux density this is measured at, maybe there is a standard flux used for measurement, i will keep looking.
wow T1 is long at 72mS, man we need to get that down with a higher flux that's 360mS at our *5
Dear Peterae.
You and others may find this website of interest.
http://www.webelements.com/manganese/isotopes.html
This might also be of interest as it relates to NMR frequency with relation to magnetic field strength. http://www.geotech1.com/cgi-bin/pages/common/index.pl?page=lrl&file=/info/nmr.dat
From there you can find many NMR analyses !!
Cheers Grum.
Thanks Grum
OK i have the wrong way around, a stronger mag flux increases the NMR, this is interesting because it could mean we have a vast range of flux in that core, starting with the T1*5 period the NMR fr will be high, but as the flux decreases and so will our NMR.
I will try some calcs when i get some time.
PS my order for Groundloop's board have arrived when i got home, the quality looks really good at first glance.
I will start doing the packaging once at work tomorrow, ready to ship out, i will send 15 to Chet and keep 15 here for UK & Europe including 1 to Groundloop for inspection.
Quote from: Groundloop on 2014.04.02, 11:47:38
Started to solder my Akyla board. :-)
I am curious which ones are the DTY OSC PHZ pots.
e.g.: OSC =
R15 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307), etc...
Quote from: Peterae on 2014.04.02, 17:14:07
OK i have the wrong way around, a stronger mag flux increases the NMR,
Darn, I noticed the inversion but forgot to mention it.
Manganese resonates at 1057Hz/Gauss (or 10.570MHz/Tesla) and 100% of naturally occurring Manganese is susceptible to NMR (and NAR).
Compare it with only 4.1% for Zinc which resonates at 267Hz/Gauss (2.67MHz/T).
P.S.
The Meyer device
diagram (http://www.overunityresearch.com/index.php?action=dlattach;topic=2194.0;attach=11534) from the 1976 article in Science et Vie, depicts a Copper rod.
Meyer's patents FR2680613 (http://freenrg.info/Patents/FR2680613%20%28MEYER%29/) and FR9110472 and FR2385255 (http://freenrg.info/Patents/FR2385255%20%28MEYER%29/) and CZ284333 (http://spisy.upv.cz/Patents/FullDocuments/284/284333.pdf) all mention an Iron rod.
The patent FR9110472 is alleged to contain a crucial detail necessary for replication of the Meyer's device.
I never read this patent.
In his embodiments, Meyer attempts to saturate the Iron rod, because without saturation, its magnetic permeability is so high, that the evil Skin Effect prevents any significant penetration of high frequency magnetic fields into the rod.
FYI:
Iron saturates at 2.14T. Only 2.2% of Iron responds to NMR and NAR and its responsive isotope, Iron-57 has nuclear resonance frequency of 138Hz/Gauss.
69% of Copper resonates at 1132Hz/Gauss and 31% of Copper resonates at 1210Hz/G.
100% of Aluminum resonates at 1110Hz/G.
Quote from: verpies on 2014.04.02, 17:17:05
I am curious which ones are the DTY OSC PHZ pots.
e.g.: OSC = R15 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307), etc...
Verpies,
You will need to look at the Eagle CAD design files to see component names and values.
GL.
Thanks for posting that verpies O0
Our TL494 has a range from 1KHz to 300KHz
so we have a range of field strength in the core between
G=fr/1057
g=1000/1057 = 0.9461G @ 1khz or 94.61 uT micro Tesla
g=300000/1057 = 283.8221G @ 300KHz or 28.38221 mT milli Tesla
to give an idea at 0 latitude the earths mag flux is 31.869 µT
I sure would like to see what the flux looks like in that core.
Now if that site was correct for Mn having a T1 of reference 0.072 s then we need to be on for 360mS which is no good because that gives us a maximun working frequency of the reciprocal giving 2.778Hz LOL
So we need to get our T1 time down by presumably using a much stronger magnetizing flux when our fet is on.
So how does B affect T1
We want a B value for a T1 time of 3 uS because if that was our 100% ON time then our TL494 has a max frequency off 300 KHz to cover it's full range of 1-300 KHz we always need a max of 3uS for T1.
We want a B value for a T1 time of 3 uS any ideas how to get our T1 time set to 3uS?
I think he means if osc is R15 on the diagram then what Circuit reference is the DTY & PHZ, i think he just wants to work out how the 3 pots fit into the circuit we have.
Quote from: Peterae on 2014.04.02, 19:41:10
I think he means if osc is R15 on the diagram then what Circuit reference is the DTY & PHZ, i think he just wants to work out how the 3 pots fit into the circuit we have.
Peter,
Yes, and he can do that by looking at the design files.
MY NAME: AKYLA NAME:
DUT = R13 = 4K7 R11 = 4K7
OSC = R12 = 10K R15 = 10K
PHZ = R18 = 500K R2 = 420K
GL.
OOPS sorry GL offcourse O0
If i was clever this maybe how i would design the Akula device.
I would use the Fet ON period for T1 biasing with a large current, i would then design a circuit so that when the fet is off i would have a small current flowing in the secondary, this should set the flux in the transformer to a certain level and maintain it at say 1.5mT for example.
Now if our flux was driven from the charged secondary load circuit, it should be possible to use the secondary voltage level or current as a feedback system by telling the TL494 to adjust it's frequency closer or further away from fr, this way the TL chip would automatically track fr and maybe even keep away from the exact fr by a certain amount which is set to the specifiacations of the transformer ratings to stop damage or worse.
Quote from: Peterae on 2014.04.02, 20:06:40
OOPS sorry GL offcourse O0
If i was clever this maybe how i would design the Akula device.
I would use the Fet ON period for T1 biasing with a large current, i would then design a circuit so that when the fet is off i would have a small current flowing in the secondary, this should set the flux in the transformer to a certain level and maintain it at say 1.5mT for example.
Now if our flux was driven from the charged secondary load circuit, it should be possible to use the secondary voltage level or current as a feedback system by telling the TL494 to adjust it's frequency closer or further away from fr, this way the TL chip would automatically track fr and maybe even keep away from the exact fr by a certain amount which is set to the specifiacations of the transformer ratings to stop damage or worse.
Peter,
Or, we could use a PIC micro controller that have PWM, A/D converters, timers and a RS232 interface to a PC,
and do everything in FW and in SW. Just a thought...........
GL.
indeed that goes through my mind also GL, that has to be the final device i think O0
QuoteNote that the rates of T1 relaxation are generally strongly dependent on the NMR frequency and so vary considerably with magnetic field strength B. Small amounts of paramagnetic substances in a sample speed up relaxation very much.
http://en.wikipedia.org/wiki/Relaxation_%28NMR%29#Microscopic_mechanism
Is zinc by any chance paramagnetic? i've found conflicting info so far, some say yes and some say no 'Scratch Head'
Quoteit depends on the charge of the ion
https://answers.yahoo.com/question/index?qid=20071115090201AAxaXqK
or
QuoteZn ion is paramagnetic whereas Zn is diamagnetic
Do we have any Zn ION's present in that core? we only need Small amounts
EDIT
maybe this is only liquids?
Dear Verpies.
As requested.
Ch1 Current in L2
Ch2 Voltage spike at LED input. (Probe on X 100)
Ch3 Voltage across R5 C5.
Cheers Grum.
Quote from: Grumage on 2014.04.02, 21:08:04
Ch1 Current in L2
Ch2 Voltage spike at LED input. (Probe on X 100)
Ch3 Voltage across R5 C5.
I'll have to think about it. Negative pulses across carbon-film resistor (R5) are strange. The TL494 will be triggered by them.
Let's verify that TL494 is really reacting to these feedback pulses by scoping pin 3 and 11 while triggering the scope on pin 5.
P.S.
Guys, thanks for not making me install Eagle just to see what those three pots were about.
I also do a lot of mundane footwork when I reply to some of your messages.
Quote from: verpies on 2014.04.02, 23:32:10
I'll have to think about it. Negative pulses across carbon-film resistor (R5) are strange. The TL494 will be triggered by them.
Let's verify that TL494 is really reacting to these feedback pulses by scoping pin 3 and 11 while triggering the scope on pin 5.
So you need much lower value of C5 to lets say 10nF for getting driver to ring (like amplifier's microphone near speakers) and have it self adjusted to resonant noise in additioon to main frequency and see what you can achieve.
Cheers!
Quote from: T-1000 on 2014.04.03, 08:20:25
So you need much lower value of C5 to lets say 10nF for getting driver to ring
1000x decrease of C5 (to 10nF) would be a significant modification of Akula's original circuit. I don't think Grum is ready for such mods yet.
BTW: What constitutes "the driver" in your sentence?
If anyone requires an Akula PCB please PM me with your full name and address, they are free of charge O0
I already have some orders and details (you know who you are :) )
Thanks
Peter
Quote from: verpies on 2014.04.03, 10:07:36
1000x decrease of C5 (to 10nF) would be a significant modification of Akula's original circuit. I don't think Grum is ready for such mods yet.
BTW: What constitutes "the driver" in your sentence?
With 10 microfarad capacitor it is steady DC and from there it goes back to driver circuit where pot should get sweet spot between pulsing and no pulsing at all. So with lower values of capacitor it should start making pulses to feed back path and that would change some bits in way circuit is supposed to work.
P.S. The driver circuit is source signal generating circuitry up to MOSFET gate.
Cheers!
All PCB board requests have been sent out this morning apart from GL which will happen a little later, i have 8 PCb's still available for UK & Europe, and there will be about 10-12 in the USA for anyone over that side of the pond still available.
About 6 or more years ago I had the hunch or intuition that a flyback converter or blocking oscillator that was bootstrapped i.e output fed back into the power source would be an ideal test bed for what I termed at that time "exotic core materials".
I built up a breadboard and fiddled with it blindly for hours.
More recently I posted a few ideas on how to loop a blocking oscillator to test for OU effects, and in a few posts in other threads alluding to the idea that perhaps exotic core materials might yield OU from the magnetic pumping of the core at the right frequency etc.
information here: http://www.overunityresearch.com/index.php?topic=604.0
I think we are on the right track as a close examination of the Akula30 Watt does indeed show a loopback of the power supply output and in addition a string of leds to soak up some of the energy and act as a clipper so that the system voltage does not soar.
The beauty of using a switchmode converter chip is much more control over current charging of the inductor, duty cycle and frequency control.
It is great that we now have some target materials to consider as fuel.
Great stuff ION
Yes it's fantastic to know where the so called free energy comes from and is not just magic ;D
You were also getting interesting noise from a common mode choke that was unexplainable which i wonder if is applicable.
I have a video of Don Smith giving a talk, he was taking questions at the end and someone asked can you please draw a simple OU circuit for us, he proceeded to draw a single transistor (I think oscillator) that had a coil wound around a high Mu rod, he said here this is a simple OU device or something similar.
I will try to hunt that video down again. ^-^
this LED (http://www.ebay.de/itm/OSRAM-Semiconductors-LECWE2B-Uni-Color-White-Chip-LED-/360697925575?pt=Bauteile&hash=item53fb458fc7) I proposed is NOT close to akula's 15V 3W specs. I just did a quick test, at 15v power consumption is more than 12watt, according to PSU display. Couldnt check more accurately for now, as i damaged all my Multimeters last days (need to repair :P) and have no clue where my 0,1ohm resistors for oscilloscope are
Next I will order and try one of these:
http://www.ebay.de/itm/OSRAM-Semiconductors-LECWE3B-Warm-White-4500K-Chip-LED-/360697852277?pt=Bauteile&hash=item53fb447175
concerning core material: I think it is only used for tuning/lowering frequency, possibly efficiency too, big akula device has air core and only uses ferrite in driver part of the Tesla Coil (unless this is where the magic happens)
kind regards and peace,
From other Planet
Quote from: From Other Planet on 2014.04.04, 14:02:05
this LED is NOT close to akula's 15V 3W specs.
That 15V rating is for 5 LEDs connected in series.
Quote from: From Other Planet on 2014.04.04, 14:02:05
I just did a quick test, at 15v power consumption is more than 12watt, according to PSU display.
...and that lone LED survived with 15V across it !? ...or did the power supply limit the current?
Quote from: verpies on 2014.04.04, 16:28:16
That 15V rating is for 5 LEDs connected in series.
...and that lone LED survived with 15V across it !? ...or did the power supply limit the current?
Are you sure about the 15v being meant in series? Then i totally misunderstood the ratings. Thats why i am looking for a 15v 3w all the time haha ;D
Yes LED survived, it is made for 16,4v according to their specs, of course gets pretty hot, so i placed it on a heat sink
(http://freenrg.info/4OUR_F/2DLk4_trad3.jpg)
I have tried to fully translate the Akula's page, using some automatic translations.
Most of time I had to guess. :-X
This sentence remains strange:
Чистоты указаны по показанием приборов И имеют ориентировочное значения
их соотношения Чистота ТТ перед.
Google:
"Purity are on the instruments and have an approximate value of their relationship Cleanliness TT before."
Reverso:
"Cleanliness they are specified on instrument reading And have specifications tentative of the value of their
correlation Cleanliness before."
Paralink:
"Cleanliness are specified under the reading of devices And have rough values of their parity Cleanliness ТТ before."
Bing:
"Purity are the indications are approximate values and their relationship Was the TT before."
Systran:
"Cleanliness are indicated on by reading instruments and have tentative of the value of their relationship cleanliness tt before."
Yes, very clean(liness).... ;D :D
========================
The Smith Conference video is here: https://www.youtube.com/watch?v=cQkYAh8Qgb4 (https://www.youtube.com/watch?v=cQkYAh8Qgb4)
About 2 hours. The stuff about the simple 'OU'(?) CTT (using metglas) begins at: 57:42
I just saw this over on the A 30w thread over on OU.com
It is a self running simulation using 555 timer. Cool! 8)
Quote from: Mr. Teslonian
I dunno if this will work or not. Just posting a link as everyone is struggling with Akula (Roman) Circuit. Just have a look at it.
https://www.youtube.com/watch?v=cD_Bwwvlc3I
--- Mr Teslonian
Edit: I wonder if this would work on a more powerful simulator like Multisim? This circuit is like the RomeroUK circuit Jean posted.
Jean,
Great detective work translating Akula's schematic. A couple of highlights:
1. Clearly the core is wrapped with a shield that is split, as to not make a complete loop. In this design though, he does not connect the shield to ground.
2. And he definitely considers NMR as the source energy, also as indicated by "Park Nuclear Technology".
Still a little uncertain of the term "TT". Looks to be associated with timing. Maybe a PLL (phase lock loop)... He is talking about matching or coordinating of time points, which in my mind would be phase synchronization. So to get the Akula device to work, we must know the event pattern in advance and be hunting for the proper response before we trigger the next cycle. From what we know about the 494 chip, things are starting to make some sense here.
Also, do we know anything about G. Kurchatov and the kind of work he is/was doing?
I would like to know what TT means also. I see it used a lot over on the Russian forums.
Quote from: 4Tesla on 2014.04.04, 19:01:08
I would like to know what TT means also. I see it used a lot over on the Russian forums.
If I had to bet, my money would be on it being Russian characters used as an acronym that is at the heart of this whole phenomena.
Quote from: Matt Watts on 2014.04.04, 18:50:35
1. Clearly the core is wrapped with a shield that is split, as to not make a complete loop. In this design though, he does not connect the shield to ground.
The "core" is an empty PVC pipe, i think he shows that in video too. Not sure now.
TT means Tesla Transformator (тесла трансформатор) = Tesla Coil, HV transformer, ...
@Jean: thanks very much for translation, maybe the russian speaking guys here can clarify on the "unclean" translated sentence
Quote from: From Other Planet on 2014.04.04, 19:23:50
The "core" is an empty PVC pipe, i think he shows that in video too.
So in this case, if there is any NMR, it would have to manifest itself in the copper coils and aluminum shield correct? There is no ferrite core.
It would also seem to me that a spectral analysis contains information of the resonance for both the windings and the NMR--one would have to know which to look for. If you added a ferrite core, now there would be three peaks to consider. Or do I have this all wrong? Maybe it has always been the NMR causing the resonance. Have we been working with the
effect instead of the
cause?
In the video where akula explains principle, he says effect is happening in copper coil of receiver (https://www.youtube.com/watch?v=wuoISZAlFAc). The shield IMHO is only part of tuning mechanism (capacity, capacitive coupling)
Quote from: From Other Planet on 2014.04.04, 19:23:50
TT means Tesla Transformator (тесла трансформатор) = Tesla Coil, HV transformer, ...
Great finding... O0 Thanks! I also was puzzled with this 'TT'. :-[
-------------
BTW(?), IMHO, it would be profitable to consult the ION's (short (and now dead)) thread..
Would it not be? http://www.overunityresearch.com/index.php?topic=604.0 (http://www.overunityresearch.com/index.php?topic=604.0)
2011-01-20 !
Sometime, some threads die, perhaps because they are ahead of one's time?
---------------------
It it question of "spectral analysis", here:
Akula video #5 translated by Wesley. ( Free Energy device)
https://www.youtube.com/watch?v=PPHipGkfSAY&
OK guys! I can not resist, should I be picked apart, by any "Veep Sir" (Andy's anagram solver).
BTW, are you aware that there are some hidden meanings in mere names?
For ex:
"Salvador Dali" = Avide à Dollar (Dollar greedy).
"NerzhDishual" = he shun lizard (for ex).
Yes, this is not "scientific"... :P
--------
IMHO, these NMR stuffs are important indeed and must add something.
But, I'm just wondering whether these are not some 'digarechou'.
(an untranslatable Breton word that can mean something like "'departure from the subject' + nitpicks
+ muddying the water".
------------
I'm also just wondering whether all these "Blocking Oscillator(s) aka Joule Thief(s)"
are not obviously and durably 'OU' only because we do not properly wire some coil.
3 Pictures.
I'm just claiming that wrapping a simple "off the shelf" inductor with a (non shorted) "off the kitchen"
bl' aluminium foil and connecting the said inductor (that is becoming an inductor-capacitor) in strange (IMHO)
ways can lead to interesting results.
This (non-shorted) f' aluminum foil is not my idea at all.
Simply, I use my intuition first, instead of my critical mind, when I see some non conventional claims.
(http://freenrg.info/4OUR_F/CCT-RLC.jpg)
(http://freenrg.info/4OUR_F/C-L-CCT.jpg)
(http://freenrg.info/4OUR_F/Coils_Cap_Small.jpg)
Quote from: 4Tesla on 2014.04.04, 18:36:52
I just saw this over on the A 30w thread over on OU.com
It is a self running simulation using 555 timer. Cool! 8)
Edit: I wonder if this would work on a more powerful simulator like Multisim? This circuit is like the RomeroUK circuit Jean posted.
Quote from: Mr. Teslonian
I dunno if this will work or not. Just posting a link as everyone is struggling with Akula (Roman) Circuit. Just have a look at it.
https://www.youtube.com/watch?v=cD_Bwwvlc3I
--- Mr Teslonian
Here is a link to the simulation:
http://falstad.com/circuit/#%24+1+5.0E-6+7.308818067910767+64+7.0+50%0Aw+240+176+208+176+0%0Ar+208+176+208+240+0+10000.0%0Aw+208+240+240+240+0%0Aw+208+240+208+272+0%0Aw+208+272+240+272+0%0Ac+208+272+208+336+0+3.0E-7+9.903216635994851E-9%0Ar+208+176+208+112+0+10000.0%0Aw+208+112+304+112+0%0AO+368+208+432+208+0%0A165+240+144+256+144+2+0.0%0Aw+304+112+368+112+0%0Aw+368+112+368+176+0%0Aw+368+208+480+208+0%0AT+480+208+544+288+0+4.0+3.0+5.824441557167402E-26+-1.9434239845923315E-26+0.999%0A162+544+208+544+288+1+2.0+1.0+0.0+0.0%0Ad+544+208+544+112+1+0.805904783%0Aw+544+112+544+80+0%0Aw+544+80+112+80+0%0Aw+208+112+112+112+0%0Aw+112+80+112+112+0%0As+112+176+112+112+0+1+true%0Av+112+336+112+176+0+0+40.0+1.5+0.0+0.0+0.5%0Aw+544+336+544+288+0%0Ag+208+336+208+352+0%0Ag+480+288+480+320+0%0Aw+208+336+112+336+0%0Aw+112+336+112+384+0%0Aw+112+384+544+384+0%0Aw+544+384+544+336+0%0Ao+5+32+0+35+0.001220703125+9.765625E-5+0+-1%0Ao+8+32+0+42+7.62939453125E-5+9.765625E-5+1+-1%0Ao+14+64+0+35+7.62939453125E-5+9.765625E-5+2+-1%0A (http://falstad.com/circuit/#%24+1+5.0E-6+7.308818067910767+64+7.0+50%0Aw+240+176+208+176+0%0Ar+208+176+208+240+0+10000.0%0Aw+208+240+240+240+0%0Aw+208+240+208+272+0%0Aw+208+272+240+272+0%0Ac+208+272+208+336+0+3.0E-7+9.903216635994851E-9%0Ar+208+176+208+112+0+10000.0%0Aw+208+112+304+112+0%0AO+368+208+432+208+0%0A165+240+144+256+144+2+0.0%0Aw+304+112+368+112+0%0Aw+368+112+368+176+0%0Aw+368+208+480+208+0%0AT+480+208+544+288+0+4.0+3.0+5.824441557167402E-26+-1.9434239845923315E-26+0.999%0A162+544+208+544+288+1+2.0+1.0+0.0+0.0%0Ad+544+208+544+112+1+0.805904783%0Aw+544+112+544+80+0%0Aw+544+80+112+80+0%0Aw+208+112+112+112+0%0Aw+112+80+112+112+0%0As+112+176+112+112+0+1+true%0Av+112+336+112+176+0+0+40.0+1.5+0.0+0.0+0.5%0Aw+544+336+544+288+0%0Ag+208+336+208+352+0%0Ag+480+288+480+320+0%0Aw+208+336+112+336+0%0Aw+112+336+112+384+0%0Aw+112+384+544+384+0%0Aw+544+384+544+336+0%0Ao+5+32+0+35+0.001220703125+9.765625E-5+0+-1%0Ao+8+32+0+42+7.62939453125E-5+9.765625E-5+1+-1%0Ao+14+64+0+35+7.62939453125E-5+9.765625E-5+2+-1%0A)
Note run away current on LED.
Quote from: 4Tesla on 2014.04.04, 18:36:52
It is a self running simulation using 555 timer. Cool! 8)
Edit: I wonder if this would work on a more powerful simulator like Multisim?
No, because the 555 timer is badly simulated in Falstad. Its supply power is less than its output power. More energy comes out of the Out pin than goes into the V
in pin.
So i have been working through the parts list for GL's board, mainly for Grumage, to compile a Farnell parts list and price.
I am going through the parts list trying to find Farnell parts as this is the logical place to order from in the UK
First i would order a resistor pack from China, you will get every value you will ever need for other projects also, pot & Cap can also be bulk bought.
Something like this would do the job 1%
http://www.ebay.co.uk/itm/2500Pcs-1-4w-1-Metal-Film-Resistor-Kit-50-Values-Assortment-Pack-Mix-1R-10MR-/350815183216?pt=UK_BOI_Electrical_Components_Supplies_ET&hash=item51ae36e170
Parts from Farnell
C1,C6,C8,C12,C14 100nF http://uk.farnell.com/vishay/k104k15x7rf53h5/cap-mlcc-x7r-100nf-50v-rad/dp/1141777 £0.36 5 Off Min order
C2,C3 22nF http://uk.farnell.com/multicomp/mcrr50223x7rk0100/cap-cer-x7r-22nf-100v-rad/dp/1216433 £1.00 5 Off Min order
C4,C9 10uF/35V http://uk.farnell.com/rubycon/35ml10mefc4x7/cap-alu-elec-10uf-35v-rad/dp/8126542 £0.43 2 Off
C5 3nF 3.3nF http://uk.farnell.com/avx/sr215c332kar/cap-mlcc-x7r-3-3nf-50v-rad/dp/1100387 £0.89 5 Off Min Order
C7 100uF/35V http://uk.farnell.com/rubycon/50yxj100mt78x11-5/cap-alu-elec-100uf-50v-rad/dp/2346269 £0.173 1 Off
C10,C11 2200uF 50V http://uk.farnell.com/multicomp/mclpr50v228m22x25/cap-alu-elec-2200uf-50v-snap-in/dp/9452885 £2.92 2 Off
C13 68pF http://uk.farnell.com/avx/mr051a680jta/cap-mlcc-c0g-np0-68pf-100v-rad/dp/669714 £0.72 1 Off
C15 4...37 pF 4-40pf http://uk.farnell.com/vishay-bc-components/bfc280908002/cap-var-top-bottom-4pf-to-38pf/dp/1215715 £4.77 1 Off Really expensive try else where
D1 1N4004 http://uk.farnell.com/vishay-general-semiconductor/1n4004-e3-54/rectifier-standard-1a-400v-do-204al/dp/2335181 £0.0675 1 Off
D2 1N4148 http://uk.farnell.com/fairchild-semiconductor/1n4148ta/diode-ultrafast-300ma-100v-do-204ah/dp/2322485 £0.009 1 Off
D3,D4 BZX85/C15 http://uk.farnell.com/vishay/bzx85c15-tap/diode-zener-15v-1-3w-5-do41/dp/1779209 £0.13 2 Off
D5,D6 15SQ045 Farnell Dont seem to do this diode???
IC1 TL494IN http://uk.farnell.com/texas-instruments/tl494in/ic-controller-pwm/dp/1470529 £0.39 1 Off
IC2 7812 http://uk.farnell.com/fairchild-semiconductor/lm7812ct/v-reg-14-5v-to-35v-to-220/dp/1467365 £0.45 1 Off
J1,J2,J3 ST,Ld,SP http://uk.farnell.com/multicomp/mc24366/terminal-blk-eurostyle-2pos-26/dp/2396251 £0.504 3 Off
L1 470uH 1A http://uk.farnell.com/wurth-elektronik/7447010/fi-line-choke-470-0uh-1000ma/dp/2082522 £1.60 1 Off
L2 EPD29/16/10 Not sure about this, i am sure others already know source
LED1 LED GREEN http://uk.farnell.com/multicomp/703-0097/led-5mm-green-150mcd-572nm/dp/2112108 £0.122 1 Off
Q1 BD139 http://uk.farnell.com/stmicroelectronics/bd139/transistor-npn-45v-1-5a-sot-32/dp/1015770 £0.235 1 Off
Q2 IRF3205 http://uk.farnell.com/international-rectifier/auirf3205z/mosfet-n-ch-55v-75a-to220ab/dp/1864527 £1.70 1 Off
R1 6,8R 2W http://uk.farnell.com/welwyn/whs2-6r8ja25/resistor-surge-6r8-2w-250v-5/dp/2251034 £0.81 5 Off Min Order
R2,R7,R19 1R 2W http://uk.farnell.com/welwyn/whs2-1rja25/resistor-surge-1r-2w-250v-5/dp/2251029 £0.985 5 Off Min order
R3 750R
R4 1K
R5 4,02K
R6 210R
R8 27K
R9 4,02K
R10 6,98K
R11 1K
R12 10K http://uk.farnell.com/bourns/3299w-1-103lf/pot-trimmer-10k-25-turn-3-pin/dp/9353704 £1.44 1 Off
R13 4K7 5K http://uk.farnell.com/bourns/3299w-1-502lf/trimmer-25-turn-5k/dp/9353801?Ntt=BOURNS+3299W+5k+pot £1.42 1 Off
R14 2K
R15 12K
R16 5,1K
R17 0,22R 2W http://uk.farnell.com/welwyn/wp2s-r22ja25/resistor-r22-2-watt-5/dp/1219190 £0.84 5 Off Min Order
R18 500K http://uk.farnell.com/bourns/3299w-1-504lf/trimmer-25-turn-500k/dp/9353828?Ntt=BOURNS+-+3299W+500k £1.43 1 Off
R20 1K
Here's also a layout with component values and location for soldering components.
Peter,
Great work on the part list.
I have a pause in my soldering, waiting for parts. I'm one resistor short and I also did
order another core for the coil. So my board will be done in a week or two. I'm
still waiting for delivery of the "Bread board" type experiment board. I will send this
board to members, probably next week, as fast as I get them.
GL.
Thanks GL
Yeah it all takes time.
@ all,
I have attached a few shots of my initial build on stripboard. I have kept as close as possible to Akula's component values. The small stripboard with 5 x 1R2 paralleled resistors makes up R3 (on Akula's schematic) , and comes out at around 0.24R, as I did not have a 0.22R. I used an array of 10 x 3W LED chips manufactured by AVAGO, type ASMT-JW32-NWXH1 (see attached data sheet).
My initial observation on power-up from my bench supply was that I could not increase the supply voltage above 6.8V without hitting my set current clamp at 5A. The 'stop' switch worked OK and current draw was only increased slightly when operated. Frequency range was 17KHz to 138KHz. I have attached a waveform scoped (x10 probe) at the mosfet drain with respect to ground. My advice to other builders is not to power-up straight from a 12V battery, as its quite possible that your LED's will pop! Pot R11 appeared to do nothing at all. The ferrite coil former was first wrapped with a gaped copper foil prior to winding the coils. Connecting the foil to earth did not appear to change anything. I tried gaping the ferrite 'E'-core's and butting them but could not detect any audible ringing. I also reversed the coil phasing, again with no noticeable changes in operation.
Thanks to Peterae for sending me a PCB which I will populate a bit further into this project.
Hoppy
Hi Hoppy
No problem, thanks for sharing your great build & Results, It will be interesting to see how the vero performs against a pcb build O0
Great Work
Peter
I found these 12v LED panels for car dome light. Any other source for LED panel/load?
http://ledpackages.com/by-function/dome-map-light.html?SID=iojlgn1g340lfn8bn8eg5qqmp4&limit=25 (http://ledpackages.com/by-function/dome-map-light.html?SID=iojlgn1g340lfn8bn8eg5qqmp4&limit=25)
Quote from: Hoppy on 2014.04.05, 14:44:57
I have attached a waveform scoped (x10 probe) at the mosfet drain with respect to ground.
How about voltage across R5
Quote from: 4Tesla on 2014.04.05, 15:42:48
I found these 12v LED panels for car dome light. Any other source for LED panel/load?
http://ledpackages.com/by-function/dome-map-light.html?SID=iojlgn1g340lfn8bn8eg5qqmp4&limit=25 (http://ledpackages.com/by-function/dome-map-light.html?SID=iojlgn1g340lfn8bn8eg5qqmp4&limit=25)
www.dx.com - good prices but shipping can be slow depending on where you are located. DX (formerly dealextreme) is based in China but has some products shipping out of U.S. (usually higher price to get them from U.S. to U.S than China to U.S. :o ) and they have a huge assortment of LED types.
Quote from: verpies on 2014.04.05, 16:05:19
How about voltage across R5
Shot taken across R5 with probe on top of resistor. Supply voltage 6.8V with LED's very bright and the mosfet getting very hot!!
Quote from: e2matrix on 2014.04.05, 16:15:39
www.dx.com - good prices but shipping can be slow depending on where you are located. DX (formerly dealextreme) is based in China but has some products shipping out of U.S. (usually higher price to get them from U.S. to U.S than China to U.S. :o ) and they have a huge assortment of LED types.
Wow! Great site! And they have 30w 15v LEDs!! :)
http://www.dx.com/p/30w-2500lm-3200k-warm-white-light-30-led-emitter-plate-15-18v-189569 (http://www.dx.com/p/30w-2500lm-3200k-warm-white-light-30-led-emitter-plate-15-18v-189569)
Quote from: Peterae on 2014.04.05, 12:03:09
So i have been working through the parts list for GL's board, mainly for Grumage, to compile a Farnell parts list and price.
Dear Peter.
Just got back to finding all these posts !! :) I really appreciate the immense effort I shall be ordering later.
Cheers Grum.
Quote from: Hoppy on 2014.04.05, 16:23:38
Shot taken across R5 with probe on top of resistor. Supply voltage 6.8V with LED's very bright and the mosfet getting very hot!!
Good evening Hoppy.
I don't think there is any doubt about that negative going pulse !! :)
I have a couple of questions.
1. Is your transformer wound 15 one end and 45 the other, split between ?
2. What is your duty cycle ?
I am still testing with a simple Mosfet PWM with less than 10% duty and current measured with DVM from 12VDC 7 AH LAB. With this set up I rarely reach even 1 Amp !!
Cheers Grum.
Quote from: 4Tesla on 2014.04.05, 16:58:17
Wow! Great site! And they have 30w 15v LEDs!! :)
http://www.dx.com/p/30w-2500lm-3200k-warm-white-light-30-led-emitter-plate-15-18v-189569 (http://www.dx.com/p/30w-2500lm-3200k-warm-white-light-30-led-emitter-plate-15-18v-189569)
Dear 4Tesla.
These are the LED's I am using for a load. I have 5 in series times two. I have then put them in parallel to provide a 10W load.
http://www.ebay.co.uk/itm/251088281787?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2648
This however must be the cheapest source !!
http://www.ebay.co.uk/itm/50x-High-Power-1W-LED-Light-Chip-Energy-Saving-Lamp-Beads-Bulb-For-DIY/151217038468?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D222003%26algo%3DSIC.FIT%26ao%3D1%26asc%3D20892%26meid%3D5994270665998511118%26pid%3D100005%26prg%3D9312%26rk%3D1%26rkt%3D6%26sd%3D251088281787&rt=nc
Cheers Grum.
Quote from: Hoppy on 2014.04.05, 14:44:57
@ all,
I have attached a few shots of my initial build on stripboard. I have kept as close as possible to Akula's component values. The small stripboard with 5 x 1R2 paralleled resistors makes up R3 (on Akula's schematic) , and comes out at around 0.24R, as I did not have a 0.22R. I used an array of 10 x 3W LED chips manufactured by AVAGO, type ASMT-JW32-NWXH1 (see attached data sheet).
...
Hoppy
Great work Hoppy! I noticed you are using resistors for load. Do you need to add a diode in series? I think someone said LEDs were also used as diode in circuit?
Quote from: Grumage on 2014.04.05, 17:45:05
Dear 4Tesla.
These are the LED's I am using for a load. ...
Cheers Grum.
Thanks Grum!
Quote from: 4Tesla on 2014.04.05, 17:49:01
Great work Hoppy! I noticed you are using resistors for load. Do you need to add a diode in series? I think someone said LEDs were also used as diode in circuit?
No, the resistors are for making up R1. See my earlier post and attachment that shows the data sheet for the LED's I used.
Quote from: Hoppy on 2014.04.05, 18:21:57
No, the resistors are for making up R1. See my earlier post and attachment that shows the data sheet for the LED's I used.
I see, thanks for clarifying for me.
Quote from: Grumage on 2014.04.05, 17:45:05
Dear 4Tesla.
These are the LED's I am using for a load. I have 5 in series times two. I have then put them in parallel to provide a 10W load.
Cheers Grum.
Hi Grum,
I found those on US ebay:
http://www.ebay.com/itm/New-50pcs-High-Power-1W-Warm-White-LED-Lamp-Bead-Chip-Light-90-100LM-3-2-3-4V-/251435918634?pt=LH_DefaultDomain_0&hash=item3a8abfa52a (http://www.ebay.com/itm/New-50pcs-High-Power-1W-Warm-White-LED-Lamp-Bead-Chip-Light-90-100LM-3-2-3-4V-/251435918634?pt=LH_DefaultDomain_0&hash=item3a8abfa52a)
Nice to have a lot of 50.. just in case some fail. :)
Quote from: Hoppy on 2014.04.05, 16:23:38
Shot taken across R5 with probe on top of resistor. Supply voltage 6.8V with LED's very bright and the mosfet getting very hot!!
Quote from: Grumage on 2014.04.05, 17:24:26
I don't think there is any doubt about that negative going pulse !! :)
Indeed, that seems to indicate, that the TL494 feedback loop makes sense after all. I had doubts about that ...and others had too.
Could you scope pin 3 while triggering on pin 5 to check if the TL494 reacts to those pulses at R5/R7.
@Grum
Did you try scoping pin 3 and 11 while triggering the scope on pin 5 ?
Quote from: Grumage on 2014.04.05, 17:24:26
Good evening Hoppy.
I don't think there is any doubt about that negative going pulse !! :)
I have a couple of questions.
1. Is your transformer wound 15 one end and 45 the other, split between ?
2. What is your duty cycle ?
I am still testing with a simple Mosfet PWM with less than 10% duty and current measured with DVM from 12VDC 7 AH LAB. With this set up I rarely reach even 1 Amp !!
Cheers Grum.
Evening Grum,
I think the spike is due to my resistor R5 being slightly inductive at 0.4uH.
My winds are as proposed by Wattsup, as I don't see how Akula's L2 could possibly have 45 turns from looking at the photo inset on his circuit schematic. L1 is 3 layers of 5 turns. L2 is 4 layers of 7 turns, with a full depth split between L1 and L2.
Duty cycle is 35% at 26KHz, falling to 16.2% at 138KHz, as measured from mosfet drain to source.
Regards
Hoppy
Quote from: verpies on 2014.04.05, 20:47:31
@Grum
Did you try scoping pin 3 and 11 while triggering the scope on pin 5 ?
Good evening Verpies.
I am still waiting on parts to arrive for the "Proper" drive board. In the mean time I have been "Transformer bashing" with a simple Mosfet PWM. In the hope of finding that elusive anomaly !! :)
I had a little accident with my cores, be warned !! Make sure they are firmly clamped together whilst testing as when loose they tend to "clack together" and break !! Luckily "THE BOSS" allowed me to buy another set so I now have two transformers ( one super glued) !! and wound conventionally. 15 T one end and 45 turn the other. My original wound the T-1000 way.
Cheers Grum.
Further investigation into the 'spike' reveals that it disappears at around 31KHz and increases fairly linearly up to max frequency.
Quote from: Grumage on 2014.04.05, 21:35:22
I had a little accident with my cores, be warned !! Make sure they are firmly clamped together whilst testing as when loose they tend to "clack together" and break !!
I think Peterae has arrived at the conclusion, that "the anomaly" in this transformer depends on these cores "clacking" together, ...just not so hard that they break.
His observations were quite deft and I don't see any logical errors in his reasoning.
So Verpies, would this be what Akula means be "resonance in resonance" ?
You have a carrier frequency tuned for the transformer and then a feedback frequency that targets the NMR/NAR effect? Would that be the correct way to look at this circuit?
Quote from: Matt Watts on 2014.04.05, 22:38:47
So Verpies, would this be what Akula means be "resonance in resonance" ?
Maybe. If that is the case then it would mean "acoustic resonance in LC resonance".
Quote from: Matt Watts on 2014.04.05, 22:38:47
You have a carrier frequency tuned for the transformer and then a feedback frequency that targets the NMR/NAR effect? Would that be the correct way to look at this circuit?
The NMR frequency will depend on the material being resonated and on LF polarizing magnetic flux density (B) during the T
1 time. A perpendicular orientation of the HF resonating magnetic field is also a prerequisite.
With NAR, the acoustic wavelength in relation to the core's size also become a factor but this additional complication is offset by the lack of the detrimental skin effect.
Further investigation into the feedback loop reveals the attached four scope shots all taken with DC coupled (x10 probe) on mosfet gate with signal referenced to ground. The first shot 07 shows the waveform with feedback connected at 47KHz. The second shot 08 shows a reduced amplitude waveform with feedback connected at 68KHz. The third shot 09 shows the waveform with feedback connected at 90KHz. The fourth shot 10 shows no waveform when feedback removed. The transition from waveform to no waveform is sharp and has the effect of switching the output LED's on and off. Disconnecting the feedback below 90KHz does have the effect of reducing the signal amplitude and LED light level slightly. This reduction increases as the frequency rises towards the 90KHz switching point. Supply voltage 6.8V.
Hoppy
The N-Ch MOSFET gets stuck in the ON state? (100% DUC)
Even without the scopeshot on pin 3, it is evident that the TL494 reacts to the feedback signal and the whole feedback loop is not bogus.
Quote from: verpies on 2014.04.06, 09:44:19
The N-Ch MOSFET gets stuck in the ON state? (100% DUC)
Even without the scopeshot on pin 3, it is evident that the TL494 reacts to the feedback signal and the whole feedback loop is not bogus.
Yes, it would appear to be intended as a form of current limiting which makes sense given an LED load. However, I still cannot see how this device operates on 12V because I connected a 12V battery with a clamp meter to measure current and fleetingly recorded at least a 12A draw without feedback and a steady7.5A (at lowest frequency setting) with feedback and LED's bright. That was with a 0.68R big power resistor in place of the 0.24R (5 x paralleled 2W 1.2R), which all very quickly smoked with and without feedback! Also, as Grum has reported, I heard a loud 'clack' from the ferrite core on connection to the battery.
If you start gluing the core, only glue to the 2 outside parts of the E Core, leave the middle faces just touching with no glue, the outside glue will hold it together OK hopefully.
The other way is to use elastic bands around the outside of the core to force the 2 halves together, at least this way there's space for inserting spacers etc.
Quote from: Peterae on 2014.04.06, 10:51:24
If you start gluing the core, only glue to the 2 outside parts of the E Core, leave the middle faces just touching with no glue, the outside glue will hold it together OK hopefully.
The other way is to use elastic bands around the outside of the core to force the 2 halves together, at least this way there's space for inserting spacers etc.
It might also be an idea to vary the DC bias current by adjustment of R1 with and without spacers to see if we can reach a point where the core can 'chatter'. I have a feeling that Akula could be testing us and having a laugh when we create smoke and crack ferrites!!!
Quote from: Hoppy on 2014.04.06, 11:02:57
I have a feeling that Akula could be testing us and having a laugh when we create smoke and crack ferrites!!!
There is always a risk of that if he monitors this forum. Is he able to?
IMO clacking/chattering core should be free to do so. Dampening materials will attenuate that ...especially in the gap.
@all
I decided to remove this post due to lack of interest. Just do what you want to do. I will make the circuit when it arrives and do this on my own. It will be much simpler then trying to explains things here.
wattsup
Grumage
where are you buying your core & Bobbins from in the UK?
Thanks
Peter
Quote from: Peterae on 2014.04.06, 14:33:02
Grumage
where are you buying your core & Bobbins from in the UK?
Thanks
Peter
Dear Peter and all.
Both Itsu and I got our bobbins and cores from here.
http://powermagnetics.co.uk/calculator. From this handy inductance calculator you can go to the shop. If you order in the AM it will usually be with you the following day (UK) for Itsu 2 days !!
We bought these. http://powermagnetics.co.uk/pace-components/ferrite-cores/etd-cores/etd49-cf139 And the bobbin. http://powermagnetics.co.uk/pace-components/coil-formers-bobbins/bobbins-for-etd-cores/etd49-90794-20p
Hope this is of help ??
Cheers Grum.
Quote from: Grumage on 2014.04.06, 16:05:09
Dear Peter and all.
Both Itsu and I got our bobbins and cores from here.
http://powermagnetics.co.uk/calculator. From this handy inductance calculator you can go to the shop. If you order in the AM it will usually be with you the following day (UK) for Itsu 2 days !!
We bought these. http://powermagnetics.co.uk/pace-components/ferrite-cores/etd-cores/etd49-cf139 And the bobbin. http://powermagnetics.co.uk/pace-components/coil-formers-bobbins/bobbins-for-etd-cores/etd49-90794-20p
Hope this is of help ??
Cheers Grum.
Grum,
Only EPCOS ETD29/16/10 bobbin and cores will fit the boards that I have designed.
GL.
To all in chat !!
OK so why is aKula not using standard preferred resistor values, the only time I've seen that is when either the resistor value is really critical for correct operation, or you get a theoretical non practical person designing the thing.
I think we know he is indeed a practical guy so then why R8 & R9 are 4.02K, i did notice that this value is orderable, also
R10 6.98K again totally weird value to use.
and then we have a 4K7 pot, that's an old world pot value, normally these days they make 5K
Quote from: Peterae on 2014.04.06, 20:23:45
OK so why is aKula not using standard preferred resistor values, the only time I've seen that is when
either the resistor value is really critical for correct operation, or
you get a theoretical non practical person designing the thing.
Yes, nonstandard resistor values usually mean critical or theoretical reasons.
...but are those values really so non-standard for 1% resistors ?
Quote from: Peterae on 2014.04.06, 20:23:45
OK so why is aKula not using standard preferred resistor values, the only time I've seen that is when either the resistor value is really critical for correct operation, or you get a theoretical non practical person designing the thing.
I think we know he is indeed a practical guy so then why R8 & R9 are 4.02K, i did notice that this value is orderable, also
R10 6.98K again totally weird value to use.
and then we have a 4K7 pot, that's an old world pot value, normally these days they make 5K
Peter,
Akyla did use 4K, 5K, 7K and 7R in the drawing. Looking at his board it seems that he used 5% or 10% resistors.
In my part design I did use 1% 4,02K, 5,1K and 6,98K and 6,8R. I think the values I have chosen is well inside the
5% and 10% tolerances of the resistors Akyla did use.
So if we use %5 resistors for the Akyla values we get:
4K can be from 3,80K to 4,20K
5K can be from 4,75K to 5,25K
7K can be from 6,65K to 7,35K
7R can be from 6,65R to 7,35R
For 10% resistors we get even more spread.
GL.
Quote from: Grumage on 2014.04.06, 18:42:51
To all in chat !!
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12541;image) (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12540)
What are the scope probes positions for this scopeshot?
Is Ch1 the voltage on pin 3 ?
Quote from: Grumage on 2014.04.06, 16:05:09
Dear Peter and all.
Both Itsu and I got our bobbins and cores from here.
http://powermagnetics.co.uk/calculator. From this handy inductance calculator you can go to the shop. If you order in the AM it will usually be with you the following day (UK) for Itsu 2 days !!
We bought these. http://powermagnetics.co.uk/pace-components/ferrite-cores/etd-cores/etd49-cf139 And the bobbin. http://powermagnetics.co.uk/pace-components/coil-formers-bobbins/bobbins-for-etd-cores/etd49-90794-20p
Hope this is of help ??
Cheers Grum.
I just got http://powermagnetics.co.uk/pace-components/ferrite-cores/etd-cores/etd49-cf129 x2 + http://powermagnetics.co.uk/pace-components/coil-formers-bobbins/bobbins-for-etd-cores/etd49-90794-20p there. Hopefully can find time to assemble 555 circuit (will see if I can make feedback as per original circuit for duty cycle) on next weekend when order will arrive. Very curious if I can find anything unusual within transformer. As in way I suggested windings there it should be interesting to see things going inside of transformer with main frequency and feedback blocking pulses... :)
Quote from: verpies on 2014.04.06, 22:38:21
What are the scope probes positions for this scopeshot?
Is Ch1 the voltage on pin 3 ?
Channel 1 - VT1-C3 (isolated ground)
Channel 2 - across C4 (isolated ground on C3)
Cheers!
Thanks GL
Actually i see he fixed those resistor values on the later circuit with the push pull pre driver transistors.
R6 = 220R was 210R
R5 = 4K7 was 4.02K
R9 = 4K7 was 4.02K
R10 = 6K8 was 6.98K
C5 = 3n3 was 3nF
but as you say if he was using 5% or 10% probably wont matter. O0
Further investigation on the feedback loop using a 0.68R power resistor has enabled me to increase the supply voltage to 8.9V without my bench supply current clamping. The feedback loop now has far less effect and I now think that this is because the mosfet gate is now being driven with a much better waveform. With R1 at the previous 0.24R, the gate voltage was around its threshold on a 6.8V supply rail, the mosfet was getting "stuck in the ON state" as Verpies suggested and which was clear from the scope shots .
I tried varying the value of R1 up to 4R and this does of course reduce running current and did result in the ferrite core 'buzzing' (no gap and ferrite faces touching) at around 3R but I could not hold this condition as the 'sweet spot' was very sharp indeed. This condition resulted at the onset of ringing in the waveform as scoped from mosfet drain to source brought about by careful adjustment of rail voltage. Its just possible that the 0.22R value for R1 was meant to be 2.2R?? This would make a lot more sense insofar as a sensible current consumption is concerned. My final build on the PCB will be with R1 at 2.2R.
Quote from: Peterae on 2014.04.07, 07:41:58
Thanks GL
Actually i see he fixed those resistor values on the later circuit with the push pull pre driver transistors.
R6 = 220R was 210R
R5 = 4K7 was 4.02K
R9 = 4K7 was 4.02K
R10 = 6K8 was 6.98K
C5 = 3n3 was 3nF
but as you say if he was using 5% or 10% probably wont matter. O0
Peter,
I did use the first circuit drawing I found, I think, attached here.
The changes I did was to use a smaller core. This because the
core I did choose is well capable of shuffling 30 Watt. And the
core from EPCOS, ETD29 can be delivered with different core
material. I have ordered N27 and N87 Ferrite cores to my bobbin,
so I can try both. The other change I did was to make the R2/C4
combo tunable. But if people do not want that then it is possible
to just solder in a fixed value resistor and capacitor.
BTW, the "Bread board" type board is "air snail mailed" to members.
The boards look like in the attachment under.
GL.
Dear All.
A question for our more erudite members !!
How does the Al value of a Ferrite change the characteristics ?? For example. Say we had a core of Al 4000 set against one of Al 2000. What would be the practical outcome ??
A little bird has just sent a frequency clue !! Try around 17 KHz. O0 With a core of Al 2000. ;)
Cheers Grum.
Nice board, next time you order from those guys, they will charge you per hole LOL ;D
Quote from: Grumage on 2014.04.07, 13:12:00
Dear All.
A question for our more erudite members !!
How does the Al value of a Ferrite change the characteristics ?? For example. Say we had a core of Al 4000 set against one of Al 2000. What would be the practical outcome ??
A little bird has just sent a frequency clue !! Try around 17 KHz. O0 With a core of Al 2000. ;)
Cheers Grum.
Hi Grum,
Basically the inductance equals AL times the number of turns squared ( L= AL*N*N). So if you have one core with AL 4000and the other with AL 2000, for the same number of turns, the first one will have double of the inductance of the second.
Cheers,
Black Bird
Quote from: Black_Bird on 2014.04.07, 15:38:08
Hi Grum,
Basically the inductance equals AL times the number of turns squared ( L= AL*N*N). So if you have one core with AL 4000and the other with AL 2000, for the same number of turns, the first one will have double of the inductance of the second.
Cheers,
Black Bird
Dear Black Bird.
Many thanks for that info. O0
Cheers Grum.
Quote from: Black_Bird on 2014.04.07, 15:38:08
Basically the inductance equals AL times the number of turns squared ( L= AL*N*N).
Yes, the A
L value of a core tells you how many Henries you will get for every turn ...squared. ( L=A
L*N
2 )
NOTE: The A
L value will decrease if you gap or cut the core ...or saturate it or bring a magnet near it, or cool it.
Heating, squeezing it with a vise (or with sound) can increase the A
L value.
P.S.
The A
L value of some cores changes even in the presence of external electric fields. (see: IEEE Transactions on Magnetics, Vol. 41, No. 10 from October 2005 by Konrad & Brudny). I think Osakosa wrote, that old Russian TV yoke ferrites were of this type.
Quote from: verpies on 2014.04.07, 20:00:57
Yes, the AL value of a core tells you how many Henries you will get for every turn ...squared. ( L=AL*N2 )
NOTE: The AL value will decrease if you gap or cut the core ...or saturate it or bring a magnet near it, or cool it.
Heating, squeezing it with a vise (or with sound) can increase the AL value.
P.S.
The AL value of some cores changes even in the presence of external electric fields. (see: IEEE Transactions on Magnetics, Vol. 41, No. 10 from October 2005 by Konrad & Brudny). I think Osakosa wrote, that old Russian TV yoke ferrites were of this type.
Dear Verpies.
As always, a thorough explanation. O0
Please read this sentence from Hoppy's post earlier today.
Quote.
I tried varying the value of R1 up to 4R and this does of course reduce running current and did result in the ferrite core 'buzzing' (no gap and ferrite faces touching) at around 3R but I could not hold this condition as the 'sweet spot' was very sharp indeed. This condition resulted at the onset of ringing in the waveform as scoped from mosfet drain to source brought about by careful adjustment of rail voltage.
Un quote.
I am finding that by rapidly altering (back and forth) my frequency pot on my PWM I seem to be able to lengthen my ringdown on switch off ?? This may of course be my imagination !! :) But to me something in Hoppy's statement above might just be key ??
At certain frequencies my signal sort of hiccups down !! As if it is sort of rebooting !! I don't really know quite how to describe it !! :)
Your thoughts ??
Cheers Grum.
Interesting observations, Grum:
QuoteI tried varying the value of R1 up to 4R and this does of course reduce running current and did result in the ferrite core 'buzzing' (no gap and ferrite faces touching) at around 3R but I could not hold this condition as the 'sweet spot' was very sharp indeed. This condition resulted at the onset of ringing in the waveform as scoped from mosfet drain to source brought about by careful adjustment of rail voltage.
Un quote.
I am finding that by rapidly altering (back and forth) my frequency pot on my PWM I seem to be able to lengthen my ringdown on switch off ?? This may of course be my imagination !! Smiley But to me something in Hoppy's statement above might just be key ??
At certain frequencies my signal sort of hiccups down !! As if it is sort of rebooting !!
Do you have any output/input power data (to go along with these intriguing observations)?
I will be interested to hear from anyone when they build Groundloops PCB.
I have suspicions that Akula quickly moved to a new version with push pull driving of the fet base.
If anyone is seeing drive problems regarding the fet, i would start looking at modding GL's PCB to use a Fet driver which we know Akula used in his latest version.
If someone need help doing this i will work out a mod, a little bit of vero with the driver chip soldered to the fet gate should do the job.
I may yet be wrong about this but the circuit that the PCB is built on has component value corrections in the later versions that then also use Push/Pull indicating to me that he never bothered updating the earlier version circuit with these value changes and instead moved straight onto a new version, and then we have the version in the video using a fet driver.
Quote from: Peterae on 2014.04.08, 10:03:54
I will be interested to hear from anyone when they build Groundloops PCB.
I have suspicions that Akula quickly moved to a new version with push pull driving of the fet base.
If anyone is seeing drive problems regarding the fet, i would start looking at modding GL's PCB to use a Fet driver which we know Akula used in his latest version.
If someone need help doing this i will work out a mod, a little bit of vero with the driver chip soldered to the fet gate should do the job.
I may yet be wrong about this but the circuit that the PCB is built on has component value corrections in the later versions that then also use Push/Pull indicating to me that he never bothered updating the earlier version circuit with these value changes and instead moved straight onto a new version, and then we have the version in the video using a fet driver.
Most likely the case Peterae!
I would go so far as to advise builders not to power up the original circuit as laid on the PCB, unless they have a current limited bench supply.
Hi Hoppy
So it is looking like your excessive current is because the Fet is not being driven very well, thanks for that O0
Dam i have been trying to find the video someone posted, which showed the Akula circuit using a fet driver, anyone got a link again for it.?
EDIT
I found it, the post was by Itsu
http://www.overunityresearch.com/index.php?topic=2358.msg36954#msg36954
https://www.youtube.com/watch?v=EwCS15pRtH0&list=PLC7684829E98CAD74
I modified my circuit to the updated circuit that shows a transistor push-pull mosfet driver (BC546 / 556 comp pair) and it now works fine without excessive heating. The feedback loop when broken will cause a cessation of output at all frequency settings. The 'stop' switch also works to cease output when the switch is closed. I think the feedback loop was likely intended as a safety switch to turn-off the PWM, should the output become open circuit. I have yet to observe anything approaching a self-running condition. However, this setup does appear to be quite efficient and a very good light output can be had from well under a 12V supply rail. It is possible that I cleared a track fault when I implemented the push-push mosfet driver, which may have caused an excessive running current, so try the original circuit before implementing the modification to enable a comparison to be made.
Nice work Hoppy Thanks
Is this your vero build or GL's PCB?
Quote from: Peterae on 2014.04.08, 17:35:41
Nice work Hoppy Thanks
Is this your vero build or GL's PCB?
Vero build.
Quote from: Hoppy on 2014.04.08, 17:16:48
The feedback loop when broken will cause a cessation of output at all frequency settings.
Why?
The TL494 outputs pulses on pins 10/11 even when the error amplifiers/comparators are not activated (pin 3 low).
Activation results in shortening of output pulses.
QuoteVero build.
O0
I've just changed my mind about the fet driver being the lastest version of the Akula device.
It has a few potential problems, he has not included the needed decoupling caps for the fet driver and these devices really do need that, this fet driver can drive with 9Amps and he is only using a 100uF on the output of the 7812, i think that the drive waveform in this version of the circuit was horrible.
The other reason i don't think this was a worker is that again he used the same odd values, which puts the push/pull transistor version as the latest and functional circuit.
I'm still pondering as to why Akula fitted the 'Stop' switch. Although this works OK, the current consumption stays much the same and unless the ferrite E-cores are held / glued together they clonk loudly when the switch is operated. Any thoughts on this?
Quote from: Grumage on 2014.04.07, 20:43:25
I am finding that by rapidly altering (back and forth) my frequency pot on my PWM I seem to be able to lengthen my ringdown on switch off ??
Where is this "ringdown" measured?
Are you referring to the switch-off of the MOSFET/IGBT or of some mechanical switch or power supply?
Quote from: Hoppy on 2014.04.08, 19:03:30
I'm still pondering as to why Akula fitted the 'Stop' switch. Although this works OK, the current consumption stays much the same and unless the ferrite E-cores are held / glued together they clonk loudly when the switch is operated. Any thoughts on this?
Good day Hoppy;
You mentioned that the stop switch funcitons OK as it kills the output when 'closed'.
and above; the E-cores vibrate and 'clonk' together when the switch is 'closed'.
The clonking sounds like a big clue, just like the Russion forum member mentioned that after about 4 days his core shattered as it was self running...........
Maybe could it be just the reverse of this and that when the stop switch is 'closed' it is recycling energy back to the circuit instead of (or) at the same time diverting energy from the LEDs and this is why they turn off?
Maybe the Stop Switch when open is when the Self running aspect of the circuit ceases to be, and when the Stop switch is closed, it is in self running mode?
Maybe the coils need to be adjusted so that the output from L1 & L2 is 'optimized' so it can feed back just a part of the current and NOT divert all the energy from the LED array, so it can stay lit.
Just a few thoughts.
take care, peace.
lost_bro
Quote from: Hoppy on 2014.04.08, 19:03:30
I'm still pondering as to why Akula fitted the 'Stop' switch. Although this works OK, the current consumption stays much the same and unless the ferrite E-cores are held / glued together they clonk loudly when the switch is operated. Any thoughts on this?
Seems to me you have something you can work with. Probably best if you don't allow those six surface faces to get damaged. Maybe use a thin layer of tape and make some notes in regards to signals. My personal preference would be to place a tiny dot of silicone on the center face of one side and let it cure. This should remain flexible enough to allow for some movement of the cores without damage. Hopefully the bobbin is tight enough it doesn't allow the cores to wobble side-to-side--you may want to add some thin shims.
To be completely honest, having not seen up-close an Akula device running, I'm not sure what exactly we need in that gap, if anything. The list verpies presented earlier enumerates the variables we will have to deal with.
Quote from: verpies on 2014.04.08, 19:45:10
Where is this "ringdown" measured?
Are you referring to the switch-off of the MOSFET/IGBT or of some mechanical switch or power supply?
Dear Verpies.
I have my (mains isolated) scope probe on the junction of C4 R2 with ref to circuit ground. And I am very basically dabbing the Pos from the 12 V battery to charge C11.
Dear Matt.
I have found that a single piece of paper between just one outer leg of the cores creates the best ring !!
Hope this is of help?
Cheers Grum.
Quote from: lost_bro on 2014.04.08, 19:50:45
Maybe the Stop Switch when open is when the Self running aspect of the circuit ceases to be, and when the Stop switch is closed, it is in self running mode?
Maybe the coils need to be adjusted so that the output from L1 & L2 is 'optimized' so it can feed back just a part of the current and NOT divert all the energy from the LED array, so it can stay lit.
Just a few thoughts.
take care, peace.
lost_bro
Hi lost_bro,
If there is any self-running to be had, then its most likely to be a case of getting the right core / winding conditions. Something that could confuse experimenters is that LED's have persistence in illumination and given the fairly large capacitor C3, my LED's stay illuminated at a very low slowly diminishing level from the cap's charge for a couple of minutes and can look quite bright in a darkened room.
My approach would be to run without the output led's as the extra burden and just note power supply current as you tweak the gap and other parameters.
Switching losses alone should be enough of a burden to overcome, when that is met additional loads can be added.
Since the output of the switcher is looped back to the input, you should begin to see a drop in current from the driving power source as you get to the sweet spot. If there is indeed extra energy to be tapped, the power supply current should drop to zero and the voltage will begin to rise.
LED's act as Zeners and if you are on the upper end of the knee it is difficult to discern efficiency improvements of the circuit because brightness does not change much on the high end of the knee.
You can switch the LED's back in when you get to the sweet spot, they will then act as clippers, soak up the extra energy and tame the voltage from soaring.
Dear All.
More of the same ??
https://www.youtube.com/watch?v=0_8OlH4kkKM
https://www.youtube.com/watch?v=juwm_W5-qAM
https://www.youtube.com/watch?v=gWjs0-WCMzs
WHY CAN'T WE DO THIS ?? IS THERE SOMETHING IN THE FERRITE ?? ;D
Cheers Grum.
Grum,
Those videos are for the Melnichenko device. I have a decent collection of that information if anyone is interested.
Quote from: Grumage on 2014.04.09, 09:41:03
Dear All.
More of the same ??
https://www.youtube.com/watch?v=0_8OlH4kkKM
https://www.youtube.com/watch?v=juwm_W5-qAM
https://www.youtube.com/watch?v=gWjs0-WCMzs
WHY CAN'T WE DO THIS ?? IS THERE SOMETHING IN THE FERRITE ?? ;D
Cheers Grum.
Grum,
Without a detailed circuit diagram, a PCB layout and a exact core type and winding, then
those videos are just nice videos on YouTube.
Ahh, I see that you have posted detailed information now. :-)
GL.
It looks like the earlier Akula version with the pot core we were going to try and build a few months back, we have diagrams and info on the core, basically the device works the same as the one we are working on right now but uses a different drive circuit, it also uses alu foil in the pot core. O0
Peter, could u please post a link/schematic of the device you are speaking about?
At the time NMR was controversial, the device above was discussed in a Private Group.
here's the link but it is a Private Group, so to be able to see it you need to PM me so i can set you up permission.
http://www.overunityresearch.com/index.php?topic=2252.msg34357#msg34357
but
You need to understand that it may not be a good idea building a device in public that produces Beta radiation C.C
when that time happens then it will need to go Private until fully documented and ready for release, and no it's not because we are covering up the device, it's to protect DIYers from irradiating themselves or worse.
Quote from: Matt Watts on 2014.04.08, 20:04:55
Maybe use a thin layer of tape and make some notes in regards to signals. My personal preference would be to place a tiny dot of silicone on the center face of one side and let it cure.
If the "magic" in the transformer depends on an acoustic effect then placing acoustically dampening materials between the core halves would attenuate this effect.
For example, if the piezo element was glued to the ferrite rod with dampening silicone, then the experiment shown in
this video (http://youtu.be/qauZ4WBwAOM?t=1m16s) would be a failure.
Vinyl tape and silicone rubber are very dampening. Paper is not so much and
sheet mica (http://en.wikipedia.org/wiki/Mica#Sheet_mica) dampens acoustic vibrations very little.
Quote from: lost_bro on 2014.04.08, 19:50:45
Maybe could it be just the reverse of this and that when the stop switch is 'closed' it is recycling energy back to the circuit instead of (or) at the same time diverting energy from the LEDs and this is why they turn off?
Maybe the Stop Switch when open is when the Self running aspect of the circuit ceases to be, and when the Stop switch is closed, it is in self running mode?
I'm sure that was the intention of this circuit's designer. In unconventional operation of this circuit (when some "magic" is happening in the transformer) the closed Stop switch recycles energy back to the circuit.
However, in conventional operation, the function of the Stop switch is just the opposite - just like Hoppy is describing.
P.S.
Beware that in Russian language an "open" switch conducts electric current, but in English language it does not!
Quote from: Grumage on 2014.04.08, 20:16:55
I have my (mains isolated) scope probe on the junction of C4 R2 with ref to circuit ground. And I am very basically dabbing the Pos from the 12 V battery to charge C11.
...but there are two junctions of C4 & R2, which one do you mean?
Sorry to be such a PITA but I hate to assUme.
Quote from: Grumage on 2014.04.08, 20:16:55
Quote from: verpies on 2014.04.08, 19:45:10
Are you referring to the switch-off of the MOSFET/IGBT or of some mechanical switch or power supply?
And I am very basically dabbing the Pos from the 12 V battery to charge C11.
Oh, so you must mean the "switch-off" of the power supply.
2 new Akula videos posted by him today:
https://www.youtube.com/watch?v=45VucygQK80
https://www.youtube.com/watch?v=3ZkeJoTNDTw
So who are we going to force to take Russian language lessons ;D
If anyone knows Akula please Send him an English translation dictionary ;D
Quote from: Peterae on 2014.04.09, 20:14:20
.............
So who are we going to force to take Russian language lessons ;D
...........
That is what I am doing, taking Russian language lessons.
If you do not intend to speak Russian and only to more or less understand what it is about,
technically-wise, it is, IMHO, not an un-reachable goal that can be achieved in some months.
Of course, If English is your only language it could be more difficult.
If you have some smattering of other languages this could help.
Привет,
Thanks++ to Matt Watts for pointing out this gnli.pdf file.
A Skeptic person would state:
No way! 'OU' (a self sustaining device, for ex) is not possible...The official equations are right. Period.
-------------
A 'Believer' would assert, for ex.:
The official equations are wrong or incomplete. The Aether exists (and must be included in the equations).
Einstein was wrong. Period.
--------------
Some persons would declare: yes, the official equations are right. Aether might exists but is not a necessity.
But, we do not take the time into account in these equations. No time in the Coulomb's law, for ex.
We assume that the 'exchanges' are made at the speed of light or near this speed, but, in some process, it is some orders of magnitude less.
You might consult 'Claus W. Turtur'.
------------
Other persons would point out that: of course, if we only use half of the equations, assuming that 'C', 'L' and 'M(ass)' are constants, we are not out of the woods 'OU'-wise.
Considering that 'C', 'L' and 'M(ass)' are *not* constant leads to very interesting results.
You also might consult 'Ronald D. Pearson' and 'Ove Tedenstig' who, both, are not assuming that 'M(ass)' has not to be constant.
-----------
You too, can derive d(L.i)/dt and d(C.v)/dt considering that 'L' and 'C' are not constant.
I tried this, but I'm not a mathematician. I got strange results. Not sure of them.
-----------------------------
Back to gnli.pdf
You must ( :)) consult: http://jnaudin.free.fr/html/tep62par.htm (http://jnaudin.free.fr/html/tep62par.htm)
last update on 10/12/97 :P
------------------------------
2 pictures:
The initial gnli.pdf CCT.
(http://freenrg.info/4OUR_F/Ggnli-Init.jpg)
My interpretation. I can be wrong.
(http://freenrg.info/4OUR_F/Ggnli-Rew2.jpg)
Or this one attached.
The Akula 60 Watt circuit T-1000 posted shows the timing to be somewhat different though and it includes an output winding.
Quote from: Grumage on 2014.04.09, 09:41:03
IS THERE SOMETHING IN THE FERRITE ??
Barium maybe or Cobalt?
Sure looks like Akula is tuning his impulse to setup some sort of momentum that continues on after the impulse signal drops.
Salve,
@Matt Watts,
Very interesting CCT ???
Unfortunately, I have not enough neurons to understand it. :-X
--------
Should I dare to submit this picture?
It is not really 'parametric' as in the gnli.pdf.
I am just wondering whether this 'bucking coil' configuration might be more simple and efficient?
(http://freenrg.info/4OUR_F/JLN_Bif-Vs-Buck-m2.jpg)
Just my 2 cents.
------------------
I am carrying on some experiments with some off-the-shelf and also some diyed coils wrapped with
some off-the-kitchen aluminium foils. Truly amazing results, IMHO.
I wish I had a good 4 channels scope to be able to prove that this (fictitious?)
'displacement currents' can do something that might be considered as 'OU'.
unfortunately, nobody seems interested in kitchen aluminium foil.
However it is an essential material for conserving sandwiches. Is it not ???
Quote from: verpies on 2014.04.09, 14:13:14
I'm sure that was the intention of this circuit's designer. In unconventional operation of this circuit (when some "magic" is happening in the transformer) the closed Stop switch recycles energy back to the circuit.
However, in conventional operation, the function of the Stop switch is just the opposite - just like Hoppy is describing.
P.S.
Beware that in Russian language an "open" switch conducts electric current, but in English language it does not!
Why did Akula provide a stop switch when opening (breaking) the feedback loop will achieve the same result? Is the real function of this switch to pulse L2 using a second oscillator??
Quote from: Matt Watts on 2014.04.09, 23:30:04
Or this one attached.
The Akula 60 Watt circuit T-1000 posted shows the timing to be somewhat different though and it includes an output winding.
The 1/3 coil shorting should be on right time so BEMF will get mixed with magnetic pulse. What happens next - you need to find out ;)
Quote from: Hoppy on 2014.04.10, 06:52:22
Why did Akula provide a stop switch when opening (breaking) the feedback loop will achieve the same result? Is the real function of this switch to pulse L2 using a second oscillator??
As far as I know it is power off switch there so you can stop it from self running...
Quote from: T-1000 on 2014.04.10, 10:55:05
As far as I know it is power off switch there so you can stop it from self running...
Thanks for your opinion on the switch function. Now I just need to get my replication self-running so that I can test this function. O0
You guys in the USA should be getting your PCB's now it's been a week and they said 5 days when posting. O0
I believe this is relative to Akula's devices
http://en.wikipedia.org/wiki/Amplifier
http://www.vk2zay.net/article/181
Electron current flow :D
Quote from: Dave45 on 2014.04.11, 11:09:44
I believe this is relative to Akula's devices
Far from it
Dear All.
The final components arrived today, so being of impetuous nature I put the whole device together !!
Now I am lost !! The thing seems to have a mind of it's own !! I started with Hoppy's suggestion using the current limit on my bench PSU. Bringing the voltage up quietly. At around 7.8 V I am getting a pulsating effect. The LED's shine like the sun pulling nearly 1.7 amps and then suddenly the current drops back to around 57 mA. Strangely moving my hand around the circuit can trigger this response.
I could really do with some advice now !! With the transformer fully disconnected the device has a standing current of 29/30 mA at 12 V. I placed my scope probe on the Mosfet gate resistor, with ref to ground. I am seeing no signal here. ??
With my transformer connected the maximum voltage I can apply without serious overload is 7.8 V . I can see the pulses being applied to L1 with ref to ground. And can vary the frequency from around 30 KHz to 123Khz. I can's seem to alter the duty cycle though !!
Oh and BTW does anyone know what the FHZ pot is supposed to do ?? Id does not seem to work !!
Sorry about the long post, but I am at a loss here !!
Cheers Grum.
Quote from: Grumage on 2014.04.11, 18:34:19
Dear All.
The final components arrived today, so being of impetuous nature I put the whole device together !!
Now I am lost !! The thing seems to have a mind of it's own !! I started with Hoppy's suggestion using the current limit on my bench PSU. Bringing the voltage up quietly. At around 7.8 V I am getting a pulsating effect. The LED's shine like the sun pulling nearly 1.7 amps and then suddenly the current drops back to around 57 mA. Strangely moving my hand around the circuit can trigger this response.
I could really do with some advice now !! With the transformer fully disconnected the device has a standing current of 29/30 mA at 12 V. I placed my scope probe on the Mosfet gate resistor, with ref to ground. I am seeing no signal here. ??
Cheers Grum.
Good day Grum
PLease see the attachment.....
The RED line will bias the MOSFET GATE directly without signal from TL494 or driver....
take care, peace
lost_bro
Grum is this with GroundLoops PCB, it's best to mention this guys,because what ever the solution is maybe applicable to a lot of people O0
The next question is going to be who is first to get the single transistor fet driver, or how we can get it going, because at some point we will need to answer the question of whether we need to mod it for push/pull drive of the fet.
We need to develop a test procedure, a step by step list of things to do, to see the fet is being driven properly without the transformer & load drawing loads of current.
Maybe something like using a bulb in place of the primary and maybe using a DC supply to provide feedback, someone more acquainted with the theory of operation should be able to work out the best practise.
Quote from: Peterae on 2014.04.11, 20:39:15
Grum is this with GroundLoops PCB, it's best to mention this guys,because what ever the solution is maybe applicable to a lot of people O0
The next question is going to be who is first to get the single transistor fet driver, or how we can get it going, because at some point we will need to answer the question of whether we need to mod it for push/pull drive of the fet.
We need to develop a test procedure, a step by step list of things to do, to see the fet is being driven properly without the transformer & load drawing loads of current.
Maybe something like using a bulb in place of the primary and maybe using a DC supply to provide feedback, someone more acquainted with the theory of operation should be able to work out the best practise.
Dear Peter.
Yes it is the NON breadboard type. I have been in chat with Groundloop he is suggesting I may have a dry joint or two !! :)
The single transistor is driving the Mosfet OK but there is some instability (probably dry joint) The LED's are the brightest I have ever seen them !! 8) A must !!
Your suggestion of a dummy load in place of the transformer seems to be a good idea. O0
Cheers Grum.
Well if you get stuck take some high res pictures under a light and we remotely check things for you O0
Quote from: lost_bro on 2014.04.11, 19:31:16
Good day Grum
PLease see the attachment.....
The RED line will bias the MOSFET GATE directly without signal from TL494 or driver....
take care, peace
lost_bro
This is a circuit path that I do not see in the original Akula circuit schematic.
@All,
The attached drawing is what I used when I designed the board. This drawing was included in the zip archive that I posted.
The TL494 built in transistor is used as a pull down device. So then you actually need a current path up to the positive rail.
Let us start with the TL494 built in transistor being ON, conducting current. The MOSFET will be off because the gate is
drained to ground. The driving transistor (2SC945) will be OFF because the base is low voltage. Now we switch OFF
the TL494 built in transistor. The voltage of the gate of the MOSFET will go high and the MOSFET will start to
switch ON current. But the base of the driving transistor (2SC945) will also go high voltage, so this transistor will also
switch on carrying current. This will make the MOSFET switch ON much faster because of higher gate loading current.
So the sole purpose of the driving transistor (2SC945) is to get a faster MOSFET switch ON time.
So, yes, lost_bro is correct, a faulty TL494 or a removed TL494 and the MOSFET will stay ON conducting current to the max.
Grumage asked what the PHZ pot-meter is doing in MY design of the circuit. This pot-meter will allow the user to
tune the time constant of the Akyla R2/C4. I do not know WHY Akyla did include those components, but IF
this is important then it is better to use some tunable components here. That was my thought. In MY design,
the variable capacitor is for fine tuning and the variable resistor is for large tuning.
Another issue, the Akyla board has ground planes. This is a lot of copper and very good heat sink. So when you
solder the GND connections, then use more heat and make sure you really DID solder the pads. For most of the
solder pads, a 18 Watt solder Iron is enough. But for the GND solder pads, I recommend a larger solder Iron of 40 Watt.
Also, if you use the recommended heat sink on the BD139 transistor, then make sure that the heat sink is lifted a little
up from the board before you solder the heat sink to the board. This because of small distance from the heat sink to
the transistor pads. The same problem are on the boards from Peter, because he did not download the newest version
of the files.
I also recommend using a variable voltage, current limited power supply, when you start testing on this board. Start low
voltage low current.
Added: I have received one board from Peter, thanks Peter, and his board are identical to mine, except that on my board
there is gold plating on all pads. On Peters board there is standard tin plating on the pads.
GL.
Quote from: Grumage on 2014.04.11, 20:57:04
Your suggestion of a dummy load in place of the transformer seems to be a good idea. O0
That is a different idea.
e.g. When the MOSFET turns on for 100µs, the current through the transformer will
gradually increase to 300mA, and with a dummy load (a light bulb) the current will immediately increase to 1A (or 10A ...depending on the light bulb).
Quote from: Grumage on 2014.04.11, 18:34:19
Oh and BTW does anyone know what the FHZ pot is supposed to do ?? Id does not seem to work !!
On the original
Akula diagram (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307), the PHZ pot did not exist. It was the fixed 420kΩ resistor R2.
Normally, this pot/resistor is a part of a high pass filter or L2 spike scrubber.
Turning this pot close to 0Ω would be a bad idea as it would short the diode D2 associated with L2. That's why R20 was added in series to prevent that.
Quote from: Hoppy on 2014.04.11, 21:22:41
This is a circuit path that I do not see in the original Akula circuit schematic.
It's
there (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307), just drawn differently.
Without TL494 present, this path will indeed pull the gate of the MOSFET high, but when the TL494 is installed, pin 11 will pull it down through R7.
From the point of view of pin 11, the transistor VT2 and R18 act as an active pullup
In other words, TL494 pulls it down and VT2 pulls it up, achieving kind of a push-pull operation.
Quote from: verpies on 2014.04.12, 07:36:41
It's there (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307), just drawn differently.
Yes, I can now see it with my good glasses on! :-[
Quote from: Grumage on 2014.04.11, 18:34:19
The thing seems to have a mind of it's own !!
I guess that is good news.
An anomalous circuit can be expected to behave unpredictably.
Quote from: Grumage on 2014.04.11, 18:34:19
I started with Hoppy's suggestion using the current limit on my bench PSU. Bringing the voltage up quietly. At around 7.8 V I am getting a pulsating effect. The LED's shine like the sun pulling nearly 1.7 amps and then suddenly the current drops back to around 57 mA. Strangely moving my hand around the circuit can trigger this response.
That is strange unless you have some floating HighZ inputs.
Do you have any activity at pin 3 ?
Quote from: Grumage on 2014.04.11, 18:34:19
I could really do with some advice now !! With the transformer fully disconnected the device has a standing current of 29/30 mA at 12 V. I placed my scope probe on the Mosfet gate resistor, with ref to ground. I am seeing no signal here. ??
...and you should.
Do you have pulses at pin 11 when the transformer is fully disconnected?
If not, does removing the Q1 (VT2) and the MOSFET, significantly increase the amplitude of pulses at pin 11, when the transformer is fully disconnected?
Quote from: Grumage on 2014.04.11, 18:34:19
With my transformer connected, the maximum voltage I can apply without serious overload is 7.8 V . I can see the pulses being applied to L1 with ref to ground.
In inductors, current pulses are much more interesting than voltage pulses.
The 0.22Ω resistor can be used as a Current Sensing Resistor (CSR) if it is
not an inductive/wirewound resistor and when the nearby 100nF capacitor is removed.
Quote from: Grumage on 2014.04.11, 18:34:19
I can's seem to alter the duty cycle though !!
Because the DTY pot is mislabeled.
Its purpose is
not to vary the duty cycle directly, but to adjust the negative slope sensitivity (from -6.6V/ms to -18V/ms) to pulses appearing at the junction of the 1Ω resistor and the bank of white load LEDs.
This pot will have no effect if you don't have quick falling edges at this junction. See
this video (http://youtu.be/exIFDDgSrng?t=1m32s).
P.S.
The inductance of the 1Ω resistor R7 (R5 on
Akula's diagram (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307)) is of
PARAMOUNT IMPORTANCE to the feedback loop in this circuit.
So don't be surprised if this circuit works completely differently with an inductive/wirewound resistor than with a non-inductive one.
Dear Verpies.
Thank you for your detailed analysis. O0
After finding my old Solon soldering iron from my youth had finally " pegged out " !! Wonder whether the new Chinese one will last as long ?? ;D I went over the board and re heated some of the ground connections. Behold, no more strange behaviour !!
PIN 3
There is activity on Pin 3 . A DC level that starts to rapidly rise and fall ( in synch ) with the core vibration. I can bring this about with adjustments to both frequency and DTY. However the device current is still through the roof. Pushing over 3 A @ 8.2 Vdc !!
BTW. I am using Carbon Film resistors throughout the circuit build.
Cheers Grum.
Quote from: Grumage on 2014.04.12, 17:40:49
There is activity on Pin 3 . A DC level that starts to rapidly rise and fall ( in synch ) with the core vibration.
That means that your signal feedback loop is functioning.
Quote from: Grumage on 2014.04.12, 17:40:49
I can bring this about with adjustments to both frequency and DTY. However the device current is still through the roof. Pushing over 3 A @ 8.2 Vdc !!
The wider the pulses on pin 3, the narrower the pulses on pin 11 (and at the MOSFET).
Narrower pulses mean less average current draw from the power supply.
Quote from: Grumage on 2014.04.12, 17:40:49
Pushing over 3 A @ 8.2 Vdc !!
Show me the scopeshot of current flowing through L1. Maybe its inductance is too low due to excessive core gapping.
Quote from: Grumage on 2014.04.12, 17:40:49
BTW. I am using Carbon Film resistors throughout the circuit build.
Is the Akula persona using all non-inductive resistors, too?
QuoteIs the Akula persona using all non-inductive resistors, too?
I'm on the wrong computer for screen shots, but having just looked at the video, they are blue resistors, so it could be carbon or wirewound, but they look like a dull blue coating which would tend to tell me they are wirewound, from my experience carbon tend to be shiny, and the 2 watt resistors in the video are dull.
Best guess is they are indeed inductive.
Also worth noting was that the vero board version has 2 pre driver transistors without heat sinks driving the fet and he is using a single high watt LED security light for the load, the video is on the first post first page.
I have several blue resistors that are all metal oxide.
Quote from: Peterae on 2014.04.12, 19:55:03
...they look like a dull blue coating which would tend to tell me they are wirewound, from my experience carbon tend to be shiny, and the 2 watt resistors in the video are dull.
That might not hold in Akula's part of the world.
Quote from: Peterae on 2014.04.12, 19:55:03
Best guess is they are indeed inductive.
That would be a significant discovery
QuoteThat might not hold in Akula's part of the world.
True, although i did search google images for blue carbon and blue wirewound to see if any existed and how they looked.
How common are 2 watt carbon resistors in Russia, here they would be carbon film these days, not that that would change the situation i suppose.
He also had a huge resistor near the LED, i cant see them making a carbon or film resistor that size, although it looks unusual to me for a wire wound resistor, i cannot imagine it being anything else other than a wirewound.
Maybe someone else with a lot of practical experience would also take a look.
EDIT this is what they look like to me
I have populated the PCB supplied by Peterae and have found that if the direct DC path to the mosfet (as shown by lost_bro in post 576), by cutting the track to the mosfet (see attached) just leaving the 750R resistor biasing the drive transistor, the transistor does not switch cleanly and with the track intact the mosfet fails to switch 'OFF', thereby passing a considerable continuous current through LI as Grum has reported.
Hoppy
Quote from: Hoppy on 2014.04.12, 21:36:13
I have populated the PCB supplied by Peterae and have found that if the direct DC path to the mosfet (as shown by lost_bro in post 576), by cutting the track to the mosfet (see attached) just leaving the 750R resistor biasing the drive transistor, the transistor does not switch cleanly and with the track intact the mosfet fails to switch 'OFF', thereby passing a considerable continuous current through LI as Grum has reported.
Hoppy
Hoppy,
If you cut the track then you need to add a 10K resistor from gate on MOSFET to ground.
This because with the track cut, there is no current path for the gate to discharge to ground.
Also, do you have a o-scope image (DC) of the gate voltage to the MOSFET?
GL.
Quote from: Groundloop on 2014.04.12, 21:42:20
Hoppy,
If you cut the track then you need to add a 10K resistor from gate on MOSFET to ground.
This because with the track cut, there is no current path for the gate to discharge to ground.
Also, do you have a o-scope image (DC) of the gate voltage to the MOSFET?
GL.
Yes, I had a gate grounding resistor in place. No scope shot at the moment as TL494 output gone completely dead. I think the voltage reg got fried! Earlier on I was over-volting C10 as if the load had become OC but when checked it was not. I've changed the regulator and have the correct output voltage at the TL494 but no output. Changed the TL494 but still no output. Time for bed.
Hoppy
Quote from: Hoppy on 2014.04.12, 21:36:13
...thereby passing a considerable continuous current through LI as Grum has reported.
I talked to Grum and his high current seems to be caused by exceeding the reverse blocking voltage of his diodes.
Quote from: verpies on 2014.04.12, 18:14:19
That means that your signal feedback loop is functioning.
The wider the pulses on pin 3, the narrower the pulses on pin 11 (and at the MOSFET).
Narrower pulses mean less average current draw from the power supply.
Show me the scopeshot of current flowing through L1. Maybe its inductance is too low due to excessive core gapping.
Is the Akula persona using all non-inductive resistors, too?
Helllo Verpies
As pointed out earlier, because the configuration is common emitter, sinking current ( with pull-up resistor), when the pulse is NARROWER @ PIN 11; then the inverse of the signal would correspond to a LONGER pulse at the MOSFET gate, ie; pulse = 10%DC @ pin 11 then 90%DC @ gate. IF the circuit is functioning properly, of course....
take care, peace.
lost_bro
EDIT; needed to change the BOLDed out part of quote
Here are screenshots of both Akula's and XName41's boards. It looks like Akula has a normal beige color resistor, while XName41 has a blue resistor.
They look like 2 or 3 watt resistors.
Beige = Carbon Film
Blue = Metal Film
(in most cases)
The load or current limiting resistor on the LED in the video looks to be wirewound.
Quote from: Groundloop on 2014.04.12, 21:42:20
Hoppy,
If you cut the track then you need to add a 10K resistor from gate on MOSFET to ground.
This because with the track cut, there is no current path for the gate to discharge to ground.
Also, do you have a o-scope image (DC) of the gate voltage to the MOSFET?
GL.
GL & all,
I've now re-instated the track and as requested attach some scope shots taken from mosfet gate to ground (DC coupled, x10). Shot 06 shows the waveform well before it breaks into audible ferrite core oscillation by adjustment of the 'Duty' pot. Shot 07 shows the waveform just before it breaks into audible oscillation. Likewise, shot 08 shows the waveform as scoped at cathode of D6 well before it breaks into audible oscillation. Shot 09 shows the waveform just before it breaks into audible oscillation. I have not found it possible to get a stable shot of the waveform whilst in oscillation (tried all trigger modes).
At all points of pot adjustments, the LED cluster shows no sign of illumination. However, during periods of ferrite core oscillation (sounds loud, like a squealing pig!) cap C10 gets hot. Originally, I had 2 x 1000uF, 35V in parallel and both of these started to blow and leak. I've now replaced these with a single 3,000uF, 63V cap.
Hoppy
Quote from: Hoppy on 2014.04.13, 09:04:55
GL & all,
I've now re-instated the track and as requested attach some scope shots taken from mosfet gate to ground (DC coupled, x10). Shot 06 shows the waveform well before it breaks into audible ferrite core oscillation by adjustment of the 'Duty' pot. Shot 07 shows the waveform just before it breaks into audible oscillation. Likewise, shot 08 shows the waveform as scoped at cathode of D6 well before it breaks into audible oscillation. Shot 09 shows the waveform just before it breaks into audible oscillation. I have not found it possible to get a stable shot of the waveform whilst in oscillation (tried all trigger modes).
At all points of pot adjustments, the LED cluster shows no sign of illumination. However, during periods of ferrite core oscillation (sounds loud, like a squealing pig!) cap C10 gets hot. Originally, I had 2 x 1000uF, 35V in parallel and both of these started to blow and leak. I've now replaced these with a single 3,000uF, 63V cap.
Hoppy
Hoppy,
Thank you for providing the o-scope shots.
>>the LED cluster shows no sign of illumination
This indicates a problem in you circuit. Also the capacitor going warm indicate a high current draw to ground somewhere.
Have you used a voltmeter and checked what voltage you get in C10? Is the LEDs connected the right way? Do the LEDs
have a 1 Ohm resistor current path to ground? Any shorts in your soldering? Etc.
GL.
Quote from: lost_bro on 2014.04.13, 01:02:01
As pointed out earlier, because the configuration is common emitter, sinking current ( with pull-up resistor), when the pulse is NARROWER @ pin 11; then the inverse of the signal would correspond to a LONGER pulse at the MOSFET gate,
Correct and because of this inversion, the functions of error amplifiers (pins 2 &15) are also inverted and their activation (pin 3 high) causes the lengthening of MOSFET's ON-state.
Longer MOSFET conduction time translates to more average current in the transformer/inductor.
Good catch :)
Quote from: Groundloop on 2014.04.13, 09:38:45
Hoppy,
Thank you for providing the o-scope shots.
>>the LED cluster shows no sign of illumination
This indicates a problem in you circuit. Also the capacitor going warm indicate a high current draw to ground somewhere.
Have you used a voltmeter and checked what voltage you get in C10? Is the LEDs connected the right way? Do the LEDs
have a 1 Ohm resistor current path to ground? Any shorts in your soldering? Etc.
GL.
GL,
I have checked my LED array as functional on a DC supply and have checked all output tracks and connections as good on the PCB and not grounding. The overall current draw varies considerably but can run high at certain pot settings. Its possible that as Verpies commented, the problem is that reverse blocking voltage of the diode is being exceeded. The LED array has the 1R resistor path to ground and there is no shorts that I can find.
Hoppy
Quote from: Hoppy on 2014.04.13, 10:38:29
GL,
I have checked my LED array as functional on a DC supply and have checked all output tracks and connections as good on the PCB and not grounding. The overall current draw varies considerably but can run high at certain pot settings. Its possible that as Verpies commented, the problem is that reverse blocking voltage of the diode is being exceeded. The LED array has the 1R resistor path to ground and there is no shorts that I can find.
Hoppy
Hoppy,
Did you check the C10 voltage? I also agree that your Schottky diodes may be "busted" by high voltage coil spikes.
GL.
Quote from: Groundloop on 2014.04.13, 10:50:01
Hoppy,
Did you check the C10 voltage? I also agree that your Schottky diodes may be "busted" by high voltage coil spikes.
GL.
Yes, I have seen voltages well in excess of its rating. What are your experiences with your replication?
Hoppy
Quote from: Hoppy on 2014.04.13, 09:04:55
Shot 06 shows the waveform well before it breaks into audible ferrite core oscillation by adjustment of the 'Duty' pot. Shot 07 shows the waveform just before it breaks into audible oscillation.
That means that the acoustic oscillation starts as the conduction time of the MOSFET decreases.
The gate driver seems to have problems with rising slew rate (BJT's job). The TL494 is responsible only for discharging the gate (falling slew rate).
Quote from: Hoppy on 2014.04.13, 09:04:55
Likewise, shot 08 shows the waveform as scoped at cathode of D6 well before it breaks into audible oscillation. Shot 09 shows the waveform just before it breaks into audible oscillation.
Too bad these waveforms are not measured in reference to the positive terminal of C11, because D6 is oriented in such a ways as to conduct only when the voltage generated by L2 exceeds the voltage across C10.
This can be seen on the scopeshot as the negative level of the waveform, which is established by the conduction of D6 above the voltage level of C10.
Quote from: Hoppy on 2014.04.13, 09:04:55
I have not found it possible to get a stable shot of the waveform whilst in oscillation (tried all trigger modes).
The core must be ratting randomly. That is not necessarily bad.
Quote from: Hoppy on 2014.04.13, 09:04:55
At all points of pot adjustments, the LED cluster shows no sign of illumination. However, during periods of ferrite core oscillation (sounds loud, like a squealing pig!) cap C10 gets hot.
Maybe the diodes are conducting in reverse or C10 is overvoltaged by strong pulses from the flyback action in L1 ...or magnetoacoustics from L2.
Dear All.
Please find attached scope shot of volt drop across R19 (1 ohm ) Ch1 and R17 Ch2 ( 0.2 ohm ) Gnd ref on + C11.
Cheers Grum.
PS. I may have got these positions wrong as I have forgotten what Verpies told me yesterday !! Should have written it down !! ;D
http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg397514/#msg397514
Ok, to save lots of time I am offering voice communications over Teamspeak 3.
Teamspeak 3 - client download:
http://www.teamspeak.com/?page=downloads
My private Teamspeak 3 server address is: sirius.proxydns.com
Please use your nickname and no password when connecting.
howto - http://youtu.be/S32AGZdY5yA?t=1m56s
Looking forward to see everyone speaking there :)
P.S> All voice communications are encrypted in this server.
Quote from: Hoppy on 2014.04.13, 10:57:41
Yes, I have seen voltages well in excess of its rating. What are your experiences with your replication?
Hoppy
Hoppy,
My replica is waiting for parts. I hope to get the parts on Monday or Tuesday.
GL.
Quote from: Grumage on 2014.04.13, 13:25:50
Please find attached scope shot of volt drop across R19 (1 ohm ) Ch1 and R17 Ch2 ( 0.2 ohm ) Gnd ref on + C11.
C12 removed ?
@ Verpies,
I've scoped around 80V DC across C11, with no discernible waveform. Its just enough to puff-up the top of my 63V rated cap C3.
I need to clear my bench next week to start some work - work but will look-in whenever I can.
Hoppy
Quote from: verpies on 2014.04.13, 15:01:36
C12 removed ?
Dear Verpies.
No C12 was not removed. These double sided boards are really tough to work with so I left it there !! I am beginning to think that I should have done a breadboard first !!
Did I pick the correct resistors to scope on ?? And was the ref position correct ??
Thinking out loud I feel that the pulse width is FAR to wide, I was seeing much more interesting results with the very narrow pulses provided by my old PWM.
Cheers Grum.
Quote from: Grumage on 2014.04.13, 16:35:03
Thinking out loud I feel that the pulse width is FAR to wide, I was seeing much more interesting results with the very narrow pulses provided by my old PWM.
Cheers Grum.
Hi Grum
Funny that you mention this about the pulse width..... I remember seeing another schematic when the thread started that indicated an Emitter Follower configuration for the MOSFET driver part of the circuit.
Well the current schematic is Common Emitter , sinking current (with pull-down resistor) and this inverts the TL494 signal (with respect to MOSFET gate), and the Emitter Follower (sourcing current, without pull-up resistor) configuration does NOT invert the signal coming from the TL494 as does the Common Emitter config.
So, as of now when the pulse is Narrow @ Pin 11 >> WIDER pulse @ MOSFET GATE, and as stated by Verpies:
" because of this inversion, the functions of error amplifiers (pins 2 &15) are also inverted and their activation (pin 3 high) causes the lengthening of MOSFET's ON-state.
Longer MOSFET conduction time translates to more average current in the transformer/inductor."
OF course more average current in the transformer translates to more average power consumed and apparently more heat generated.
So, the Emitter-Follower dirver config. that I saw on the other Akula 30watt schematic would pulse the MOSFET gate 'in time' (non-inverted with respect to gate) with the pulse width of the TL494 logic.
This would give a shorter pulse 'in sycn' with the pulse width generated by the TL494 in response to it feedback loop.....
I guess the point of this commentary is that it strikes me odd to believe that both so very different circuit driver topologies can be correct, when apparently their action is so very different!
Let me know your ideas on this.
take care, peace.
lost_bro
Quote from: lost_bro on 2014.04.13, 17:00:09
I guess the point of this commentary is that it strikes me odd to believe that both so very different circuit driver topologies can be correct, when apparently their action is so very different!
take care, peace.
lost_bro
Good point, which begs the question, exactly what is the two versions of this circuit correct for?? :-\
Hoppy
Quote from: lost_bro on 2014.04.13, 17:00:09
...it strikes me odd to believe that both so very different circuit driver topologies can be correct, when apparently their action is so very different!
Good point.
What driver topology is shown by the pertinent Akula video?
Quote from: Grumage on 2014.04.13, 16:35:03
No C12 was not removed.
Unfortunately this needs to be done because a CSR cannot be bypassed by a capacitor. Cutting one leg of the cap is enough too...
Quote from: Grumage on 2014.04.13, 16:35:03
Did I pick the correct resistors to scope on ?? And was the ref position correct ??
Yes x2 if you were referring to
Groundloop's schematic (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12309) diagram component designations.
Quote from: Grumage on 2014.04.13, 16:35:03
Thinking out loud I feel that the pulse width is FAR to wide, I was seeing much more interesting results with the very narrow pulses provided by my old PWM.
Lost_bro might have a good take on the inverted MOSFET conduction.
Quote from: verpies on 2014.04.13, 18:38:22
Good point.
What driver topology is shown by the pertinent Akula video?
Hello Verpies
OK, I remember that I posted over at Over unity.com many weeks ago something pertaining to a discrepancy with the video in as much, if I remember correctly that the Schematic shown on said video did NOT match the schematic shown in the link of the same. (different driving circuits)
Don't remember now which was which, but when I have time will check my posts at O.U.com.
take care, peace.
lost_bro
If you want to play around with the pins on the chip to swap the internal driver over, i would bend the 4 legs so they don't plug into the chip socket and then use 4 pins from another chip socket plugged into the 4 sticking out pins and run wires to where you need to connect them, it saves cutting tracks O0
Hi all,
i am back home, so i picked up where i left off and as my leds arrived, i hooked them up and made the same measurements as before.
I have my R1 (0,22 Ohm) starting to smoke after a few seconds which is similar as Hoppy has if i remember well.
I still have to read up on what is published up till now.
Anyway, here a short video of what i did, and a screenshot of the severall signals measured when momentarely powering on the circuit.
Yellow trace is the gate voltage,
blue the drain voltage,
green the drain current and
purple the pin 11 signal
The cores "clack" together when the voltage is applied.
When i remove the current limiter from my PS, the current running into the circuit is 10A at 14V, the video and screenshot where taken when the current was limited to 7A causing the voltage to drop to 12V.
https://www.youtube.com/watch?v=ag_ry97wktA&feature=youtu.be
Regards Itsu
Quote from: Itsu on 2014.04.13, 21:07:10
Anyway, here a short video of what i did, and a screenshot of the severall signals measured when momentarely powering on the circuit.
Yellow trace is the gate voltage,
Good day Itsu
I will comment on the Yellow trace: 8.4 volts is NOT enough voltage to drive the MOSFET into conduction fast enough, usually one would want a signal in amplitude of about 10-11 volts just to make sure the MOSFET is completely ON.
take care, peace.
lost_bro
Quote from: Itsu on 2014.04.13, 21:07:10
Yellow trace is the gate voltage,
blue the drain voltage,
green the drain current and
purple the pin 11 signal
A scopeshot of current flowing in the windings and voltage signal on pin.3 (with different pot settings) when the scope is triggered on MOSFET's gate voltage signal, would reveal much more.
Quote from: Hoppy on 2014.04.13, 21:19:48
Waveform 11 is scoped at the cathode of D6 with the LED brightly lit.
That's 80V above ground! ...+ spikes.
What was the voltage across C10 and C11 during this measurement?
Quote from: lost_bro on 2014.04.13, 21:23:32
8.4 volts is NOT enough voltage to drive the MOSFET into conduction fast enough, usually one would want a signal in amplitude of about 10-11 volts just to make sure the MOSFET is completely ON.
8V significantly exceeds V
GS_ON thus this voltage is sufficient to fully turn ON most MOSFETs (see attachment).
You are correct, that 8V might not be sufficient to turn on some MOSFETs
fast enough but they will turn ON eventually with that V
GS.
Whether that happens in 100µs or in 100ns depends on the total gate charge of a particular MOSFET.
Quote from: verpies on 2014.04.13, 23:35:41
8V significantly exceeds VGS_ON thus this voltage is sufficient to fully turn ON most MOSFETs (see attachment).
You are correct, that 8V might not be sufficient to turn on some MOSFETs fast enough but they will turn ON eventually with that VGS.
Whether that happens in 100µs or in 100ns depends on the total gate charge of a particular MOSFET.
Hi Verpies,
Yes, it is above Vgs, but looking at the waveform in yellow and speaking of QGD, It looks to me like the MOSFET will spend too much time passing thru ( or should I say; sleeping) in the Miller Plateau....
When the MOSFET switches, the gate voltage becomes clamped to the plateau voltage (Miller Plateau) and stays there until sufficient charge has been added/ removed for the device to switch.
The wave form, Leading edge *should* be much more abrupt. This is Definitely NOT nano-pulsing.
Take care, peace.
lost_bro
Quote from: verpies on 2014.04.13, 23:15:34
That's 80V above ground! ...+ spikes.
What was the voltage across C10 and C11 during this measurement?
I've deleted that post (619) as my probe was incorrectly set to x1 C.C The correct scoped voltages are as follows: -
Across C10 with 'STOP' button normally closed: 11.4V
Across C10 with 'STOP' button open: 15.4V
Across C11 with 'STOP' button normally closed: 7.4V
Across C11 with 'STOP' button open: 7.6V
Hoppy
Quote from: lost_bro on 2014.04.13, 21:23:32
Good day Itsu
I will comment on the Yellow trace: 8.4 volts is NOT enough voltage to drive the MOSFET into conduction fast enough, usually one would want a signal in amplitude of about 10-11 volts just to make sure the MOSFET is completely ON.
take care, peace.
lost_bro
Thanks Lost, i guess my current limiter on my bench PS (set at 7A) probably limited the voltage to this 8.4V as it should have put 12V on the gate via the driving transistor.
I will try again with the current unlimited to see if this changes the gate signal.
Regards Itsu
Quote from: verpies on 2014.04.13, 23:05:48
A scopeshot of current flowing in the windings and voltage signal on pin.3 (with different pot settings) when the scope is triggered on MOSFET's gate voltage signal, would reveal much more.
verpies,
i can do that, any specific winding (L1 / L2) or pot or both?
Regards Itsu
You don't have two current probes to measure current in two windings simultaneously. You could use the R3 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) as a CSR if it is non-inductive but then your probe's ground clip would gave to be at the positive terminal of C11 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307), which would create a ground loop if you wanted to scope pin.3 and the gate at the same time in reference to pin 7 (GND) :(
Pot R11 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) affects the signal at pin.3 the most.
Quote from: lost_bro on 2014.04.14, 01:52:52
It looks to me like the MOSFET will spend too much time passing thru ( or should I say; sleeping) in the Miller Plateau....
Yes, the gate spends a long time at the Miller Plateau but is faster switching called for by the designer of this circuit?
Who said that nanosecond pulsing is desirable in this application at all?
Dear All.
Please find attached scope shot of...
Ch 1. Signal at gate of Mosfet.
Ch 2. Voltage at L1.
Ch 3. Signal on pin 3.
All with ref to Gnd.
Cheers Grum.
Quote from: verpies on 2014.04.14, 12:13:05
You don't have two current probes to measure current in two windings simultaneously. You could use the R3 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) as a CSR if it is non-inductive but then your probe's ground clip would gave to be at the positive terminal of C11 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307), which would create a ground loop if you wanted to scope pin.3 and the gate at the same time in reference to pin 7 (GND) :(
Pot R11 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) affects the signal at pin.3 the most.
Ok, 2 screenshots, first one with:
gate voltage (yellow)
pin 3 voltage (blue)
L1 current (green)
second one with:
gate voltage (yellow)
pin 3 voltage (blue)
L2 current (green)
picture is from bench PS no current limit at full load (same as the screenshots)
R11 has no influence on any of the traces.
Next i will swap one of the L1 or L2 leads and remeasure.
EDITTED: i had 2 the same screenshots presented, now i have corrected that.
Be aware that the current needs to be taken times 2 due to my current terminator settings.
Regards Itsu
You all seem to have a duty of 50% or more, is that right ? Hence high current draw.
Great work all,
Grum, I tried to save your hijacked thread over at OU.com
http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg397851/#msg397851 (http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg397851/#msg397851)
http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg397397/#msg397397 (http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg397397/#msg397397)
Quote from: Peterae on 2014.04.14, 19:54:34
You all seem to have a duty of 50% or more, is that right ? Hence high current draw.
Correct, 56% (or 44% depending how/where you look :) )
But there is no way to change this, the one pot (R15) changes the frequency, the other pot (R11) the voltage on pin1 which does not change the duty cycle.
Regards Itsu
Quote from: Itsu on 2014.04.14, 19:59:24
Correct, 56% (or 44% depending how/where you look :) )
But there is no way to change this, the one pot (R15) changes the frequency, the other pot (R11) the voltage on pin1 which does not change the duty cycle.
Regards Itsu
Does the chip allow the duty cycle to be changed? Maybe another pot is needed? Pins 3 or 4?
Datasheet:
http://www.ti.com/lit/ds/symlink/tl494.pdf (http://www.ti.com/lit/ds/symlink/tl494.pdf)
This might be helpful:
http://pisarski.info/?p=19
Quote from: 4Tesla on 2014.04.14, 20:03:38
Does the chip allow the duty cycle to be changed? Maybe another pot is needed?
It does, but we have discussed this before, and making changes to this would change the circuit we are trying to replicate which should be avoided.
Regards Itsu
Quote from: Itsu on 2014.04.14, 20:15:58
It does, but we have discussed this before, and making changes to this would change the circuit we are trying to replicate which should be avoided.
Regards Itsu
But if it doesn't work as is, why not change?
Here i swapped the L1 leads.
The screenshot with L2 now looks almost identical of that of Grumage above.
Noticeable is also that the input current went back to 8A.
Same probe points as before.
R11 again did not make any difference.
Regards itsu
Im also thinking the duty cycle described by you guys is way too high, wasting lots of energy here IMHO. Btw did you notice, that akula uses different types of wire for L1 and L2, though in schematic both are specified as same diameter?
Dear All.
With a great deal of Skyping over the weekend and today I have found out a bit more. For both Hoppy and myself with L2 connected the LED's are extinguished and current increases drastically !! Just running L1 I have a very efficient LED driver !! So efficient I took out the string (due to overheating !!) I am now unable to continue as some of the LED's were damaged when the "meltdown" occurred !! China to the rescue again !! :)
I will have to take a back seat for a couple of weeks while the new ones are on the way.
Cheers Grum.
Fair enough about the duty, as you say if it's set using component values then it must be as intended, although i understand the duty changes depending on the chips output pins, which are different in the dual transistor version, maybe that version has 0-50%.
Other Planet
He is using fiberglass high temp sheathing that's a sort of woven fabric, in one winding he has stretched it to make it's diameter smaller due to the required length i think, so is the wore thicker on one winding? or is it just the sheathing has been pulled.
Quote from: Peterae on 2014.04.14, 20:44:52
Other Planet
He is using fiberglass high temp sheathing that's a sort of woven fabric, in one winding he has stretched it to make it's diameter smaller due to the required length i think, so is the wore thicker on one winding? or is it just the sheathing has been pulled.
What are the differences of this type of insulation other than thickness and max temperature compared to PVC/teflon/... ? Maybe different dielectric behaviour?
Is akula also using different conducting medium in L2?
What i also tried was opening the stop switch, then activating the start switch (power on).
This causes the LEDs to almost explode, so needing to switch off again the start switch asap.
The leds will keep on glowing for some minutes after this start switch is opened (power removed) gradually decreasing their lights.
Closing the stop switch will cause this glowing to stop immediately.
Video here: https://www.youtube.com/watch?v=c6ldTllVXLY&feature=youtu.be
This was with the L1 leads swapped.
Regards Itsu
My original strip-board was built as the push-pull version and this also pulled a heavy current at supply voltages much above 9V. However, it does appear to be a fairly efficient low voltage LED driver if the load is matched to the supply voltage. IMO this is what Akula designed it to be, thus his conversion of a mains operated LED lamp to battery operation. My 2 parallel strings of 5 x 3W LED chips is not a 30W loading, so given that Akula presumably used a full 30W load, then a battery voltage of 12V may turn out to be optimum for this loading in terms of best efficiency. I will when I have the time, increase loading using 4 x 12V / 9W LED lamps in parallel and report on my results. I will also revert back to BC546 / 556 push-pull driver for comparison.
Hoppy
Thank you so much Peter and GL, I received the two " Shark-01" boards in todays morning mail.
I will get to building them promptly right after "tax day" tomorrow.
On another note, there seem to be two boards, one by Matt Watts, one by GL. Are there more?
Would it be a good idea to put all the latest info to build the a given board into a pdf appropriately named e.g Shark-01.pdf?
Notes scattered through the forum could be added and the rev # updated as errors / improvements are found.
I have been very busy with some crisis on the home front I'm afraid I may missed some important stuff.
I'll try to go back over all 20+ pages as time permits to see if I did.
Thanks again guys.
Quote from: Peterae on 2014.04.14, 19:54:34
You all seem to have a duty of 50% or more, is that right ? Hence high current draw.
With pins 13 & 14 connected together, the TL494 can output duty cycle from 0% to 48%, but because its output is inverted, the MOSFET can be driven only from
100% to 52% duty cycle.
The only way to change these boundaries is to ground pin 13 or to
not invert the TL494's output to the MOSFET's gate.
Quote from: Grumage on 2014.04.14, 17:52:24
Please find attached scope shot of...
Ch 1. Signal at gate of Mosfet.
Ch 2. Voltage at L1.
Ch 3. Signal on pin 3.
All with ref to Gnd.
Negative gate pulses, WTF?
Quote from: Itsu on 2014.04.14, 18:43:23
Ok, 2 screenshots, first one with:
gate voltage (yellow)
pin 3 voltage (blue)
L1 current (green)
Slow gate rise time. Miller plateau is clearly visible.
VT2 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) is not doing its job well :(
Pin 3 voltage not showing any activity, unless these 1V spikes are not artifacts caused by EMI. The waveform at pin 3 should have a 5V amplitude as shown in
this video (http://youtu.be/exIFDDgSrng?t=1m37s).
Is the junction of
R5 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) &
R7 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) showing any negative pulses or quickly falling edges?
Quote from: Itsu on 2014.04.14, 18:43:23
R11 has no influence on any of the traces.
That would be expected if there is no negative or quick activity on the junction of
R5 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) &
R7 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307)Quote from: Itsu on 2014.04.14, 18:43:23
Next i will swap one of the L1 or L2 leads and remeasure.
So, I will not analyze transformer currents at this time.
Today, I received 14 of the "Shark-01" boards - finally arrived! O0
Peter has some left for those on the east side of the Atlantic pond, I understand.
So I am very happy to send to anyone in the US/west of the pond one of these boards, that is, to anyone interested in doing the build.
Quote from: ION on 2014.04.14, 21:48:42
Thank you so much Peter and GL, I received the two " Shark-01" boards in todays morning mail.
Would it be a good idea to put all the latest info to build the a given board into a pdf appropriately named e.g Shark-01.pdf?
Notes scattered through the forum could be added and the rev # updated as errors / improvements are found.
Thanks again guys.
Totally agree with ION -- it would be helpful to have the latest info on the components etc. for the build.
O0
@ verpies,
Here are two scope shots at junction of R5 / R7, referenced to ground. Shot 11 is with 'STOP' switch open and LED array lit. Shot 12 is with 'STOP' switch closed and LED array extinguished.
Hoppy
Hi,
My intuition tells the second step should be finding right materials for transformer.
If you manage to find sources selling ferrite cores made fro hard ferrite(this is essential for effect, these black ferrite keep magnetic field residue!) please let me know. In USSR most pulse transformers was made from this type of ferrite.
Also on top of that you should try using iron wire instead of copper as they have opposite magnetic properties...
Cheers!
Quote from: Hoppy on 2014.04.15, 07:28:14
Here are two scope shots at junction of R5 / R7, referenced to ground. Shot 11 is with 'STOP' switch open and LED array lit. Shot 12 is with 'STOP' switch closed and LED array extinguished.
Both of these waveforms appear to have sufficiently steep falling edges (>18V/ms) to trigger the comparator in TL494 at pins 1&2.
The 2
nd waveform (with 'STOP' switch closed) also has -500mV negative excursions, that would trigger the second comparator in TL494 at pins 15&16.
Do these comparators take notice of these pulses? Can you see the evidence of this on pin 3 ?
I wish I could see scopeshots like this from Itsu and Grumage.
P.S.
Are you using an inverting MOSFET driving circuit topology like
this (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12101) or non-inverting like
this (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12299)?
Fantastic thanks for letting me know they have arrived O0
Dear Verpies.
I am using inverting ( single transistor ) Mosfet driver.
BTW. why? WTF !! ???
Cheers Grum.
Quote from: verpies on 2014.04.15, 09:08:51
Both of these waveforms appear to have sufficiently steep falling edges (>18V/ms) to trigger the comparator in TL494 at pins 1&2.
The 2nd waveform (with 'STOP' switch closed) also has -500mV negative excursions, that would trigger the second comparator in TL494 at pins 15&16.
Do these comparators take notice of these pulses? Can you see the evidence of this on pin 3 ?
I wish I could see scopeshots like this from Itsu and Grumage.
P.S.
Are you using an inverting MOSFET driving circuit topology like this (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12101) or non-inverting like this (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12103)?
I can try that tonight.
By the way, i am using an inverting MOSFET driving circuit topology, but looking at the screenshot i published last sunday,
the signal from pin 11 (TL494 put) see the purple trace, has the same polarity/shape as the gate signal, see the yellow trace.
So to me that means it is NOT inverted; how could that be?
Also notice the signal shape is identical, including the Miller plateau signature, how could that be on pin 11?
Another question i have is about the (unnamed) diode right after the start switch (D1 in GL diagram).
In my case it is an 1n4007 (1A), but after all my tests showing 10A going into the circuit, it still measures OK.
Guess its a matter of time this diode will pop, right? Need to use a more sturdy one here?
Regards Itsu
Quote from: verpies on 2014.04.15, 09:08:51
P.S.
Are you using an inverting MOSFET driving circuit topology like this (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12101) or non-inverting like this (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12103)?
Verpies,
Like the first schematic (Akula's original schematic) with the single transistor.
Waveform 14 is scoped at pin 3 to ground with the 'STOP' switch open and LED fully lit. Waveform 15 is scoped at pin 3 to ground with 'STOP' switch closed.
Hoppy
Quote from: Itsu on 2014.04.15, 09:55:31
the signal from pin 11 (TL494 put) see the purple trace, has the same polarity/shape as the gate signal, see the yellow trace.
So to me that means it is NOT inverted; how could that be?
Because the signal is
not inverted between pin 11 and MOSFET's gate.
The signal is inverted between the NOR gate and the MOSFET's gate.
Quote from: Itsu on 2014.04.15, 09:55:31
Also notice the signal shape is identical, including the Miller plateau signature, how could that be on pin 11?
Because only a 7Ω resistor separates pin 11 and the gate of the MOSFET.
Quote from: Itsu on 2014.04.15, 09:55:31
Another question i have is about the (unnamed) diode right after the start switch (D1 in GL diagram).
In my case it is an 1n4007 (1A), but after all my tests showing 10A going into the circuit, it still measures OK.
Guess its a matter of time this diode will pop, right? Need to use a more sturdy one here?
Yup
Quote from: Grumage on 2014.04.15, 09:53:37
I am using inverting ( single transistor ) Mosfet driver.
I know.
The allegedly successful replication in
this video (http://www.youtube.com/watch?v=OOG6YQ_6K6k) uses the
non-inverting (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12690) TL494 -->MOSFET driving circuit topology that matches
this schematic (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12686).
...and
this schematic (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12688) is equivalent to it signal-wise.
Quote from: Grumage on 2014.04.15, 09:53:37
BTW. why? WTF !! ???
Because, gates of N-ch enhancement mode MOSFETs need to be positive in reference to the source terminal in order to turn on.
Also, where from are you getting the negative voltage in reference to ground ?
I found a bit more time for playing today and I now have 3 x 9W LED lamps in parallel as a load and have found that the duty pot can now be adjusted to bring the LED light level down to given level or completely off. The current draw is still high at a supply voltage not much above 9.5V before my bench PSU clamps and with a 12V battery, the draw is over 10 Amps.
Hoppy
Quote from: Hoppy on 2014.04.15, 12:31:13
Waveform 14 is scoped at pin 3 to ground with the 'STOP' switch open and LED fully lit. Waveform 15 is scoped at pin 3 to ground with 'STOP' switch closed.
Waveform 14 shows no significant activity. The deviation from a flat line seems to be just HF interference.
Waveform 15 shows significant activity since 3.2V on pin 3 is sufficient to decrease the output duty cycle of the TL494 to 3% at the NOR gate.
DTY = 3% at the NOR gate after inversion yields DTY = 97% at the MOSFET, so it is not surprising, that the supply current consumption skyrockets.
In other words, increased voltage at pin.3 causes increased duty cycle at the MOSFET due to the polarity inversion after the NOR gate.
The feedback loop of TL494 appears to be positive. Positive feedback loops are unusual in PWM power circuits because they create run-away duty cycles.
Quote from: verpies on 2014.04.15, 15:20:08
Waveform 14 shows no significant activity. The deviation from a flat line seems to be just HF interference.
Waveform 15 shows significant activity since 3.2V on pin 3 is sufficient to decrease the output duty cycle of the TL494 to 3% at the NOR gate.
DTY = 3% after inversion yields DTY = 97% at the MOSFET, so it is not surprising, that the supply current consumption skyrockets.
In other words, increased voltage at pin.3 causes increased duty cycle at the MOSFET due to the polarity inversion.
The feedback loop of TL494 appears to be positive. Positive feedback loops are unusual in SMPS.
So this would make some sense of the NC 'STOP' switch if we had no polarity inversion and explain why Akula revised his schematic to show the push-pull driver.
Quote from: verpies on 2014.04.15, 15:20:08
Waveform 14 shows no significant activity. The deviation from a flat line seems to be just HF interference.
Waveform 15 shows significant activity since 3.2V on pin 3 is sufficient to decrease the output duty cycle of the TL494 to 3% at the NOR gate.
DTY = 3% at the NOR gate after inversion yields DTY = 97% at the MOSFET, so it is not surprising, that the supply current consumption skyrockets.
In other words, increased voltage at pin.3 causes increased duty cycle at the MOSFET due to the polarity inversion after the NOR gate.
The feedback loop of TL494 appears to be positive. Positive feedback loops are unusual in PWM power circuits because they create run-away duty cycles.
Verpies,
If we connect the gate resistor (6,8R) to pin 8 on TL494 instead, will the feedback be negative then?
GL.
Quote from: Hoppy on 2014.04.15, 15:31:12
So this would make some sense of the NC 'STOP' switch if we had no polarity inversion and explain why Akula revised his schematic to show the push-pull driver.
...and later
the schematic (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12688) with the integrated non-inverting driver
IXDD609 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12688).
Quote from: Groundloop on 2014.04.15, 15:40:11
If we connect the gate resistor (6,8R) to pin 8 on TL494 instead, will the feedback be negative then?
No :(
Here the screenshot of the signal on the Junction R5/R7 (purple trace) together with the gate signal (yellow) and pin 3 (Blue):
Stop switch normally closed
Regards Itsu
Here the same with the stop switch opened:
Regards Itsu
For those that have not yet populated their PCB's and wish to go straight for the push-pull version 2, I have attempted to list out the stages of modification: -
1) Cut pin 9 from IC or holder if used.
2) Bend out pin 10 and wire to R1 pin on VT2 side. Do not fit a resistor in R1 position. Link IC pins 11 & 12 on underside of PCB.
3) Build push-pull circuit (New PNP transistor Q3, 100R, 2 x 1K, 15V mosfet gate zener and 1N4148 diode - across 10R) on a small piece of strip-board and connect into circuit with input to R1 pin on Q1 side (to connect with new wire link).
Connect new PNP transistor (BC556) emitter to Q1 emitter and its collector to a conveniently located ground connection. Connect output to the mosfet gate (Q2).
4) Do not fit C12, R3 & R19. Replace Q1 with a BC546 transistor.
5) Cut PCB track between junction of R1 / Q1 base and junction of D2 cathode.
I have not done this myself yet, so it will need checking for accuracy.
Hoppy
Quote from: Itsu on 2014.04.15, 16:10:06
Here the screenshot of the signal on the Junction R5/R7 (purple trace) together with the gate signal (yellow) and pin 3 (Blue):
Yup, the pulses are so short that the comparators (pins 2 & 15) are not even noticing it.
Most likely your R5 and C5 are too good of a quality ;) , ...or your MOSFET switching is weak, inverted or both.
Quote from: Hoppy on 2014.04.15, 16:46:41
For those that have not yet populated their PCB's and wish to go straight for the push-pull version 2, I have attempted to list out the stages of modification: -
And for those already having a non-inverting integrated MOSFET driver such as the IXDD609 or UCC27511 or TC4428 and wanting to go to
ersion 3 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12688), the list of modifications is below:
IN REFERENCE TO Akula's DIAGRAM (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307):1) Remove R16, R17, R18, C12, VT2, VD1.
2) Connect pin 11 of the TL494 to pin 12.
3) Bend out pin 10 out of the socket and connected it to where the junction between R19 and C12 used to be.
NOTE: at this point you can connect the junction of R19&C12 to the gate of the MOSFET even without the integrated driver. The downside of this is that the MOSET will not switch as fast as with the integrated driver.4) Bypass the power supply pins of the integrated MOSFET driver chip with a 470nF (or greater) ceramic capacitor (directly soldered across the supply pins at the underside of the driver chip)
5) Superglue the integrated MOSFET driver chip on its back to the power MOSFET.
6) Connect the positive power supply pin of the integrated MOSFET driver chip to the output pin of the three-terminal 12V linear voltage regulator (or to the positive terminal of C6) with at least 0.5mm dia. wire (short & stranded wire is better)
7) Connect the negative power supply pin (a.k.a. GND) of the integrated MOSFET driver to the ground pin of the three-terminal 12V linear voltage regulator (or to the negative terminal of C6) with at least 0.5mm dia. wire (short & stranded wire is better).
8) Connect the non-inverting output of the integrated MOSFET driver chip to the gate of the MOSFET through a 4.7Ω carbon resistor.
9) Connect the non-inverting input of the integrated MOSFET driver chip to where the junction between R19 and C12 used to be
Quote from: verpies on 2014.04.15, 15:41:46
...and later the schematic (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12688) with the integrated non-inverting driver IXDD609 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12688).
Hello All
Interesting note regarding the IXYS series of MOFET driver ICs:
Some of the IXYS series MOSFET drivers utilize integrated Schmidt triggers for wave forming logic. I remember on a couple of projects we used a few of them with excellent results...... gives 10s of nano-second leading edge rise times. Miller plateau all but eliminated.
I have not had time to check the spec. sheet for the IXDD609 series, (as posted in schematic by Verpies) to see it they have the integrated Schmidtt trigger logic and/or Active Miller clamping.
When my board arrives I will try the Schmidt trigger mod.
take care, peace
lost_bro
EDIT Yes, it does have the integrated Schmidt trigger for waveform shaping..... see attachments
Quote from: Hoppy on 2014.04.15, 16:46:41
For those that have not yet populated their PCB's and wish to go straight for the push-pull version 2, I have attempted to list out the stages of modification: -
And for those already having a non-inverting integrated MOSFET driver such as the IXDD609 or UCC27511 or TC4428 and wanting to go to
version 3 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12688), the list of modifications is below:
IN REFERENCE TO Groundloop's PCB (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12309):1) Remove R1, R2, R3, C1, Q1, D2.
2) Connect pin 11 of the TL494 to pin 12.
3) Bend out pin 10 out of the socket and connected it to where the junction between R4 and C1 used to be.
NOTE: at this point you can connect the junction of R4/C1 to the gate of the MOSFET even without the integrated driver. The downside of this is that the MOSET will not switch as fast as with the integrated driver.4) Bypass the power supply pins of the integrated MOSFET driver chip with a 470nF (or greater) ceramic capacitor (directly soldered across the supply pins at the underside of the driver chip)
5) Superglue the integrated MOSFET driver chip on its back to the power MOSFET.
6) Connect the positive power supply pin of the integrated MOSFET driver chip to the output pin of the three-terminal 12V linear voltage regulator (or to the positive terminal of C7) with at least 0.5mm dia. wire (short & stranded wire is better)
7) Connect the negative power supply pin (a.k.a. GND) of the integrated MOSFET driver to the ground pin of the three-terminal 12V linear voltage regulator (or to the negative terminal of C7) with at least 0.5mm dia. wire (short & stranded wire is better).
8) Connect the non-inverting output of the integrated MOSFET driver chip to the gate of the MOSFET through a 4.7Ω carbon resistor.
9) Connect the non-inverting input of the integrated MOSFET driver chip to where the junction between R4 and C1 used to be
If somebody wants to use the TC4428 integrated MOSFET driver, below are its connections:
The bypass capacitor C1 should be soldered similarly to the photo below, but on the underside of the chip and across the power supply pins 3 & 6.
...and the same for the UCC27511 integrated MOSFET driver chip:
In this video (http://www.youtube.com/watch?v=bYEuV-ji4tk) the author of a similar and allegedly self-running LED flashlight circuit, shows the internal construction of its transformer.
Such videos are uncommon. I've never seen a transformer gutted like this by its constructor.
A schematic diagram that was reverse-engineered by magpwr (http://www.overunity.com/profile/magpwr.9705/) directly from this video (http://youtu.be/gWjs0-WCMzs), is attached below:
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12713)
Quote from: verpies on 2014.04.15, 20:12:52
In this video (http://www.youtube.com/watch?v=bYEuV-ji4tk) the author of a similar and allegedly self-running LED flashlight circuit, shows the internal construction of its transformer.
Such videos are uncommon. I've never seen a transformer gutted like this by its constructor.
Quite revealing verpies:
Two copper shield layers noted.
Outer winding appeared darker in color. Would that be from heat or is that wire a coated black annealed iron wire?
Cardboard spacer between center of pot core.
Teflon tape between layers.
Anything else noticed and relevant?
For the hell of it, I'll initially try the ICL7660 inverting driver as a comparison to the two versions of the bipolar transistor drivers. I'm currently right out of non-inverting drivers.
Hoppy
Quote from: Matt Watts on 2014.04.15, 21:58:03
Anything else noticed and relevant?
The copper shields are coated with some adhesive. The inner shield is gapped. The outer one might be also.
Perhaps more details can be gleaned from this video, by people with better eyes. Such as:
1) Turn count
2) Layer count
3) Layer advancement
4) The top winding direction in relation to the bottom winding direction ( a.k.a.
dot convention (http://en.wikipedia.org/wiki/Dot_convention) )
5) Soldered connections on those coppers shields.
Quote from: Hoppy on 2014.04.15, 22:06:20
For the hell of it, I'll initially try the ICL7660 inverting driver as a comparison to the two versions of the bipolar transistor drivers. I'm currently right out of non-inverting drivers.
Yes, you can do that as long as you feed this inverting driver with the signal from pin 11. Don't forget a pullup resistor around 200Ω from pin 11 to +Vcc.
At a risk of being boring, I will reiterate for others what modifications should be made to the original Akula circuit for using an
inverting integrated MOSFET driver to drive the MOSFET without inverting the total signal path (because of double inversion):
IN REFERENCE TO Akula's DIAGRAM (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307):1) Remove R18, R19, C12, VT2, VD1.
2) Pull out one side of R16 and one side of R17 (the sides at their junction) out of the PCB and connect them together in the air (not touching the PCB)
3) Optionally, you can change R17 to a 0Ω resistor (e.g. wire).
4) Bypass the power supply pins of the integrated MOSFET driver chip with a 470nF (or greater) ceramic capacitor (directly soldered across the supply pins at the underside of the driver chip)
5) Superglue the integrated MOSFET driver chip on its back to the power MOSFET.
6) Connect the positive power supply pin of the integrated MOSFET driver chip to the output pin of the three-terminal 12V linear voltage regulator (or to the positive terminal of C6) with at least 0.5mm dia. wire (short & stranded wire is better)
7) Connect the negative power supply pin (a.k.a. GND) of the integrated MOSFET driver to the ground pin of the three-terminal 12V linear voltage regulator (or to the negative terminal of C6) with at least 0.5mm dia. wire (short & stranded wire is better).
8) Connect the output of the integrated MOSFET driver chip to the gate of the MOSFET through a 4.7Ω carbon resistor.
a) use the non-inverting output, or
b) ...or use the inverting output
9) Connect the input of the integrated MOSFET driver chip to the junction between R16 and R17 that is hanging in the air.
a) if pt.8a was done, then use the inverting input
b) if pt.8b was done, then use the non-inverting input
For example the TC4428 or the UCC27511 driver can be used in this configuration like this:
At a risk of being boring, I will reiterate for others what modifications should be made to the original Akula circuit for using an inverting integrated MOSFET driver to drive the MOSFET without inverting the total signal path (because of double inversion):
IN REFERENCE TO Groundloop's PCB (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12309):
1) Remove R2, R4, C1, Q1, D2.
2) Pull out one side of R1 and one side of R3 (the sides at their junction) out of the PCB and connect them together in the air (not touching the PCB)
3) Optionally, you can change R1 to a 0Ω resistor (e.g. wire).
4) Bypass the power supply pins of the integrated MOSFET driver chip with a 470nF (or greater) ceramic capacitor (directly soldered across the supply pins at the underside of the driver chip)
5) Superglue the integrated MOSFET driver chip on its back to the power MOSFET.
6) Connect the positive power supply pin of the integrated MOSFET driver chip to the output pin of the three-terminal 12V linear voltage regulator (or to the positive terminal of C6) with at least 0.5mm dia. wire (short & stranded wire is better)
7) Connect the negative power supply pin (a.k.a. GND) of the integrated MOSFET driver to the ground pin of the three-terminal 12V linear voltage regulator (or to the negative terminal of C6) with at least 0.5mm dia. wire (short & stranded wire is better).
8) Connect the output of the integrated MOSFET driver chip to the gate of the MOSFET through a 4.7Ω carbon resistor.
a) use the non-inverting output, or
b) ...or use the inverting output
9) Connect the input of the integrated MOSFET driver chip to the junction between R1 and R3 that is hanging in the air.
a) if pt.8a was done, then use the inverting input
b) if pt.8b was done, then use the non-inverting input
For example the TC4428 or the UCC27511 driver can be used in this configuration like this:
Quote from: verpies on 2014.04.15, 22:07:09
The copper shields are coated with some adhesive. The inner shield is gapped. The outer one might be also.
Perhaps more details can be gleaned from this video, by people with better eyes. Such as:
1) Turn count
2) Layer count
3) Layer advancement
4) The top winding direction in relation to the bottom winding direction ( a.k.a. dot convention (http://en.wikipedia.org/wiki/Dot_convention) )
5) Soldered connections on those coppers shields.
Cannot discern turn count, but outer winding is one complete layer; inner winding is almost two complete layers with no tape between. It appears winding direction is opposite between primary and secondary. Inner winding has two parts; one of which is only a small number of turns and is positioned on one side of the bobbin, not centered over the gap.
Quote from: Matt Watts on 2014.04.15, 22:57:39
It appears winding direction is opposite between primary and secondary.
Didn't he drop the spool and picked it up
turned around, making it appear that the winding direction is opposite ?
Yes, between being clumsy and the camera focus changing, it is hard to tell.
I do now see the TWO windings on the inner portion. He unspools to a point where he pulls the bobbin out of his hand. When he picks it back up, he breaks the entire winding which looks to be about a dozen turns.
When he disassembles the transformer, is there any indication that the core is magnetized, rather than ferromagnetic?
Grumpy I did an analysis of the core sometime ago, one of his cores had the manufacturers part code on, unfortunately they were no longer in business but i found the data which suggested it was MnZn ferrite material, it appears to be a black almost crumbly looking core material, it has a very low resistance something like 6Ohms per meter.
Guys if he is using a double sticky sided tape on the copper foil then that will also insulate it when a layer is placed on top of itself, i dont think there would be any shorted turns in that copper.
@All,
First test run of my Akyla board. I have connected a 12 Volt 3,5 Watt LED lamp to the output.
The Input is 8,2 Volt @ 0,26 Amp. The voltage over the LED lamp is 13,7 Volt. LED lamp
seems to be full brightness. Nothing is heating up in my circuit and the MOSFET is cold.
I'm running with the "STOP" switch open. I did try to close the "STOP" switch but the
input current just went up to over 1 Ampere so I guess I did get the windings on the L2
coil the wrong way around. Only the OSC adjustment had any effect. I just adjusted
the OSC pot-meter until I got the highest voltage output. I'm running a N27 core right
now. My coil data: L1 = 76 turn 0,5mm, L2 = 186 turn 0,5mm, L1/L2 Ratio = 1:2,447.
Coil is standard with L1 first then L2 on top of that. The LED lamp is slightly warm.
Added: Test 2
IN = 08,00 Volt @ 0,22 A = 1,76 Watt
OUT= 12,63 Volt @ 0,10 A = 1,26 Watt
COP = 0,72
GL.
@ all,
I have now modified the GL PCB to drive the mosfet from an ICL7600 chip. Current consumption is much better at 2.4A @ 12V (as read from my bench PSU) on max frequency of 131.6KHz. Lowest frequency is 18.9KHz. See attached scope shot 16 taken at mosfet gate and shot 17 across R17.
The 'duty' pot can be adjusted to allow a sharp transition from LED lit to LED extinguished by operation of the 'STOP' switch. Duty cycle is 42% at top of frequency range and 43.9% at bottom of range.
Must now get some work done!
Nice neat build GL. :)
I note that you have a lot more turns on your L1 & L2 coils than Akula.
Hoppy
Quote from: Hoppy on 2014.04.16, 11:11:50
Nice neat build GL. :)
I note that you have a lot more turns on your L1 & L2 coils than Akula.
Hoppy
Hoppy,
Yes, that is correct. I want lower current usage and I wanted to see if more turns makes
any different result for the feedback from L2. But I think I got my L2 backwards.
GL.
Hi guys :)
Akula showed more details as per my request - https://www.youtube.com/watch?v=_DkjdsQ9vug
In first part he showed effect what happens when Tesla coil effect is applied perpendicularly to normal windings on 90 degrees. Obviously this big toroid was not good success for first intended OU device. The ground wire is used there as well.
In second part of video the circuit board is from LEDs 1W schematics using toroid from 1st part. This is where it was OK on self running it seems...
Cheers!
Quote from: Hoppy on 2014.04.16, 11:08:19
scopeshot 17 across R17.
With C12
removed on GL's PCB ?
Quote from: T-1000 on 2014.04.16, 11:26:18
Obviously this big toroid was not good success for first intended OU device.
What's under all these windings?
A ferrite core?
Quote from: verpies on 2014.04.16, 11:26:38
With C12 removed on GL's PCB ?
No, C12 is in.
Quote from: T-1000 on 2014.04.16, 11:26:18
Hi guys :)
This is where it was OK on self running it seems...
Cheers!
Do we take it then, that the two 30W versions we are working on never self-ran?
Quote from: verpies on 2014.04.16, 11:47:29
What's under all these windings?
A ferrite core?
I can assume, - yes, most likely just still question what made of ferrite - hard or soft. All OU effects I observerd over last few years was with hard ferrite http://en.wikipedia.org/wiki/Ferrite_%28magnet%29#Hard_ferrites
Quote from: Hoppy on 2014.04.16, 11:52:53
Do we take it then, that the two 30W versions we are working on never self-ran?
Interesting question which can be answered only after 1:1 replication.
I can assume it was self running if there was NMR effect involved.
Also I had conversation with Delamorto (where this saga began from with LEDs self runners in Youtube) and asked him to check something on ferrite because he did not know what type of ferrite it was and the compass needle was reacting to that ferrite even without pre-magnetization so this ferrite is presumably hard type which was same as in Lithuanian Yoke experiment material...
Cheers!
Quote from: T-1000 on 2014.04.16, 12:09:16
the compass needle was reacting to that ferrite
A ferrite can concentrate Earth's magnetic field and make the compass needle react to this concentration.
The inductance with known turn count, core size and gap width, as well as resistivity between two drops of salt water (5mm apart - not touching!) and the sound the ferrite makes after being struck while being suspended on a thread, can aid in ferrite identification better than the compass test.
Quote from: verpies on 2014.04.16, 15:22:21
A ferrite can concentrate Earth's magnetic field and make the compass needle react to this concentration.
If you try that with soft ferrite - won't work ;)
Also the hard ferrite has distinctive black color and soft ones from China - greyish...
Color is a much better indicator than a compass test.
Quote from: Hoppy on 2014.04.16, 11:49:44
No, C12 is in.
So R17 bypassed by a cap is not a good Current Sensing Resistor.
Quote from: verpies on 2014.04.16, 17:20:33
So R17 bypassed by a cap is not a good Current Sensing Resistor.
Yes, I do know that and just wanted to record the waveform for comparative purposes for other builders.
Quote from: verpies on 2014.04.16, 17:19:01
Color is a much better indicator than a compass test.
Do you maybe have a close up picture of a ferrite type u think will work?
And thank you very much for ur proposed integrated MOSFET driving mod, i think i will try directly with that mod when i assemble my akula 30w the next days. I will use the MIC4422BM, its working pretty good in my other builds at the moment, as long as switching frequency is below 300kHz, more than 300-400kHz destroys it.
kind regards and peace,
From other Planet
Dear All.
I carried out the mod's to the circuit today. I have used a TC4428 A.
Unfortunately I have no gate pulse at all !! Do I need to isolate the Mosfet gate from the board ?? As I see it R4 ( GL's board ) is also in circuit !!
HELP please !!
Cheers Grum.
Grum see verpies post here
http://www.overunityresearch.com/index.php?topic=2358.msg37705#msg37705
It says remove R4 O0
Dear All.
Ok. I have now got a pulse on Pin 10, but only at around 8.1 V dc. If I raise the voltage higher the pulse vanishes !! I started with a process of, install TL 494. Then I attached the TC 4428 A this increased the current to 1.6 A !! NOT GOOD methinks !!
So although things are starting to work I still need some advice.
Cheers Grum.
Quote from: verpies on 2014.04.15, 17:17:29
And for those already having a non-inverting integrated MOSFET driver such as the IXDD609 or UCC27511 or TC4428 and wanting to go to ersion 3 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12688), the list of modifications is below:
IN REFERENCE TO Akula's DIAGRAM (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307):
1) Remove R16, R17, R18, C12, VT2, VD1.
2) Connect pin 11 of the TL494 to pin 12.
3) Bend out pin 10 out of the socket and connected it to where the junction between R19 and C12 used to be.
NOTE: at this point you can connect the junction of R19&C12 to the gate of the MOSFET even without the integrated driver. The downside of this is that the MOSET will not switch as fast as with the integrated driver.
4) Bypass the power supply pins of the integrated MOSFET driver chip with a 470nF (or greater) ceramic capacitor (directly soldered across the supply pins at the underside of the driver chip)
5) Superglue the integrated MOSFET driver chip on its back to the power MOSFET.
6) Connect the positive power supply pin of the integrated MOSFET driver chip to the output pin of the three-terminal 12V linear voltage regulator (or to the positive terminal of C6) with at least 0.5mm dia. wire (short & stranded wire is better)
7) Connect the negative power supply pin (a.k.a. GND) of the integrated MOSFET driver to the ground pin of the three-terminal 12V linear voltage regulator (or to the negative terminal of C6) with at least 0.5mm dia. wire (short & stranded wire is better).
8) Connect the non-inverting output of the integrated MOSFET driver chip to the gate of the MOSFET through a 4.7Ω carbon resistor.
9) Connect the non-inverting input of the integrated MOSFET driver chip to where the junction between R19 and C12 used to be
I followed this mod using an ucc27511.
screenshot shows:
gate voltage (yellow)
drain voltage (blue)
pin3 signal (purple
drain current (green)
lowest frequency setting (16KHz) input 14V @ 10A
Regards Itsu
Guys,
When playing with this circuit do not forget to attach 10-20W bulb from car in series from battery so if it lights up in full, there i something wrong in circuit. We are talking about 12V very low power battery to power up final circuit here.. ;)
Quote from: Itsu on 2014.04.16, 21:10:21
I followed this mod using an ucc27511.
screenshot shows:
gate voltage (yellow)
Regards Itsu
Good Day Itsu
Now, that's the type of Gate drive pulse I like to see! O0
take care, peace
lost_bro
Sorry for the late reply but the O.U.R. forum was down.
If it wasn't down, you'd have had this reply at 9pm GMT..
Quote from: Grumage on 2014.04.16, 18:58:33
I carried out the mod's to the circuit today. I have used a TC4428 A.
Did you use the TC4428 in the
inverting mode (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12727) or
non-inverting mode (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12703) ?
Quote from: Grumage on 2014.04.16, 18:58:33
Unfortunately I have no gate pulse at all !!
What do you have at the input of the TC3328 ?
Quote from: Grumage on 2014.04.16, 18:58:33
Do I need to isolate the MOSFET gate from the board ??
No.
The removal of C1, D2 and R3 (even one sidedly) isolates the gate of the MOSFET from everything except the Zeners.
Quote from: Grumage on 2014.04.16, 18:58:33
As I see it R4 ( GL's board ) is also in circuit !!
R4 does not touch the gate because it is either removed or separated by removed C1 and D2 on Groundloop's PCB.
R3 does not touch the gate because it is pulled out of the PCB on one side or completely.
Quote from: Grumage on 2014.04.16, 21:09:49
Ok. I have now got a pulse on Pin 10,
The fact that you are mentioning pulses on pin 10, means that pin 11 must be connected to +Vcc and you are using the TC4428 in the
non-inverting mode (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12703).
This also means that you'd done the mods to Groundloop's PCB according to
this list (http://www.overunityresearch.com/index.php?topic=2358.msg37695#msg37695) and pin 10 is isolated from the PCB and connected to one side of R4 (the side that used to connect to C1).
The latter is connected to pin 4 of the TC4428.
Is the above correct?
Quote from: Grumage on 2014.04.16, 21:09:49
...but only at around 8.1 V DC.
That 8.1VDC is the power supply voltage applied across C8, isn't it?
Quote from: Grumage on 2014.04.16, 21:09:49
If I raise the voltage higher the pulse vanishes !!
What happens at the pin 3 of the TL494 as the pulse at pin 10 vanishes ?
Quote from: Itsu on 2014.04.16, 21:10:21
gate voltage (yellow)
drain voltage (blue)
pin3 signal (purple
drain current (green)
What was the state of the STOP switch and the polarity of L2 when this scopeshot was taken?
Gate voltage ref GND - too good (less than 50% duty cycle as expected). Pulses at pin 3 can decrease this duty cycle even down to 0%.
Drain voltage ref GND - Enough to break down a 50V rated diode (D5). @
Lost_bro, help me to figure out what causes that little falling edge in the second half of the OFF period (? voltage across L2 relative to voltage across C11 ?).
Pin3 voltage ref GND - Hmm, are those ±2V spikes an EMI artifact, Dirty ground (e.g. ground loop) or real steering pulses ? What is going on at pin 2 ?
Drain current - Drain current does not mean much. L1 or L2 current is significant.
Quote from: Itsu on 2014.04.16, 21:10:21
input 14V @ 10A
That's too much. We need to get the feedback loop (pins 2, 15, 3) to work properly so it can decrease the MOSFET's duty cycle and decrease the current draw with it.
The feedback loop is now negative so it is capable of accomplishing that.
P.S.
Please do not push this circuit to 14V 10A because sth will break. We should be striving for minimum current draw.
Quote from: verpies on 2014.04.17, 00:40:34
What was the state of the STOP switch and the polarity of L2 when this scopeshot was taken?
Stop switch is in its normally closed position. When in open position i will specifically mention this.
QuoteGate voltage ref GND - too good (less than 50% duty cycle as expected). Pulses at pin 3 can decrease this duty cycle even down to 0%.
too good Hmmm is that good? :D
QuoteDrain voltage ref GND - Enough to break down a 50V rated diode (D5). @Lost_bro, help me to figure out what causes that little falling edge in the second half of the OFF period (? voltage across L2 relative to voltage across C11 ?).
Not sure which D5 is, but my IDH12SG60C schottky diodes are rated 600V/12A
QuotePin3 voltage ref GND - Hmm, are those ±2V spikes an EMI artifact, Dirty ground (e.g. ground loop) or real steering pulses ? What is going on at pin 2 ?
Let me check tonight
QuoteDrain current - Drain current does not mean much. L1 or L2 current is significant.
OK, Let me check tonight
QuoteThat's too much. We need to get the feedback loop (pins 2, 15, 3) to work properly so it can decrease the MOSFET's duty cycle and decrease the current draw with it.
The feedback loop is now negative so it is capable of accomplishing that.
OK, when increasing the frequency, the current input goes down. At 75KHz its below 1A and i can keep it on continuously, so i can experiment better.
QuoteP.S.
Please do not push this circuit to 14V 10A because sth will break. We should be striving for minimum current draw.
Right, i will operate in the higher frequency range.
Thanks, regards Itsu
Quote from: verpies on 2014.04.17, 00:40:34
That's too much. We need to get the feedback loop (pins 2, 15, 3) to work properly so it can decrease the MOSFET's duty cycle and decrease the current draw with it.
The feedback loop is now negative so it is capable of accomplishing that.
I have found a point where the 'Duty' pot does progressively reduce the duty cycle but only when the supply voltage rail is reduced to around 8.5V. Of course at the duty cycle shown in the attached scope shot (taken at the mosfet gate), LED brightness is much reduced but I see no signs of self-running.
Hoppy
Another reply I found these videos on the project 30w akula!!!
http://www.youtube.com/watch?v=plORzm-4i18
http://www.youtube.com/watch?v=Qu2lN1w-zBA
Quote from: TutorialFE on 2014.04.17, 09:09:45
In another reply I found these videos on the project 30w akula!!!
http://www.youtube.com/watch?v=Qu2lN1w-zBA
...but that video does not show if the power supply was eventually disconnected
Quote from: Itsu on 2014.04.17, 08:11:38
Stop switch is in its normally closed position. When in open position i will specifically mention this.
but what about the L2 polarity. Which way is it connected?
Quote from: Itsu on 2014.04.17, 08:11:38
too good Hmmm is that good? :D
It's good if it does not hurt the desired operational principle. I don't know what the author of this circuit intended.
A crisp gate waveform like you have now is close to perfection in digital systems. It creates a lot of EMI, for sure.
Fortunately your UCC MOSFET driver is very universal and you can degrade its driving speed by increasing resistors at its outputs by e.g. 100x.
At this point, I do not know why degrading the performance if the UCC would be necessary (except for decreasing the EMI) but it is possible nonetheless.
Quote from: Itsu on 2014.04.17, 08:11:38
Not sure which D5 is, but my IDH12SG60C schottky diodes are rated 600V/12A
It is the diode connected to the drain of the MOSFET. On Groundloop's PCB it is called D5. It does not have a designation on Akula's diagram :/
Anyway, with the 600V rating you do not have to worry about it breaking down in reverse at these voltage levels.
Quote from: Itsu on 2014.04.17, 08:11:38
OK, when increasing the frequency, the current input goes down.
That is an indication that the L1 is limiting the current, because during a shorter MOSFET's ON time the current in L1 does not have the time to build up much.
Quote from: Hoppy on 2014.04.17, 08:55:25
I have found a point where the 'Duty' pot does progressively reduce the duty cycle but only when the supply voltage rail is reduced to around 8.5V.
Interesting. How does the feedback signal look at pin 15?
Quote from: verpies on 2014.04.17, 10:52:10
Interesting. How does the feedback signal look at pin 15?
See attached scope shot taken at pin 15. The duty pot adjustment is quite critical and only a small adjustment in the pot results in the progressive reduction of duty cycle. At anything above about 8.6V, the duty cycle locks into around 43%.
Hoppy
Maybe could be usefull for further information .....
Quote from: verpies on 2014.04.17, 10:23:38
but what about the L2 polarity. Which way is it connected?
well let me try to explain :-)
i have followed the V2 diagram from grumage, see below
Then i have followed the below picture, meaning 1 coil is CW, the other CCW.
Finally to comply to the dot notation, i have connected the both outer leads of the coils as represented by the dots on the exploded view of the transformer
Regards Itsu
@verpies
I many of your posts when you refer to circuit components I cannot even see those on the diagrams.
@all
I've been following the progress while working on FTPU stuff that I cannot stop. hehehe.
Just got two circuit boards, GL2014 and Shark-01. Thanks for the sends. I think I will stand by and see you guys move forward a little more. I am too novice in the circuit end to do major mods, etc, but.......
That R3 really bugs me. Seems to me it should be a diode pointing to source instead. That would change everything. Maybe another one of Akulas diagram tricks.
Can any one please try these;
1) Open SB1, run circuit and scope across L1 until L1 is in resonance, then close sb1. What frequency and what happens?
Then try...
2) Open SB1, run circuit but scope across L2 until L2 is in resonance, then close sb1. What frequency and what happens?
Why is this consuming what seems like 140 watts when it's supposed to only produce 30 watts? It should be running at 20 watts max if the OU is double less loses for looping.
Choked DC on R3 to L2 continuous then to R2 then feeding the load while filling C3 and pushing against the diode after L1. This means L2 is holding the core hostage at all times. The core is never left to resonate. L1 is simply banging its head on a wall. Then add the fact that some have overlapped the center gap with the secondary and that side is held hostage as well.
Seems to me the frequency of such a scheme has to be much higher then where you guys are to be able to counter the constant effect of L2 on the core. Otherwise that R3 is a big mistake. How can you pass energy looped back to source via R3?
140 watts when run at brute force may sound OK, but for 30 watts, this should be running at resonance with very little feed to get OU and looping. I know you will see waveforms rise anywhere between 0 and 300k range but those are usually token areas where the best resonances can only occur higher up. Plus no one has wound the coils in tuned mode so this whole thing is just a crap shot hoping for the right mix, but the gold could be buried right from the start under non-tuned coil winds. I had posted how (#517) but removed it because there was no interest. Guys should consider putting their circuit aside and use their FG on the core and see where it really comes to life, then work around that. I have ordered a few cores so eventually I will do it myself since for me this is a prerequisite in order to know where to run the coils.
wattsup
Quote from: verpies on 2014.04.17, 00:40:34
@Lost_bro, help me to figure out what causes that little falling edge in the second half of the OFF period (? voltage across L2 relative to voltage across C11 ?).
Hello Verpies,
If I told you I knew for sure what the artifact was, I would be telling you a lie.... >:-)
But, I'll throw in my 2 cents worth anyway....
Ok, After the initial *discharge spike* of the L1, we see a gradual linear discharging of C11 thru L1 From 20 > 16vdc (possible feeding the L2 and/or LED array?)..... then about 20us into the affair we see a very fast discharge, abruptly pulling the signal towards Zero potential which just as fast reaches an equilibrium @ what appears to be Vcc potential (from the scope shot appears about 12 - 13vdc).
Question: What is the operational Vcc of this Circuit when the scope shot was taken?
Question: What is the Voltage Value of C11 in this circuit when the scope shot was taken?
Question: What is happening simultaneously @ LED array at this moment?
Question: What are the inductances of L1 & L2?
Question: What about the copper tape to ground shown in the Akula diagram, Is it being used here?
I played around with LTspice last night just to see if I could duplicate this effect, and with standard models I could *not* duplicate it. I did notice some interesting signal results when the system was about to enter resonance. RLC.... R1; C11; L1;....
take care, peace
lost_bro
Quote from: verpies on 2014.04.17, 00:40:34
P.S.
Please do not push this circuit to 14V 10A because sth will break. We should be striving for minimum current draw.
Good day all:
Yes, has bothered me now for a while: the 7812 is a 1 amp voltage regulator..... the LM317 =1.5amps so, how can this circuit consume 10 or more amps and be expected to self run from a 1 amp regulator?
And...... 1amp X 12vdc = 12watts power..... so, how can this be a 30 watt device using just a single 1amp voltage regulator @ 12Vcc?
Please let me know what I have missed here...
take care, peace
lost_bro
Quote from: lost_bro on 2014.04.17, 14:49:17
Good day all:
Yes, has bothered me now for a while: the 7812 is a 1 amp voltage regulator..... the LM317 =1.5amps so, how can this circuit consume 10 or more amps and be expected to self run from a 1 amp regulator?
And...... 1amp X 12vdc = 12watts power..... so, how can this be a 30 watt device using just a single 1amp voltage regulator @ 12Vcc?
Please let me know what I have missed here...
take care, peace
lost_bro
Hi lost_bro,
The regulator only supplies the PWM. The regulator is drawn back-to-front from convention.
Hoppy
Quote from: Hoppy on 2014.04.17, 15:09:17
Hi lost_bro,
The regulator only supplies the PWM. The regulator is drawn back-to-front from convention.
Hoppy
Good day Hoppy.
Ok, so timing between L2 & L1 is extremely important: ie L2 must discharge into L1 when MOSFET is ON so L1* discharge pulse* can recharge L2 during MOSFET OFF cycle.......
Has anyone checked the timing between the L1 & L2?
take care, peace
lost_bro
Quote from: Hoppy on 2014.04.17, 11:39:24
See attached scope shot taken at pin 15. The duty pot adjustment is quite critical and only a small adjustment in the pot results in the progressive reduction of duty cycle. At anything above about 8.6V, the duty cycle locks into around 43%.
That waveform bothers me.
I wonder if it would still exist if you disconnected the feedback point form the junction of R6 & R7 (R5 & R7 on Akula's diagram) and connected it to ground near where the R7 is grounded (R5 on Akula's diagram).
If it still exists then it means that it is just reacting to EMI.
Quote from: wattsup on 2014.04.17, 14:00:49
@verpies
I many of your posts when you refer to circuit components I cannot even see those on the diagrams.
Because when I reply to Itsu I use the component designations from
Akula's diagram1 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307).
...and when I reply to Grumage I use the component designations from
Groundloop's diagram (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307), because Grum he is using Groundloop's PCB.
...with Hoppy I forgot what board he is using so am am stating both ;)
I know it's a mess ...but what can I do besides linking to a diagram each time I write a component designation ! ;(
Quote from: Hoppy on 2014.04.17, 15:09:17
The regulator only supplies the PWM. The regulator is drawn back-to-front from convention.
Yes, and its terminals are not marked on
Akula's diagram (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307).
L1 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) and
L2 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) are supplied from the positive terminal of
C11 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307)
Quote from: wattsup on 2014.04.17, 14:00:49
Just got two circuit boards, GL2014 and Shark-01. Thanks for the sends.
...
That R3 really bugs me. Seems to me it should be a diode pointing to source instead. That would change everything.
Do you mean R3 on the PC boards that you just received or on the Akula's
diagram1 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) or
diagram2 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12686) or
diagram3 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12688).
Why would a diode change everything ?
Hy GL!!
I would let you know that I ordered all the parts for the device.
I would ask if you have any information regarding how the winding needs to
be done on this little trasformator. (just to begin....)
Which type of copper wire(solid wire ?) to be use and what about the thickness
and how many turns and in which direction?
:-\
Quote from: TutorialFE on 2014.04.17, 16:13:38
Hy GL!!
I would let you know that I ordered all the parts for the device.
I would ask if you have any information regarding how the winding needs to
be done on this little trasformator. (just to begin....)
Which type of copper wire(solid wire ?) to be use and what about the thickness
and how many turns and in which direction?
:-\
TutorialFE,
I recommend that you do not solder the transformer to the PCB. Use wires from the PCB to the transformer legs.
That way you can experiment with different turns and wires etc. Also, look at the original Akyla drawing for
information on turns etc. I did use 0,5mm enameled copper wire on my transformer, but feel free to experiment.
GL.
Quote from: verpies on 2014.04.17, 16:11:43
Do you mean R3 on the PC boards that you just received or on the Akula's diagram1 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) or diagram2 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12686) or diagram3 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12688).
Why would a diode change everything ?
R3 in diagram 1 or 2 between the DC feed and the SB1 on top of L2. Having a diode there pointing to the feed side would permit L2 to return to the source without resistance and permit that L2 to resonate either alone or with the primary and also would permit Grumage to pulse his primary without his secondary half wind located under his primary to seize the core.
wattsup
Quote from: wattsup on 2014.04.17, 16:53:26
R3 in diagram 1 or 2 between the DC feed and the SB1 on top of L2.
wattsup
hello Wattsup
OK, now that you mention R3.... on a simulation that was posted many days or weeks ago??? I did change the value of R3 (@47R) and it did modify the function of the circuit. It would allow the LED array to be lit while at the same time allowing *some* energy back to the source at C11..........
Did not try a diode though, interesting (voltage drop & rectification).....
take care, peace
lost_bro
I think the 'duty' pot's main function appears to be to disable the PWM at a set rail voltage, probably to protect the LED load from damage due to over-current. It switched the PWM output and load LED's 'OFF' cleanly at a set supply voltage from around 8V to over 13V.
Hoppy
Ok I've check I've got 0.5 mm solid copper wire..
Do you have also instruction about the setting with the trimmer?
Dear All.
Short video of the problem of Pin 3 Cut off !!
https://www.youtube.com/watch?v=VfQlOhOocso
Remedies welcome !! :)
Cheers Grum.
Quote from: lost_bro on 2014.04.17, 14:36:33
Hello Verpies,
If I told you I knew for sure what the artifact was, I would be telling you a lie.... >:-)
But, I'll throw in my 2 cents worth anyway....
Ok, After the initial *discharge spike* of the L1, we see a gradual linear discharging of C11 thru L1 From 20 > 16vdc (possible feeding the L2 and/or LED array?)..... then about 20us into the affair we see a very fast discharge, abruptly pulling the signal towards Zero potential which just as fast reaches an equilibrium @ what appears to be Vcc potential (from the scope shot appears about 12 - 13vdc).
Question: What is the operational Vcc of this Circuit when the scope shot was taken?
Question: What is the Voltage Value of C11 in this circuit when the scope shot was taken?
Question: What is happening simultaneously @ LED array at this moment?
Question: What are the inductances of L1 & L2?
Question: What about the copper tape to ground shown in the Akula diagram, Is it being used here?
I played around with LTspice last night just to see if I could duplicate this effect, and with standard models I could *not* duplicate it. I did notice some interesting signal results when the system was about to enter resonance. RLC.... R1; C11; L1;....
take care, peace
lost_bro
Question: What is the operational Vcc of this Circuit when the scope shot was taken?
14V
Question: What is the Voltage Value of C11 in this circuit when the scope shot was taken?
10V + a 2V ripple, see purple trace on the below scopeshot
Question: What is happening simultaneously @ LED array at this moment?
its fully lit, at 14V + spikes, see blue trace on the below scopeshot
Question: What are the inductances of L1 & L2?
L1 500uH
L2 7mH
Question: What about the copper tape to ground shown in the Akula diagram, Is it being used here?
Yes see http://www.overunityresearch.com/index.php?topic=2358.msg36930#msg36930
The yellow trace again is the gate voltage, the green trace the drain current (as always need to be taken x2)
Regards Itsu
Quote from: verpies on 2014.04.17, 00:40:34
Pin3 voltage ref GND - Hmm, are those ±2V spikes an EMI artifact, Dirty ground (e.g. ground loop) or real steering pulses ? What is going on at pin 2 ?
Here a screenshot of pin 3 (purple trace) and pin 2 (blue trace) together again with the gate voltage (yellow trace) and the L2 current (green trace)
I lost my low current setting in high frequency mode after starting to use the ucc27511 driver.
High frequency (max. 116KHz) still pulls 4.5A @ 14V, and R3 and R1 are getting hot fast.
2e screenshot is with same settings as first one, except the green (current) trace now is from L1.
Regards Itsu
Quote from: Grumage on 2014.04.17, 18:54:10
Dear All.
Short video of the problem of Pin 3 Cut off !!
https://www.youtube.com/watch?v=VfQlOhOocso
Remedies welcome !! :)
Cheers Grum.
Hi Grum,
to me it looks like the same problem that Hoppy is reporting when he says that turning the duty cycle pot can cleanly cut off the leds and PWM (TL494).
Regards Itsu
Quote from: Hoppy on 2014.04.17, 17:14:16
I think the 'duty' pot's main function appears to be to disable the PWM at a set rail voltage, probably to protect the LED load from damage due to over-current.
Absolutely not !
The DTY pot adjusts the negative slope sensitivity that occurs at the junction of R7 & R6 (R5 & R7 on Akula's diagram )
See this
video (http://youtu.be/exIFDDgSrng?t=1m23s).
Quote from: Grumage on 2014.04.17, 18:54:10
Short video of the problem of Pin 3 Cut off !!
Temporarily remove one side of R6 on Groundloop's PCB. (R7 on Akula's diagrams).
This will disable the feedback loop and pin 3 of TL494 will stay low.
Please make a short video about what happens as the power supply voltage is increased slowly above the critical voltage while the Ch2 is connected to the junction of R6 and R7.
I cleaned up a little Akula's diagram v2 appearing in this video (http://youtu.be/EwCS15pRtH0?t=4m12s).
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12797)
I did not change this circuit electronically in any way. I only added missing component designations and cleaned it up a little graphically.
@Groundloop
Why on Groundloop's PCB, C4 is connected to pin 15 but on Akula's diagram the same capacitor (C8) is connected to pin 13 & 14 ?
Akula's diagram1 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) also shows this discrepancy.
Quote from: lost_bro on 2014.04.17, 14:36:33
Ok, After the initial *discharge spike* of the L1, we see a gradual linear discharging vof C11 thru L1
With MOSFET removed, that can happen only if voltage on C11 is greater than voltage on C3
NOTE: The replies below use the component designations from Akula's chematic diagram v2.1:
If the MOSFET is in place and switching rapidly, then the junction of L1 & D5 will have a high voltage in reference to ground after the MOSFET opens (in order to keep L1 current continuous).
When this voltage becomes higher than the voltage across C3, then D5 will conduct current and charge up C3 (potentially to to a very high voltage which is clamped down only by the LED bank and D6+L2+R3).
Quote from: lost_bro on 2014.04.17, 14:36:33
From 20 > 16vdc (possible feeding the L2 and/or LED array?)
D6 will conduct and allow current through L2 only if the voltage across C3 + the EMF generated by L2 is
greater than the voltage on C11.
Any current flowing through D6 will charge up C11.
BTW: Current flowing through the scrubber composed of R2 and C4 is minimal or momentary.
The above is the
MAJOR POWER LOOP of the entire circuit.
Quote from: lost_bro on 2014.04.17, 14:36:33
..... then about 20us into the affair we see a very fast discharge, abruptly pulling the signal towards Zero potential which just as fast reaches an equilibrium @ what appears to be Vcc potential (from the scope shot appears about 12 - 13vdc).
Can you rephrase that? 20μs after the gate edge or sth else ?
Quote from: lost_bro on 2014.04.17, 14:36:33
Question: What is happening simultaneously @ LED array at this moment?
The LED array always attempts to discharge C3 above its combined voltage drop (e.g. 15V). In effect the LED array attempts to clamp C3 to a voltage <= 15V.
A resistive load would discharge C3 from the beginning (from 0V), and the circuit would behave very differently with such load.
Quote from: lost_bro on 2014.04.17, 14:36:33
Question: What are the inductances of L1 & L2?
Mutual coupling coefficient is also unknown between L1 and L2.
Quote from: lost_bro on 2014.04.17, 14:36:33
Question: What about the copper tape to ground shown in the Akula diagram, Is it being used here?
I do not know but it is alleged that the copper tape is essential for the OU effect.
In a different, albeit related, 3V perpetual flashlight, the copper tape is allegedly placed like this around the inner winding (secondary):
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12782)
Quote from: verpies on 2014.04.17, 19:53:52
Absolutely not !
The DTY pot adjusts the negative slope sensitivity that occurs at the junction of R7 & R6 (R5 & R7 on Akula's diagram )
See this video (http://youtu.be/exIFDDgSrng?t=1m23s).
It most definitely does cease PWM output with my replication!
Quote from: Hoppy on 2014.04.17, 17:14:16
I think the 'duty' pot's main function appears to be to disable the PWM at a set rail voltage,
Quote from: Hoppy on 2014.04.17, 21:37:17
It most definitely does cease PWM output with my replication!
I believe you, but it is a secondary side effect, not a primary effect of the "rail voltage".
The R5 & C5 LC current has more direct causal connection to the feedback loop and pin 3 behavior, than the "rail voltage"
Quote from: verpies on 2014.04.17, 20:22:32
I cleaned up a little Akula's diagram v2 appearing in this video (http://youtu.be/EwCS15pRtH0?t=4m12s).
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12784)
I did not change this circuit electronically in any way. I only added missing component designations and cleaned it up a little graphically.
@Groundloop
Why on Groundloop's PCB, C4 is connected to pin 15 but on Akula's diagram the same capacitor (C8) is connected to pin 13 & 14 ?
Akula's diagram1 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307) also shows this discrepancy.
@All,
Seems that I messed up the C4 connection!
I apologize for the error. It can be fixed by lifting
the plus wire on C4 out of the PCB, and then
solder a wire from the plus on C4 to pin 14.
Again, I apologize for my mistake!
GL.
Quote from: Groundloop on 2014.04.17, 21:50:48
Seems that I messed up the C4 connection!
Just help everyone to fix it and add this correction to v2 of your PCB.
Some people might even miss this post and be unaware of it.
Rising edges from the "feedback takeoff point" appearing on pin 4 through C4 (C8 on Akula's diagram) can inhibit the TL494's outputs through the Dead Time Control circuitry.
This is most likely what Grumage and Hoppy are struggling with.
Quote from: wattsup on 2014.04.17, 16:53:26
R3 in diagram1 or diagram2 between the DC feed and the SB1 on top of L2.
Having a diode there pointing to the feed side would permit L2 to return to the source
There is already a diode in series with the L2. On
diagram v2.1 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12784) this diode is designated D6.
Another diode in series would not make much sense. R3 is there for a purpose, as a CSR or current sharing resistor or a limiter.
Quote from: wattsup on 2014.04.17, 16:53:26
without resistance and permit that L2 to resonate either alone or with the primary
Do you mean an LC resonance involving L2 or an acoustic resonance in the core ?
If you mean LC resonance then adding another diode in series with L1 & D6 would not change anything.
Quote from: verpies on 2014.04.17, 21:51:29
Just help everyone to fix it and add this correction to v2 of your PCB.
Some people might even miss this post and be unaware of it.
Rising edges from the "feedback takeoff point" appearing on pin 4 through C4 (C8 on Akula's diagram) can inhibit the TL494's outputs through the Dead Time Control circuitry.
This is most likely what Grumage and Hoppy are struggling with.
Verpies,
I have updated the design and corrected the error. I have posted the new design
to my work bench. I also posted how to fix the C4 error.
GL.
Copper foils in that orientation will create a radial electric field perpendicular to the coil windings.
This is a biasing technique mention in TPU-related theory where two perpendicular fields are used to induce a current. One of the forces has to rotate (i.e. change position) and the other force is static. Magnetic fields can also be used. This is why I asked if there was any indication that that core is magnetized.
Quote from: Groundloop on 2014.04.17, 22:40:07
I have updated the design and corrected the error. I have posted the new design
to my work bench.
If I was proficient with Eagle I would help you renumber the component designations to match Akula's diagrams, to avoid all that confusion.
Quote from: verpies on 2014.04.17, 22:48:16
If I was proficient with Eagle I would help you renumber the component designations to match Akula's diagrams, to avoid all that confusion.
Verpies,
A lot of work but I can do it on the latest version posted.
GL.
Quote from: verpies on 2014.04.17, 21:21:01
With MOSFET removed, that can happen only if voltage on C11 is greater than voltage on C3
NOTE: The replies below use the component designations from Akula's chematic diagram v2.1:
D6 will conduct and allow current through L2 only if the voltage across C3 + the EMF generated by L2 is greater than the voltage on C11.
Any current flowing through D6 will charge up C11.
The above is the MAJOR POWER LOOP of the entire circuit.
Hi Verpies
Thanks for cleaning up the schematic....
OK, C11 once charged will supposedly discharge slowly thru L1 (Choke). The purpose of the L1 (choke) would be to act a surge suppressor for the transients so the 7812 doesn't burn up. This feedback loop as you stated is the major power loop for the OU system, right? So, as C11 charges up at the same time it would be discharging into the L1 (choke) to feed the 7812 to power the TL494.
But until the system is free running, the input voltage/current must pass thru L1 (choke) from opposite side to charge C11 the then thru L1 Primary to the Drain of the MOSFET (when switching)...... So C11 is held @ Vcc as a minimum voltage thru L1 (choke). As stated C11 can charge to a higher voltage when L1 EMF (primary) and C3 voltage exceed Vcc.
As a side note, I remember one of the Russian members at O.U.com mentioning something about the capacitor needing to be on the opposite side of the L1 (choke). I will have to look for that info.....
When I moved the capacitor to the *other* side of the L1 (choke) I noticed that the simulations in LTspice showed quasi resonant condition.... I have attached the LTspice file if anyone is interested.
My idea is this: What if a small boost/buck converter (1-3 amp variety) was substituted in place of the 7812 volt reg. When the voltage dipped below the set value, the boost converter would raise it, and when it went above, the buck would lower it.... This *could* give more leeway to the operation of the control circuitry to get it free running. Usually these small boost/bucks converters have decent efficiency ratings..
take care, peace
lost_bro
Quote from: Itsu on 2014.04.17, 19:43:37
Here a screenshot of pin 3 (purple trace) and pin 2 (blue trace) together again with the gate voltage (yellow trace)
I am still not sure if the signal at pin 3 can be trusted. The spikes on it become negative (EMI?) and the outputs of TL494's error amplifiers just cannot go below ground naturally.
Also, these spikes appear to go only 500mV above ground, and for them to have any effect on TL494's internal PWM comparator, they'd have to be at least 600mV above ground.
So my verdict is that the outputs of the error amplifiers (pin 3) are not getting activated and your feedback loop is not functioning.
To troubleshoot this you need to scope the "feedback takeoff point".
Quote from: Itsu on 2014.04.17, 19:43:37
the L2 current (green trace)...2e screenshot is with same settings as first one, except the green (current) trace now is from L1.
The comparison between L1 and L2 current reveals that they flow at the same time like in a regular transformer.
I don't know if this is what the author had intended.
I was expecting a current flow like in a flyback transformer (alternating between L1 and L2). My expectation was that L2 skims off the energy from C3 that the LEDs were not able to use up, because D6 conducts only when C3 is overcharged, precisely meaning V
C3 + V
L2 > V
C11. In the present configuration V
L2 = +3*V
L1 when the MOSFET is ON because of the transformer's 1:3 turn ratio and L2 polarity.
For my expected circuit behavior, L2 must be
reversed making V
L2 = -3*V
L1 when the MOSFET is ON, so the current in D6 & L2 cannot flow when the MOSFET is ON. L2 current can flow through D6 only when C3 had been overcharged by the flyback pulse via D5.
See
this simulation (http://falstad.com/circuit/#%24+1+4.9999999999999996E-6+0.41233529972698213+35+12.0+43%0A162+448+176+448+208+1+3.0+1.0+1.0+1.0%0A162+448+208+448+240+1+3.0+1.0+1.0+1.0%0A162+448+240+448+272+1+3.0+1.0+1.0+1.0%0A162+448+272+448+304+1+3.0+1.0+1.0+1.0%0A162+448+304+448+336+1+3.0+1.0+1.0+1.0%0Av+160+240+160+176+0+0+40.0+12.0+0.0+0.0+0.5%0AT+480+96+528+96+0+5.0E-4+3.0+1.1447183270729688E-9+3.103087965831985+0.7%0Ad+480+128+480+176+1+0.805904783%0Aw+448+176+480+176+0%0A159+352+240+352+272+0+0.1+1.0E10%0Aw+240+48+240+96+0%0Aw+240+96+240+176+0%0AR+304+256+272+256+0+2+500.0+5.0+0.0+0.0+0.2%0Ar+448+336+448+384+0+1.0%0Aw+240+240+240+384+0%0Ac+480+176+480+384+0+0.002+13.69457264381914%0Aw+448+384+480+384+0%0Ac+400+336+400+384+0+1.0E-5+0.05536795223256917%0Aw+448+384+400+384+0%0Aw+400+384+352+384+0%0Aw+240+384+352+384+0%0Aw+400+336+448+336+0%0Aw+480+128+352+128+0%0Aw+352+128+352+240+0%0Aw+352+272+352+384+0%0Ad+512+176+576+176+1+0.805904783%0Ar+576+208+512+208+0+420000.0%0Ac+576+240+512+240+0+1.0E-10+-0.8342177226628422%0Aw+512+176+512+208+0%0Aw+512+208+512+240+0%0Aw+576+176+576+208+0%0Aw+576+208+576+240+0%0Aw+480+176+512+176+0%0Aw+560+128+576+128+0%0Aw+560+96+576+96+0%0Aw+576+128+576+176+0%0Ar+240+48+352+48+0+1.0%0Ar+240+96+352+96+0+0.22%0Aw+352+96+480+96+0%0As+576+48+576+96+0+0+false%0Aw+352+48+576+48+0%0As+304+256+336+256+0+0+false%0Ag+240+384+240+416+0%0As+160+176+240+176+0+1+false%0Ac+240+176+240+240+0+0.002+9.760765449934386%0Ar+160+240+240+240+0+0.1%0Aw+528+96+560+128+0%0Aw+560+96+528+128+0%0Ap+528+96+528+128+0%0Ax+102+198+155+202+0+16+12VDC%0Ax+251+193+277+197+0+14+C11%0Ax+491+302+509+306+0+14+C3%0Ao+37+8+0+33+2.5+25.6+0+-1%0Ao+36+8+0+289+2.5+6.4+0+-1%0Ao+44+8+0+33+9.353610478917778+23.945242826029514+0+-1%0A) for what I mean.
Quote from: verpies on 2014.04.17, 21:44:54
I believe you, but it is a secondary side effect, not a primary effect of the "rail voltage".
The R5 & C5 LC current has more direct causal connection to the feedback loop and pin 3 behavior, than the "rail voltage"
It is nonetheless a feature that Akula may have intended in his design and makes sense insofar as limiting current to the LED array.
Once the PWM output has ceased, there may still be an amount of DC current through the LED array depending on the combined forward voltage of the series connected LED elements in the array and the supply voltage level. The typical forward voltage for my LED array is 3.6V x 5 LED elements = 18V for my array. However, current conduction will start as low as 2.6V per element, so I do see an increasing level of illumination as the supply voltage is increased beyond a PWM cut-off setting of 13V. Those builders using ordinary 12V LED lamps either singly or in parallel, will be forward conducting at a much lower voltage level.
Hoppy
Do you still believe in the intended PWM cutoff directly related to the rail voltage level after fixing the C4 error on Groundloop's PCB ?
Quote from: verpies on 2014.04.18, 08:41:12
Do you still believe in the intended PWM cutoff directly related to the rail voltage level after fixing the C4 error on Groundloop's PCB ?
Yes, the cap fix makes no difference to the PWM cut-off. I can see no observable affect of the cap being completely removed on the PWM output waveform.
Quote from: verpies on 2014.04.17, 23:22:12
I am still not sure if the signal at pin 3 can be trusted. The spikes on it become negative (EMI?) and the outputs of TL494's error amplifiers just cannot go below ground naturally.
Also, these spikes appear to go only 500mV above ground, and for them to have any effect on TL494's internal PWM comparator, they'd have to be at least 600mV above ground.
So my verdict is that the outputs of the error amplifiers (pin 3) are not getting activated and your feedback loop is not functioning.
To troubleshoot this you need to scope the "feedback takeoff point".
The comparison between L1 and L2 current reveals that they flow at the same time like in a regular transformer.
I don't know if this is what the author had intended.
I was expecting a current flow like in a flyback transformer (alternating between L1 and L2). My expectation was that L2 skims off the energy from C3 that the LEDs annot not use, because D6 conducts only when C3 is overcharged, precisely meaning VC3 + VL2 > VC11. In the present configuration VL2 = +3*VL1 when the MOSFET is ON because of the transformer's 1:3 turn ratio and L2 polarity.
For my expected circuit behavior, L2 must be reversed making VL2 = -3*VL1 when the MOSFET is ON, so the current in D6 & L2 cannot flow when the MOSFET is ON. L2 current can flow through D6 only when C3 had been overcharged by the flyback pulse via D5.
See this simulation (http://falstad.com/circuit/#%24+1+4.9999999999999996E-6+0.41233529972698213+35+12.0+43%0A162+448+176+448+208+1+3.0+1.0+1.0+1.0%0A162+448+208+448+240+1+3.0+1.0+1.0+1.0%0A162+448+240+448+272+1+3.0+1.0+1.0+1.0%0A162+448+272+448+304+1+3.0+1.0+1.0+1.0%0A162+448+304+448+336+1+3.0+1.0+1.0+1.0%0Av+160+240+160+176+0+0+40.0+12.0+0.0+0.0+0.5%0AT+480+96+528+96+0+0.01+3.0+8.502254733588899E-10+1.0747372309418068+0.999%0Ad+480+128+480+176+1+0.805904783%0Aw+448+176+480+176+0%0A159+352+240+352+272+0+0.1+1.0E10%0Aw+240+48+240+96+0%0Aw+240+96+240+176+0%0AR+304+256+272+256+0+2+500.0+5.0+0.0+0.0+0.2%0Ar+448+336+448+384+0+1.0%0Aw+240+240+240+384+0%0Ac+480+176+480+384+0+0.002+16.059954830831582%0Aw+448+384+480+384+0%0Ac+400+336+400+384+0+1.0E-5+1.0434352769434803%0Aw+448+384+400+384+0%0Aw+400+384+352+384+0%0Aw+240+384+352+384+0%0Aw+400+336+448+336+0%0Aw+480+128+352+128+0%0Aw+352+128+352+240+0%0Aw+352+272+352+384+0%0Ad+512+176+576+176+1+0.805904783%0Ar+576+208+512+208+0+420000.0%0Ac+576+240+512+240+0+1.0E-10+-0.8079059429112156%0Aw+512+176+512+208+0%0Aw+512+208+512+240+0%0Aw+576+176+576+208+0%0Aw+576+208+576+240+0%0Aw+480+176+512+176+0%0Aw+560+128+576+128+0%0Aw+560+96+576+96+0%0Aw+576+128+576+176+0%0Ar+240+48+352+48+0+1.0%0Ar+240+96+352+96+0+0.22%0Aw+352+96+480+96+0%0As+576+48+576+96+0+0+false%0Aw+352+48+576+48+0%0As+304+256+336+256+0+0+false%0Ag+240+384+240+416+0%0As+160+176+240+176+0+1+false%0Ac+240+176+240+240+0+0.002+10.78182233832853%0Ar+160+240+240+240+0+0.1%0Aw+528+96+560+128+0%0Aw+560+96+528+128+0%0Ap+528+96+528+128+0%0Ax+102+198+155+202+0+16+12VDC%0Ao+15+16+0+34+20.0+3.2+0+-1%0Ao+44+16+0+289+7.654505172902097+156.76426594103498+0+-1%0Ao+25+16+0+33+160.0+1.6+0+-1%0A) for what I mean.
I will concentrate on the feedback signal to see what is happening there tonight.
Good to see that your simulation also shows ± 10A being pulled when closing the 12V switch.
A short video about the present circuit and the way it behaves when powering off, especially pin3 signal is going up/down shortly after power off.
Also some temperature measurements are done on the severall components.
https://www.youtube.com/watch?v=ofRW4OeFExk&feature=youtu.be
Regards Itsu
Quote from: Itsu on 2014.04.18, 09:07:34
A short video about the present circuit and the way it behaves when powering off,
I little girl just run up to me with her hands in the air, yelling: "Aaaaa, go, go... Itsu has made another video!" :)
Quote from: lost_bro on 2014.04.17, 23:16:06
OK, C11 once charged will supposedly discharge slowly thru L1 (Choke). The purpose of the L1 (choke) would be to act a surge suppressor for the transients so the 7812 doesn't burn up.
Yes
Quote from: lost_bro on 2014.04.17, 23:16:06
This feedback loop as you stated is the major power loop for the OU system, right?
In my posts I use two phrases containing the word "loop"
The "major power loop" consisting of L2, SB1, R3, C11, R1, L1, D5, C3, D6 in that order (with Q1 as a catalyst).
The small signal "feedback loop" consisting of the "feedback takeoff point", R7, C15, pin 2 & pin 15, culminating in pin 3.
Quote from: lost_bro on 2014.04.17, 23:16:06
So, as C11 charges up at the same time it would be discharging into the L1 (choke) to feed the 7812 to power the TL494.
Yes, I like to think about the VR1 as the origin of the auxiliary power supply.
Quote from: lost_bro on 2014.04.17, 23:16:06
But until the system is free running, the input voltage/current must pass thru L1 (choke) from opposite side to charge C11
Yes, in other words it appears that the author had intended to charge the major energy tank (C11) from the external battery via L3 in the
bootstrap mode and from L2 via D6 in the self-running mode.
Quote from: lost_bro on 2014.04.17, 23:16:06
the then thru L1 Primary to the Drain of the MOSFET (when switching).
That is the major discharge path for C11.
Quote from: lost_bro on 2014.04.17, 23:16:06
...So C11 is held @ Vcc as a minimum voltage thru L1 (choke).
What is Vcc in your analysis? Voltage across C11 or C6 or C3 ?
Quote from: lost_bro on 2014.04.17, 23:16:06
As stated C11 can charge to a higher voltage when L1 EMF (primary) and C3 voltage exceed Vcc.
IMO only D6 and L2 provide the recharge path for C11.
I don't count the series LC tank formed by C11 & L1 during the power-up transient.
Quote from: lost_bro on 2014.04.17, 23:16:06
My idea is this: What if a small boost/buck converter (1-3 amp variety) was substituted in place of the 7812 volt reg. When the voltage dipped below the set value, the boost converter would raise it, and when it went above, the buck would lower it.... This *could* give more leeway to the operation of the control circuitry to get it free running. Usually these small boost/bucks converters have decent efficiency ratings.
Yes, a buck/boost converter would be able to maintain the auxiliary voltage with wider input voltage range on C11 via L3.
Quote from: verpies on 2014.04.18, 09:16:36
I little girl just run up to me with her hands in the air, yelling: "Aaaaa, go, go... Itsu has made another video!" :)
Sweet. :)
Quote from: verpies on 2014.04.17, 20:04:09
Temporarily remove one side of R6 on Groundloop's PCB. (R7 on Akula's diagrams).
This will disable the feedback loop and pin 3 of TL494 will stay low.
Please make a short video about what happens as the power supply voltage is increased slowly above the critical voltage while the Ch2 is connected to the junction of R6 and R7.
Dear Verpies.
As you have since discovered the C4 problem would you like me to show a before, and after, the capacitor fix, with relation to R 6 removal ??
Cheers Grum.
Quote from: Grumage on 2014.04.18, 09:50:21
As you have since discovered the C4 problem would you like me to show a before, and after, the capacitor fix, with relation to R6 removal ??
Removing R6 (R7 on Akula's diagram) disables the feedback loop as stated in
this message (http://www.overunityresearch.com/index.php?topic=2358.msg37772#msg37772).
After you do the C4 fix (C8 on Akula's diagram) you might find out that pin 3 does not go high for higher supply voltages anymore, and disabling/troubleshooting the feedback loop is not necessary.
A short video about scoping the "feedback takeoff point"
* on Ch2 while the power supply voltage is increased slowly, is still called for.
You may trigger your scope on Ch1 connected to pin 10 so you are far away from EMI generating components (TC4428, MOSFET, L1 & L2)
* "feedback takeoff point" is the junction of R7 and R6 on Groundloop's PCB (R5 & R7 on Akula's diagrams)
Quote from: verpies on 2014.04.18, 09:12:58
I little girl just run up to me with her hands in the air, yelling: "Aaaaa, go, go... Itsu has made another video!" :)
I am glad you guys are having fun about my video's, it won't prevent me making them though ;D
What about increasing R5 (Akula diagram) to say 10 Ohm? Will it not increase the feedback signal to a level where it does affect the duty cycle?
Regards Itsu
The components arrived this morning but is missing the Toroidal Choke.....and the large heat sink
Quote from: TutorialFE on 2014.04.18, 10:31:23
The components arrived this morning but is missing the Toroidal Choke.....
Disassemble on old computer power supply.
You will find a suitable choke inside.
Quote from: Itsu on 2014.04.18, 10:30:17
What about increasing R5 (Akula diagram) to say 10 Ohm? Will it not increase the feedback signal to a level where it does affect the duty cycle?
It could.
Also, changing R5 to an inductive wirewound resistor will dramatically change the signal at the "feedback takeoff point".
Quote from: verpies on 2014.04.18, 10:33:40
Disassemble on old computer power supply.
You will find a suitable choke inside.
Don't tell him that, he might find at least half the components he needs inside. :)
@ Verpie Disassemble on old computer power supply.
You will find a suitable choke inside.
Hy Verpies I don't have the device to measure uH ...
This are about an year of serching ferrite in the litter....
But I don't know which one is good... Do you have any suggest?
Hi,
There is new video from akula (obviously it was attempted to delete very quickly, so no original link) - https://www.youtube.com/watch?v=ogMcs6iBJwY
The control circuit in the left is creating main transformer driving frequency and resonant frequency to the secondary primary.
The box in the right is impulse power supply (unless it was modified).
To me it is almost same what we did in Lithuanian Yoke experiment, just we did not self loop on the time.
Good luck!
Could someone tell to Akula to do a TutorialFE seriously with all information (All) to replicate in all the World with electronic components......
I'm very glad to see that Him is good constructor about Free Energy but his knowledge never go on market.... so He is wasting time to do movie without the ALL information... ^-^
More video more probably he will begin a Slave..... >:(
Tell him also open a Paypall account to receive money from everyone.. ;)
No spark gaps, no ground wire and not even heatsinks on the MOSFETs. Looks like a typical core from a flyback transformer. Good stuff T-1000. Maybe we can get lucky and see how close the schematic matches what we are working on.
Quote from: TutorialFE on 2014.04.18, 12:15:25
@ Verpie Disassemble on old computer power supply.
You will find a suitable choke inside.
Hy Verpies I don't have the device to measure uH ...
This are about an year of serching ferrite in the litter....
But I don't know which one is good... Do you have any suggest?
Use the core that has approximately the same size as the drawing on the PCB.
If the core has more than one winding, then remove the thinnest winding.
GL.
Whats that Steho Energy AG in the video
http://www.psiram.com/ge/index.php/Steho_Energy_AG
https://www.facebook.com/steho.energyag
Quote from: Peterae on 2014.04.18, 13:59:31
Whats that Steho Energy AG in the video
Not sure, but worth checking into.
http://www.google.com/translate?hl=en&ie=UTF8&sl=auto&tl=en&u=http%3A%2F%2Fesoterischewarnwelten.blogspot.com%2F2012%2F02%2Fsteho-energy-ag-leistungsverstarker.html
@GL
Ok It has got only one wire thikness 0,7 mm...
Thanks
Quote from: TutorialFE on 2014.04.18, 12:15:25
But I don't know which one is good... Do you have any suggest?
The one Groundloop suggested is fine or the one to the left of it.
Thanks Matt, thats interesting so i wonder if the Russian named who owns the most shares is actually Akula.
This all reminds of the the Return of the Hungarian
QuoteThe basic principle of deriving free energy from wave-fields can be understood even without higher mathematics. If we would have to summarize the essence of Mr. Vajda?s discovery in few sentences without maths, the following explanation could be given:
When two waves with identical polarization, frequency, phase and amplitude propagate in the same direction and meet (and merge) in free space, then their amplitudes will add together and the amplitude of the resultant wave will be double that of a single input wave. This physical phenomenon is called superposition or interference of the waves, when (under the above conditions) the
amplitude of the resultant wave is calculated by simply adding together the amplitudes of the incoming waves.
The energy content of a wave is directly proportional with the square of its amplitude. This fact has a profound impact on the energy balance of the wave-fields.
Calculating the energy balance of the above example, we get that if two units of energy enter the system, then the energy of the output resultant wave will be (calculated as the square of the
resultant?s amplitude, that is) four times that of one single input wave (and not only double). As we see, two units of energy enter the system and four units leave, that means we have gained two times more energy then what we have feed into it. If we take two units of energy from the output and feed it back into the input, then there are still two units remaining for utilization and the process can go on continuously.
http://www.overunity.com/3068/the-return-of-the-hungarian-free-energy-from-wave-fields/#.U1FF-aJrFZM
Quote from: T-1000 on 2014.04.18, 12:15:54
Hi,
There is new video from akula (obviously it was attempted to delete very quickly, so no original link) - https://www.youtube.com/watch?v=ogMcs6iBJwY
Good day All:
OK, the SMPS in the video says '100-240@.35a to 5vdc @ 5amps' and when he reads the LED voltage he shows 12.44vdc. So, the SMPS is ONLY powering the ICs? This is the equivalent of the 7812 reg in the 30watt design..... Where is the 12.44vdc generated??
LEDs are two parallel strings of 4 in series.....
That type of SMPS is a 50/60hz AC input..... So the output Frequency from the 2nd coil side is 50/60hz???? What is the input frequency?
If you notice also the Yellow color tampering seal on the side of the SMPS has been half way removed..... Maybe the SMPS has been modified?
ON the primary side of the coil, two of the wires on the return side are joined together at the ferrite and then return to the circuit board as one cable only.
So the Primary side of the coil has THREE wires that connect to the Circuit board. Is that wire connected back unto itself or does it connect to a copper foil????
Two ICs and a minimum of components.... maybe an integrated drive IC?
Both ICs have the numbers scraped off :'(
It is a shame the quality of the video is so blurry....
take care, peace
lost_bro
Quote from: lost_bro on 2014.04.18, 15:56:52
That type of SMPS is a 50/60hz AC input..... So the output Frequency from the 2nd coil side is 50/60hz???? What is the input frequency?
If you notice also the Yellow color tampering seal on the side of the SMPS has been half way removed..... Maybe the SMPS has been modified?
Very likely zero frequency with the device powered from a battery or super cap inside the SMPS. :(
Dear Verpies and All.
Having carried out the Capacitor mod earlier today. Please see YouTube video below.
https://www.youtube.com/watch?v=yfRb80SYAI8
Awaiting further instructions !! :)
Cheers Grum.
I finish soldering...
There are 5 little problems:
1 in the elfa order list is missing C7 100uF/35
2 the holeof big resistor are a little to close..
3 The Foil trimmer is to close to the Capacitor 2200uF 50V
4 2 more capacitor 100 nF 50V
5 Distrelec didn't ship me the toroid .. I temporary fix one similar..
Quote from: Grumpy on 2014.04.17, 22:47:32
Copper foils in that orientation will create a radial electric field perpendicular to the coil windings.
Indeed
Quote from: Grumage on 2014.04.18, 18:21:49
Having carried out the Capacitor mod earlier today. Please see YouTube video below.
It shouldn't do that.
What is on Ch2?
Quote from: TutorialFE on 2014.04.18, 18:40:24
I finish soldering...
There are 5 little problems:
1 in the elfa order list is missing C7 100uF/35
2 the holeof big resistor are a little to close..
3 The Foil trimmer is to close to the Capacitor 2200uF 50V
4 2 more capacitor 100 nF 50V
5 Distrelec didn't ship me the toroid .. I temporary fix one similar..
TutorialFE,
1. The missing 100uF is added on the corrected board design.
2. Yes but is it not a problem to have the resistors a little up in the air, better cooling.
I also made the board holes a little larger. More easy to fit the resistors.
3. I know, it depends on the type used. Your solution looks OK to me.
Made more room around the foil capacitor on the new design.
4. There should be 5 pc. of 100nF, 2 pc. of 22nF, 1 pc. of 3nF and 1 pc. of 68pF.
5. Your replacement toroid is OK.
Remember to lift the plus wire of the C4 electrolytic capacitor and solder a wire from the plus of C4 to pin 14 on the TL494!
Also, did you lift the heat sink for the BD transistor a little off the PCB so that the heat sink do not short the pads underneath?
(This is fixed on the new design.)
I also recommend NOT to solder the L1/L2 bobbin into the board. Use some wires from the board to the L1/L2 coils. This will
make it easier to do experiments.
GL.
@GL I would know also how to turn the trimmers....
Thank you for your patience
QuoteWhat about increasing R5 (Akula diagram) to say 10 Ohm? Will it not increase the feedback signal to a level where it does affect the duty cycle?
Quote from: verpies on 2014.04.18, 10:35:09
It could.
Also, changing R5 to an inductive wirewound resistor will dramatically change the signal at the "feedback takeoff point".
My R5 (1 Ohm) is a wirewound resistor, so must have some inductance.
I changed R5 to a 10 Ohm resistor, and the signal across R5 increases as expected, but no change in the pin 3 signal or duty cycle, see screenshots
first one with normal 1 Ohm R5, 2th with 10 Ohm R5.
Signals are:
gate (yellow),
voltage across R5 (blue)
signal on pin3 (purple)
current through L2 (green)
Regards Itsu
@Itsu
Thanks for the vid. Noticed that your L1 choke has few turns while the others have many turns. Is that irrelevant?
@TutorialFE
Thanks for your photos of your circuit build. I am building up my confidence to start the circuit on my own and those photos will help. Better not to solder in the E-core, plus maybe put long wires at D5 and R3 in case you need to change it out (I will have my Diode Carousel at D5).
@verpies
If the L1 choke is there for when C11 can feed it, why not add a diode across it pointing to R1. This way the starting energy can enter without overtaxing the choke.
I still say R3 is the triple wrong way. A diode there will let the L2 move to the L1 impulse only. This way you will consume much less energy plus L1 will now be able to pulse and get rid of the flux via L2 thus run more smoothly because right now it is creating a mess. Consumption would drop dramatically and then leave the field open to do the fine tuning of the system because this will allow for new resonant frequencies.
Only at that point would remain a question of wind method, turns, D5 diode choice, etc. So the wind method is still a variable as well. My money is on standard mode.
You asked about why the L1/L2 current is the same timing. I guess it will depend if the E-core was wound to T-1000 method or standard method. If T-1000 method with two layers of L2 under the L1, then L1 is hitting the L2 right away without any out of phase core flux bouncing first. The core becomes irrelevant.
Actually in the present set-up, I see the E-core as being useless. May as well send the juice directly to load. When L1 gets its pulse, it has a north and south polarity that is clear, but meanwhile the L2 is being force fed positive from R3 and positive from R2. So where do the north-south flux polarities in L2 go to create positive and negative potential difference in L2? There is no where for the L2 negative to go so it has to cancel out somewhere. If R3 was a diode, you would no longer force feed L2 with positive leaving L2 to set-up its potential difference all while lowering total power consumption.
Thanks for the diagram. Now we're takin the same language (or almost).
wattsup
Quote from: Itsu on 2014.04.18, 21:10:55
gate (yellow),
No comment ;)
Quote from: Itsu on 2014.04.18, 21:10:55
voltage across R5 (blue)
The current through the LED bank increases slightly each time the MOSFET turns off and charges up C3 through D5 due to the flyback pulse generated by L1 as the MOSFET opens.
This is normal and expected behavior of a boost converter. So far so good.
The narrow spike, at the moment the MOSFET turns OFF is EMI or is due to R5's inductance. It does not last long enough to charge up the differentiating capacitor at pin 2 and trigger the associated comparator. To trigger it, the pulse would have to be
high for a longer time and then abruptly fall. You have the abrupt fall but not the former. Also, a feedback pulse going below ground would trigger the second comparator at pin 15.
You have neither waveform feature in your feedback signal so your feedback loop is not doing anything :(
Quote from: Itsu on 2014.04.18, 21:10:55
signal on pin3 (purple)
That one is starting to annoy me. Apparently it is idle but the scope shows positive and negative spikes on it. The TL494 cannot produce negative spikes so I must assume that those spikes are an artifact due to EMI. Time for EMI countermeasures!
Quote from: Itsu on 2014.04.18, 21:10:55
current through L2 (green)
Is this channel AC-coupled?
How can the current in L2 be bidirectional if there is D6 in series with L2 ?
I can see that you did not reverse the polarity of L2 and the current in L1 flows at the same time as the current in L2 - like in a plain transformer >:(
If L2 was reversed, its current would start to increase the moment the MOSFET turned OFF.
Quote from: wattsup on 2014.04.18, 21:11:48
If the L1 choke is there for when C11 can feed it, why not add a diode across it pointing to R1. This way the starting energy can enter without overtaxing the choke.
Because such diode would provide a low impedance path to spikes generated by L2 even if this diode was pointing that direction.
L3 is not overtaxed. It is not a part of the major power loop - only the auxiliary power flows through it. Together with C13 and C6, its job is to filter out spikes from this auxiliary power rail because the sensitive TL494 is powered from it. If this choke is not doing its job, then EMI shows up at the TL494. Just look at Itsu's impossible negative spikes on pin 3.
Quote from: wattsup on 2014.04.18, 21:11:48
I still say R3 is the triple wrong way. A diode there will let the L2 move to the L1 impulse only.
No. There is already a diode in series with L2. Why can't you see it?
I redrew the diagram especially for you and swapped the position of R3 with the D6 block. Electronically that swap makes no difference because series connections are commutative... but maybe it will open your eyes graphically.
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12849)
In practical terms, the closing of the 'STOP' switch causes an increase in supply current and decrease in LED lumen level, irrespective of coil phasing, so I can see no advantage in having L2 in circuit.
Hoppy
I noticed something strange in the Pcb board...
Groundloop tracks so thin can cause problems in electronics???
Hi,
Here is higher quality video from akula - https://www.youtube.com/watch?v=7V8vC0UL-XY
There he showed PCB connnections and some measurements as well.
Cheers!
Quote from: wattsup on 2014.04.18, 21:11:48
@Itsu
Thanks for the vid. Noticed that your L1 choke has few turns while the others have many turns. Is that irrelevant?
Well, i copied it from the original picture which had 7 turns on a similar sized core.
As i have it connected via a PCB connector i can swap it out easely which i did with severall other (more turns) chokes, but no real difference is noted.
Regards Itsu
@ T-1000 I think we are close to the replica .....
Quote from: T-1000 on 2014.04.19, 09:16:18
Here is higher quality video from akula
There he showed PCB connnections and some measurements as well.
He clearly wants to be a teacher and is pretty good at it, especially the one handed camera work.
What do we need to do to form a classroom and let Akula be the instructor until we all understand and can do it ourselves?
http://www.youtube.com/watch?v=pUW-3yXQIWQ
Seems he tries pretty hard to teach us what is needed to make this work.
Quote from: verpies on 2014.04.18, 21:38:50
Quotevoltage across R5 (blue)
QuoteYou have neither waveform feature in your feedback signal so your feedback loop is not doing anything :(
How is that with other builders? Is the feedback loop also not doing anything?
Quotesignal on pin3 (purple)
QuoteThat one is starting to annoy me. Apparently it is idle but the scope shows positive and negative spikes on it. The TL494 cannot produce negative spikes so I must assume that those spikes are an artifact due to EMI. Time for EMI countermeasures!
Ok, where is my stack of 0.1uF caps......
Quotecurrent through L2 (green)
QuoteIs this channel AC-coupled?
How can the current in L2 be bidirectional if there is D6 in series with L2 ?
AC coupled would display a little sine wave symbol after the BW (bandwidth reduced to 150MHz) symbol, so no, not AC coupled
I measured D6 and it seems to function correctly :D
EDITTED: its an AC current probe, guess this would ALWAYS show AC only, Right?
QuoteI can see that you did not reverse the polarity of L2 and the current in L1 flows at the same time as the current in L2 - like in a plain transformer >:(
If L2 was reversed, its current would start to increase the moment the MOSFET turned OFF.
Well that could be because the way i insert the current probe into the circuit (when i turn it 180° it will show the opposite)
Notice that i sometimes invert this current channel to what i think would be good (current flowing when gate signal is high), but thinking about it that would not be the correct thinking in ALL situations, so i have to come up with a solid solution on how to insert the current probe into the circuit.
Regards Itsu
Quote from: TutorialFE on 2014.04.19, 09:14:12
Groundloop's tracks are so thin, that they can cause problems in electronics???
The thin darker lines are the gaps between tracks...not the tracks themselves.
Quote from: T-1000 on 2014.04.19, 09:16:18
Hi,
Here is higher quality video from akula - https://www.youtube.com/watch?v=7V8vC0UL-XY
There he showed PCB connnections and some measurements as well.
Cheers!
It looks to me like the Kapanadze 9V battery latching circuit trick. The short battery pulse produces HV from the coil into the SMPS which latches a switch to supply battery (in SMPS) current back to the device - simple! He is relying on the Kapanadze 9V battery ruse to convince us that he has achieved self-running. :(
Hoppy
I will from now on use the below current probe setup when measuring the L1 / L2 currents.
I adjust the L1 current screenshot to show positive current when the gate is positive and use the same setting for the L2 current.
That way both should show the actual direction of the current in relation to each other.
Regards itsu
Quote from: T-1000 on 2014.04.19, 09:16:18
Here is higher quality video from akula - https://www.youtube.com/watch?v=7V8vC0UL-XY
There he showed PCB connnections and some measurements as well.
I will let Hoppy have a go at it.
I predict that he will take it apart because Akula did not take the AC/DC-DC converter apart.
Maybe even the phrase "hidey hole" will be resurrected.
Quote from: Hoppy on 2014.04.19, 08:31:51
In practical terms, the closing of the 'STOP' switch causes an increase in supply current and decrease in LED lumen level, irrespective of coil phasing, so I can see no advantage in having L2 in circuit.
The conventional circuit analysis supports this conclusion.
However if there is any unconventional magic* to be had in this system then it will manifest itself exactly across this L2 as a high power spike that will recharge C11 through D6 ...and most importantly, such L2 pulse will cause an abrupt decrease the voltage at the 'feedback takeoff point".
If you were following my conventional circuit analysis, TL494 feedback loop is configured to react exactly to such abrupt voltage decrease at this point. AFAIR: -dV/dt > -6V/ms
* e.g. magnetoacoustic pulse or beta current pulse due to NAR or NMR, etc...
Quote from: verpies on 2014.04.19, 10:59:26
The conventional circuit analysis supports this conclusion.
Howeve,r if there is any unconventional magic* to be had in this system then it will manifest itself exactly across this L2 as a high power spike that will recharge C11 through D6 ...and most importantly, such L2 pulse will cause an abrupt decrease the voltage at the 'feedback takeoff point".
If you were following my conventional circuit analysis, TL494 feedback loop is configured to react exactly to such abrupt voltage decrease at this point. AFAIR: -dV/dt > -6V/ms
* e.g. magnetoacoustic pulse or beta current pulse due to NAR or NMR, etc...
I think its high time we all wise-up to this Akula chap!!
Quote from: Hoppy on 2014.04.19, 11:09:16
I think its high time we all wise-up to this Akula chap!!
I don't think this reply was intended as a response to
this post (http://www.overunityresearch.com/index.php?topic=2358.msg37843#msg37843).
This reply is befitting more of
this post (http://www.overunityresearch.com/index.php?topic=2358.msg37842#msg37842).
Quote from: Itsu on 2014.04.19, 09:46:52
How is that with other builders? Is the feedback loop also not doing anything?
Their circuits do not work like yours. There is something wrong with them.
I think this will change once they do the C4 (C8 Akula) fix properly.
Quote from: Itsu on 2014.04.19, 09:46:52
its an AC current probe, guess this would ALWAYS show AC only, Right?
Right.
I thought that your current probes were Hall-based and capable of sensing DC current.
Too bad :(
You can always go back to using R3 and R1 (with C2 removed) as CSRs (if they are not inductive), but then you'll need to keep your probe ground clips on the positive terminal of C11.
Except for the L2 polarity, I think that your circuit functions correctly. IMO L2 must be connected in such a way that its current increases when the MOSFET is on the OFF state.
The reason you don't have the proper signal at the "feedback takeoff point" is because L2 is not generating strong enough pulses or is not connected properly.
Those pulses are the unconventional magic pulses that are supposed to be generated by a properly wound transformer over an appropriate core.
Take a look at
this (http://falstad.com/circuit/#%24+1+4.9999999999999996E-6+0.7389056098930651+41+12.0+43%0A162+416+176+416+208+1+3.0+1.0+1.0+1.0%0A162+416+208+416+240+1+3.0+1.0+1.0+1.0%0A162+416+240+416+272+1+3.0+1.0+1.0+1.0%0A162+416+272+416+304+1+3.0+1.0+1.0+1.0%0A162+416+304+416+336+1+3.0+1.0+1.0+1.0%0Ar+416+336+416+384+0+1.0%0Ag+208+384+208+416+0%0Ax+219+193+245+197+0+14+C11%0Ax+455+246+473+250+0+14+C3%0Aw+208+240+208+384+0%0Av+464+80+464+48+0+0+40.0+100.0+0.0+0.0+0.5%0Aw+416+384+448+384+0%0Aw+416+176+448+176+0%0Aw+448+176+448+224+0%0Aw+448+224+496+224+0%0Aw+560+224+528+224+0%0Av+560+288+560+224+0+0+40.0+15.0+0.0+0.0+0.5%0Ar+448+288+560+288+0+0.001%0Ar+96+240+208+240+0+0.001%0Av+96+240+96+176+0+0+40.0+15.0+0.0+0.0+0.5%0Aw+128+176+96+176+0%0Aw+160+176+208+176+0%0Ac+208+176+208+240+0+0.002+15.000430638948425%0Ad+448+176+448+128+1+0.6%0Aw+416+176+384+176+0%0Ac+384+176+384+336+0+1.0000000000000001E-7+14.34295474314519%0Aw+384+336+416+336+0%0As+496+224+528+224+0+1+true%0As+128+176+160+176+0+1+true%0Aw+448+288+448+384+0%0Ax+44+195+82+199+0+16+%2B15V%0Ax+569+242+607+246+0+16+%2B15V%0Ax+464+153+485+157+0+16+D6%0Ax+486+68+529+72+0+16+-100V%0Aw+464+48+464+32+0%0Aw+432+48+432+32+0%0Aw+464+32+432+32+0%0Aw+432+32+320+32+0%0Aw+208+176+208+32+0%0A160+448+128+448+96+1+0.001+1.0E10%0Aw+432+48+432+96+0%0Aw+464+80+464+96+0%0As+304+112+336+112+0+1+true%0AR+304+112+272+112+0+0+40.0+3.0+0.0+0.0+0.5%0Ax+385+364+403+368+0+14+R5%0Ax+257+56+275+60+0+14+R3%0Ar+208+32+320+32+0+1.0%0Ax+483+45+548+49+0+14+EMF+of+L2%0Ax+221+93+424+97+0+14+Push+to+simulate+magic+L2+pulse%0Aw+432+112+336+112+0%0Ax+70+153+196+157+0+14+Push+to+charge+C11%0Ax+463+202+581+206+0+14+Push+to+charge+C3%0Ac+448+224+448+288+0+0.002+14.586591801813986%0Aw+416+384+384+384+0%0Aw+384+384+208+384+0%0Ax+294+230+366+234+0+14+LED%27s+stray%0Ax+292+246+368+250+0+14+capacitance%0Ax+313+265+337+269+0+14+and%0Ax+304+279+352+283+0+14+reverse%0Ax+288+294+373+298+0+14+recovery+time%0Ao+5+8+0+289+0.29230032746618057+1.4965776766268446+0+-1%0Ao+52+8+0+290+20.0+0.2+0+-1%0A) abridged simulation of the major power loop.
The current in R5 can rapidly fall and even reverse only and only if L2 somehow generates a strong negative EMF.
This EMF will charge up C11 through D6 and will rapidly discharge C3, even to the point of reversing C3's voltage.
Push the momentary switch on the simulation to simulate the "magic pulse" occuring in L2 and observe how quickly the current falls and even reverses in R5. (you can increase LED's stray capacitance to make that negative current pulse larger)
The whole circuit (including the TL494 feedback loop) seems to be designed to expect powerful negative pulses from L2.
According to conventional electronic analysis, L2 will never generate such a pulse ...but according to McFreey, Gunderson, etc... it just might.
Anyway, that's what the circuit is waiting for.
Quote from: verpies on 2014.04.19, 10:20:12
The thin darker lines are the gaps between tracks...not the tracks themselves.
I've missed the right word... sorry.... I've replace the picture with the right description...
Quote from: TutorialFE on 2014.04.19, 14:05:53
I've missed the right word... sorry.... I've replace the picture with the right description...
At these voltages there is no risk of an arcing across those varnished gaps. If you notice any arcing between tracks, notify Groundloop immediately.
The increased inter-track capacitive reactance is negligible at these frequencies.
It is easier to manufacture printed circuits with wider gaps. That's probably why the Chinese do it in their "normal circuits".
@verpies SORRY-SORRY AND SORRY AGAIN.
First, let's now put the main skeleton on the table. Our only reference of this circuit working is via x-name41. No video of Akula was ever shown with this circuit working. If x_name41 made his video based on the original Akula 30 watts circuit diagram, then it did not work in loop and that is why he faked his video probably because he was so pissed off in following the circuit to the letter and not showing looping. So he spiked his circuit with two feed wires from underneath. I know he faked it and there is solid proof. So x_name41 was not able to self-run it, only fake it. I am convinced that the problem is R3 and he could not figure it out himself. Probably if he is reading this forum, he will do the change himself on his circuit. This does not mean the Akula device is fake, only that there are mistakes in the diagram and we need to work them out.
I really appreciate your extra effort on my behalf and am sorry to be so repetitive. A real pain I know, but then you or anyone else would need to answer these questions just to make quadruple sure when looking at your original diagram. I am sure some others are asking themselves the same thing.
1) Why would you accept to pass dc + directly through R3 - then L2 - then R2- then to load, continuously? This is like a straight line to load and since load is at the end of the line, the DC+ entering R3 is still positive when it enters L2 and still positive when exiting L2 and still positive when entering R2 so still positive when exiting R2 and when hitting the load. L2 is always positive going to load.
You can pulse L1 as much as you want and L2 bleed will not change. You can switch the R2/R3 around with diode and cap (like on your second diagram you made for me) but that does not change this continuous DC bleed to load at all times. @Itsu said his R3 was really hot? That's not a good path to OU by wasting energy in a resistor, especially when it is not required.
2) Then how do you expect to refill C11 if L2 is so busy passing DC to load? You are saying D6 is passing a high voltage spike to L2 then R3 then to C11. But why would you want to more resistance in this path when it is not required. In any case any spike through D5 will hit the load (path of least resistance) before it can pass L2 and R3. So that logic is not fluent enough to work.
3) Then why do you think L1 is not only going to the Mosfet Q1, but just before that it is bypassed by D5 and sent to load as well? This part is OK and for me this is the only main out-of-the-box feature of the Akula circuit. Instead of feeding the load with L2, he is feeding the load with L1 via D5. That's all he needs to do to feed the load. He then must be using the free L2 to replenish the C11. There is no other way. But L2 cannot replenish C11 and feed the load at the same time because it cannot respond to the L1 pulse off the E-core because L2s' internal polarity is always held positive by R3.
4) How do you think the mosfet is influencing what L1 is sending to the load? The pulsed mosfet is useless because pulsing L1 is useless as per the diagram. You pulse a coil to impart to the another coil otherwise you are stuck with overflux in the core, but if the other coil is already full of DC, then how do you expect it to react? It cannot.
5) What adverse effect would you anticipate if R3 was replaced by a diode? Is such a change going to risk blowing the circuit? I can't see how it could.
For me, when I consider all of these conditions as the diagram is draw presently, none of it makes sense if R3 is not replaced with a diode pointing to C11. Nothing else, just a diode at that point prohibits DC+ from entering the top of L2 but provides a clear and simple path so L2 can refill C11 ASAP. If you fill C11 fast enough you can have a self-runner but you need an available and receptive L2 to accomplish that.
The diode D6 can still be there and there is no problem having two diodes one on each side of L2 because they both react differently depending on the polarities and intensities present in L2. My opinion on the D6 diode is it is never really used except if L1 really created some very high spikes that could pass the D6 and enter the L2 but if the coils are well tuned, that should never occur. The other reasons is if D6 is held in reverse threshold mode by extreme changes in L2. But those extremes cannot occur if R3 is holding it hostage.
Just having that one D6 is not the same thing as also replacing the R3 with a diode because the source of polarity changes is supposed to originate from inside L2 and not something that can be controlled by D6. If you check D6 temperature, it is probably not hot at all because it is not being used for much.
Look I know I may risk ("risk" huh more like "for sure") coming off as the resident shmuck of the forum, being such a pain and all. But I think everyone should stop and carefully answer the above questions. There is no point to continue if this is not resolved in a logical manner. We seem to think it is OK to force feed through L2 then to load, I do not think, well actual I know, that if L2 is feeding the load as it is doing, then impulses from L1 to L2 are useless. L2 is just a conductor. Add the fact that some are using the T-1000 wind and this is compounded even greater.
Maybe asked again, why is R2 present in the circuit. For me it is if R3 was a diode then when C11 is full and cannot take anymore, if the stress in L2 increases further it will dump more and more to load instead of over saturating L2 and killing the overall effect. It's a question of balance but you cannot achieve that balance if one side is tipped way over the scale as R3 is doing now.
I won't bring this up again but I will make my own circuit "eventually" and do it my way.
wattsup
PS: This post took 3 hours to prepare (last night and this morning) because I wanted to make sure I did not vex you since I hold you in top regard and I consider this part of my duty to point out something so wrong. Akula is not infallible and I would not be surprised if he put R3 there (instead of a diode) on purpose to see who will figure it out. That fact that he is not coming forward any more openly is his way of saying "There is no free lunch man". Use your brains and figure it out. His way of saying, "if you can find the problem, then you have proven to me that you have finally learned well what the circuit is doing".
Thanks for your insights, wattsup.
NOTE -- I have sent "Akula circuit boards" to Mexico (first class mail), and to California and Hawaii (Priority Mail); on their way!
@ Wottsup,
I've tried your diode idea and it makes no difference to the observable operation of the circuit for the reasons that Verpies has already pointed out to you.
Don't take our word for it, build it yourself and see.
Happy
Quote from: wattsup on 2014.04.19, 14:57:45
This does not mean the Akula device is fake, only that there are mistakes in the diagram and we need to work them out.
We have to be objective and consider that possibility.
There are several Akula's 30W perpetual flashlight schematic diagrams in circulation:
Diagram v1.0 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12307)
Diagram v2.0 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12686) Diagram v2.1 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12797)
Diagram v2.2 (http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/dlattach/attach/136959/)
Diagram v3.0 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12688)The Diagram v1.0 does not match the functionally of other diagrams because in v1.0 the pulses to the the MOSFET's gate are inverted and because of this the duty cycle can vary only between 53% and 100% while the TL494 feedback loop is positive.
In all other versions gate pulses are not inverted and the duty cycle can vary between 0% and 47% while the TL494 feedback loop is negative.
That allegedly
successful replication (https://www.youtube.com/watch?v=9zilk7itV6k) was done according to diagram v2.0
v2.0 is functionally and electronically equivalent to v2.1 and v2.2 (the differences between them are only graphical).
v2.0 is also functionally equivalent to v3.0 (but not electronically).
v2.0 is functionally and electronically
different from v1.0
The point of the above is that indeed all of these schematics cannot be correct, because they are not the same and one is just the opposite of the others.
Quote from: wattsup on 2014.04.19, 14:57:45
Why would you accept to pass dc + directly through R3 - then L2 - then R2- then to load, continuously?
No, because R2 is practically an open circuit (420kΩ).
To transfer 30W across such resistance would require 3550 Volts. None of the components would be able to withstand that much voltage.
Quote from: wattsup on 2014.04.19, 14:57:45
...the DC+ entering R3 is still positive when it enters L2 and still positive when exiting L2
Not if L2 generates an EMF that is opposite and greater than the voltage across C11.
When -V
L2 > V
C11 then V
C11 + V
L2 < 0 (negative).
BTW: D6 conducts when V
C11 + V
L2 < V
C3.
See
this (http://falstad.com/circuit/#%24+1+4.9999999999999996E-6+0.7389056098930651+41+12.0+43%0A162+416+176+416+208+1+3.0+1.0+1.0+1.0%0A162+416+208+416+240+1+3.0+1.0+1.0+1.0%0A162+416+240+416+272+1+3.0+1.0+1.0+1.0%0A162+416+272+416+304+1+3.0+1.0+1.0+1.0%0A162+416+304+416+336+1+3.0+1.0+1.0+1.0%0Ar+416+336+416+384+0+1.0%0Ag+208+384+208+416+0%0Ax+219+193+245+197+0+14+C11%0Ax+455+246+473+250+0+14+C3%0Aw+208+240+208+384+0%0Av+464+80+464+48+0+0+40.0+100.0+0.0+0.0+0.5%0Aw+416+384+448+384+0%0Aw+416+176+448+176+0%0Aw+448+176+448+224+0%0Aw+448+224+496+224+0%0Aw+560+224+528+224+0%0Av+560+288+560+224+0+0+40.0+15.0+0.0+0.0+0.5%0Ar+448+288+560+288+0+0.001%0Ar+96+240+208+240+0+0.001%0Av+96+240+96+176+0+0+40.0+15.0+0.0+0.0+0.5%0Aw+128+176+96+176+0%0Aw+160+176+208+176+0%0Ac+208+176+208+240+0+0.002+15.000430638948425%0Ad+448+176+448+128+1+0.6%0Aw+416+176+384+176+0%0Ac+384+176+384+336+0+1.0000000000000001E-7+14.34295474314519%0Aw+384+336+416+336+0%0As+496+224+528+224+0+1+true%0As+128+176+160+176+0+1+true%0Aw+448+288+448+384+0%0Ax+44+195+82+199+0+16+%2B15V%0Ax+569+242+607+246+0+16+%2B15V%0Ax+464+153+485+157+0+16+D6%0Ax+486+68+529+72+0+16+-100V%0Aw+464+48+464+32+0%0Aw+432+48+432+32+0%0Aw+464+32+432+32+0%0Aw+432+32+320+32+0%0Aw+208+176+208+32+0%0A160+448+128+448+96+1+0.001+1.0E10%0Aw+432+48+432+96+0%0Aw+464+80+464+96+0%0As+304+112+336+112+0+1+true%0AR+304+112+272+112+0+0+40.0+3.0+0.0+0.0+0.5%0Ax+385+364+403+368+0+14+R5%0Ax+257+56+275+60+0+14+R3%0Ar+208+32+320+32+0+1.0%0Ax+483+45+548+49+0+14+EMF+of+L2%0Ax+221+93+424+97+0+14+Push+to+simulate+magic+L2+pulse%0Aw+432+112+336+112+0%0Ax+70+153+196+157+0+14+Push+to+charge+C11%0Ax+463+202+581+206+0+14+Push+to+charge+C3%0Ac+448+224+448+288+0+0.002+14.586591801813986%0Aw+416+384+384+384+0%0Aw+384+384+208+384+0%0Ax+294+230+366+234+0+14+LED%27s+stray%0Ax+292+246+368+250+0+14+capacitance%0Ax+313+265+337+269+0+14+and%0Ax+304+279+352+283+0+14+reverse%0Ax+288+294+373+298+0+14+recovery+time%0Ao+5+8+0+289+0.29230032746618057+1.4965776766268446+0+-1%0Ao+52+8+0+290+20.0+0.2+0+-1%0A) abridged simulation of the major power loop.
Do you think that your remaining questions are valid in light of these objections above?
Dear Verpies.
I have had a busy day today so I have only now got back to the PCB.
I still have the same situation with regards to Video 2 with R6 (one side lifted) from the board.
Thanks to a.king21 I now have a compliment of 25 1 W LED's connected to the output.
There is obviously something wrong with my electronics !! Do you have any suggestions ??
Cheers Grum.
Evening Grum,
How are your LED chips arranged? There needs to be no more than 5 in each series string.
Hoppy
Quote from: Hoppy on 2014.04.19, 18:03:17
Evening Grum,
How are your LED chips arranged? There needs to be no more than 5 in each series string.
Hoppy
Good evening Hoppy.
Yes you are correct !! I now have 5 strips of 1 W LED's, 5 per strip and 5 in parallel.
Cheers Grum.
Quote from: Grumage on 2014.04.19, 17:50:58
I still have the same situation with regards to Video 2 with R6 (one side lifted) from the board.
Well did you pull out +C4 and connect it to R5 (next to pins 13 & 14) ?
Quote from: Grumage on 2014.04.19, 17:50:58
There is obviously something wrong with my electronics !! Do you have any suggestions ??
Take out one side of R6 and let me know if the output of the TL494 is still inhibited when you increase the supply voltage above 8.6VDC.
Quote from: verpies on 2014.04.19, 14:01:16
You can always go back to using R3 and R1 (with C2 removed) as CSRs (if they are not inductive), but then you'll need to keep your probe ground clips on the positive terminal of C11.
Except for the L2 polarity, I think that your circuit functions correctly. IMO L2 must be connected in such a way that its current increases when the MOSFET is on the OFF state.
OK, i used R3 (measured 1 Ohm) and R1 (measured 0.2 Ohm) as CSR's (i have removed C2).
The both ground leads of the probes are at the C11 plus side, but probe and ground lead are as close to the resistors as possible to exclude any extra inductance.
The both channels 1 and 2 are inverted to show the correct current direction.
I have the current probe first in L1 (first screenshot), then L2 (2th screenshot) to compare with the currents through the resistors.
Probe situated as mentioned in my last post.
Data:
1st screenshot
L1 (current probe) 6.1A * 2 = 12.2A
R1 (yellow) 3.6V / 0.2 Ohm = 18A (for voltages i did NOT take the value in the boxes (as they have some spikes), but the signal in the division divides)
R3 (blue) 8.5V / 1 Ohm = 8.5A
2th screenshot
L2 (current probe) 3.56A * 2 = 7.12A
R1 (yellow) 3.6V / 0.2 Ohm = 18A
R3 (blue) 8.5V / 1 Ohm = 8.5A
Regards Itsu
Quote from: lost_bro on 2014.04.17, 23:16:06
As a side note, I remember one of the Russian members at O.U.com mentioning something about the capacitor needing to be on the opposite side of the L1 (choke). I will have to look for that info.....
When I moved the capacitor to the *other* side of the L1 (choke) I noticed that the simulations in LTspice showed quasi resonant condition.... I have attached the LTspice file if anyone is interested.
take care, peace
lost_bro
I did just that, but besides a cleaner drain voltage signal (nice square wave, so no more jump half way) no noticeable other effects are seen.
Regards Itsu
@verpies
Thanks again. I totally forgot about the 420k value of R2.
I know what to do now. I will simply work on the E-core driving method and just dispense with the all the circuitry for the frequency control. So I will make the coils and the other particulars as shown below, wind a tuned primary and secondary and do all my tests with my FG having full control to better understand what is going on in this device.
Something is wrong and I will find it. hehehe
@Happy
Thanks for trying the diode at R3. Listen man, it is too complicated for me to explain so while you guys work out the circuit end I will do it myself and prepare the tuning of the coils and set-up a procedure for testing and report back. This part, the hunting part is my specialty and I would have to be standing behind you to explain everything and even then the circuit frequency drive limitations would just drive me crazy. Better to work the output end with full control first, then to bring it into a stand alone. With the set-up below, if I hit a good effect I should be able to remove the battery and make it run even with the FG in place, the point would be made.
@verpies
One last question, I promise.
L2 needs a ground somewhere otherwise what does L2 do when the primary is pulsed, L2 will develop a voltage differential, so where the does negative go if both sides of L2 are positive. Is it possible that D6, R2 and C4 are arranging a type of floating ground for L2?
This won't change the my test method but it is the only other question that I have not manged to resolved for myself.
wattsup
Quotesignal on pin3 (purple)
Quote from: verpies on 2014.04.18, 21:38:50
That one is starting to annoy me. Apparently it is idle but the scope shows positive and negative spikes on it. The TL494 cannot produce negative spikes so I must assume that those spikes are an artifact due to EMI. Time for EMI countermeasures!
zooming in on this:
green trace is pin 3
purple trace is pin 2
blue trace is junction R12 and C7
yellow trace is gate signal
What i see is ONLY negative pulses on pin 3, not (hardly) on pin 2 or the junction R12/C7
To me this means that these negative pulses are coming from pin 3 (TL494), right?
Regards Itsu
Quote from: wattsup on 2014.04.19, 21:04:25
L2 needs a ground somewhere otherwise what does L2 do when the primary is pulsed? Will L2 will develop a voltage differential?
Yes.
Note that "voltage differential" does not necessarily result in current flow. (e.g. when the circuit is open or a diode blocks the flow).
Quote from: wattsup on 2014.04.19, 21:04:25
so where the does negative go if both sides of L2 are positive. Is it possible that D6, R2 and C4 are arranging a type of floating ground for L2?
"Ground" can be assumed at an any point. You can pick and choose.
You could assume the cathode of D6 as the ground because that choice is arbitrary and inconvenient.
For this circuit it is much more convenient to assume the negative terminal of C3 as the ground.
When L2 develops a voltage differential between its terminals, it will behave as a floating battery in series with the C3 capacitor. Depending on the polarity of this "differential", L2's voltage (V
L2) will add or subtract with the voltage across C3 (V
C3) just like the voltage of two different batteries connected in series.
If V
C3 plus V
L2 becomes greater than the voltage across C11 (V
C11) then the diode D6 will allow current to flow ...and if V
C3 plus V
L2 remains less then V
C11 then D6 will block the current.
From the point of view of V
C3 + V
L2 the V
C11 is a parallel voltage source. As you probably know, no current flow is possible between two parallel voltage sources of equal voltage.
"Normal" transformers work like gears and their primary and secondary currents flow simultaneously in opposite directions. (their magnetic coupling coefficient is close to unity, too)
"Flyback" transformers have diodes in series with their secondary windings that prohibit such simultaneous flows of currents. In such transformers primary and secondary currents flow alternately (non-simultaneously) in the same direction.
I believe that the transformer in this Akula's circuit was designed to operate in the "flyback" mode.
Quote from: wattsup on 2014.04.19, 21:04:25
I will simply work on the E-core driving method and just dispense with the all the circuitry for the frequency control. So I will make the coils and the other particulars as shown below, wind a tuned primary and secondary and do all my tests with my FG having full control to better understand what is going on in this device.
That's fine. The video mentioned in
this message (http://www.overunity.com/14524/3v-ou-flashlight/msg398591/#msg398591) might show you waveforms that you might find helpful in hunting for the magic effect.
As always, the
construction details (http://youtu.be/bYEuV-ji4tk?t=1m6s) and core of the transformer are of paramount importance.
Quote from: Itsu on 2014.04.19, 19:13:46
OK, i used R3 (measured 1 Ohm) and R1 (measured 0.2 Ohm) as CSR's (i have removed C2).
The both ground leads of the probes are at the C11 plus side, but probe and ground lead are as close to the resistors as possible to exclude any extra inductance.
The both channels 1 and 2 are inverted to show the correct current direction.
R1 (yellow)
R3 (blue)
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12875)
You still have the L1 and L2 currents reaching maxima and zeros
simultaneously. That's a characteristic current transfer in a "normal transformer". That transformer mode is wrong in my opinion.
For a "flyback transformer" mode (the correct one IMO), L2 should be reversed and these waveforms should look like this:
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12891)
Ch1 (yellow) : Gate voltage in reference to ground
Ch2 (blue) : Voltage across R1, with C2 removed (L1 Current)
Ch3 (green) : Voltage across R3 (L2 Current)
Quote from: Itsu on 2014.04.19, 21:14:13
green trace is pin 3
purple trace is pin 2
blue trace is junction R12 and C7
yellow trace is gate signal
What i see is ONLY negative pulses on pin 3, not (hardly) on pin 2 or the junction R12/C7
To me this means that these negative pulses are coming from pin 3 (TL494), right?
I would not be so sure.
The amplitude of positive spikes on pin 2 is 4x higher than on pin 3.
Do you have these spikes on the supply pins of the TL494 (pins 7, 12, 14), too ?
Quote from: verpies on 2014.04.19, 23:25:19
You still have the L1 and L2 currents reaching maxima and zeros simultaneously. That's a characteristic current transfer in a "normal transformer". That transformer mode is wrong in my opinion.
For a "flyback transformer" mode (the correct one IMO), L2 should be reversed and these waveforms should look like this:
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12891)
Ch1 (yellow) : Gate voltage in reference to ground
Ch2 (blue) : Voltage across R1, with C2 removed (L1 Current)
Ch3 (green) : Voltage across R3 (L2 Current)
Ok, but what does that mean? Does it mean the design of this circuit is wrong, or did i make a measurement/connection mistake somewhere?
The difference between a normal transformer and a flyback transformer is the diode in the secondary, right?
Well we do have one in the form of D6, so why does it not work in flyback transformer mode?
Thanks, regards Itsu
Quote from: verpies on 2014.04.19, 23:51:30
I would not be so sure.
The amplitude of positive spikes on pin 2 is 4x higher than on pin 3.
Do you have these spikes on the supply pins of the TL494 (pins 7, 12, 14), too ?
I thought that the negative spikes are an indication of EMI, those are not present on pin2 and the junction.
Let me check tonight on spikes on the supply pins, i did that before, and found them almost clean, but i will check again.
Regards Itsu
If I can give my opinion as electronic spin doctors not so ...
I see the project 30w very difficult to replicate so you do not even know if it works ...
maybe it was an idea that remained on paper and a real simulation of the circuit but never swich on....
Xname41 .... it seems to me that at the beginning of his video says exatly "IDIOTS" so I think it's definitely a fake replica also because it has not posted any videos or other replication schemes ...
In my opinion the 30w project replication should be left aside and focus on the latest invention of 8 LEDs and try to replicate that ....
it seems to me that Akula has project it right to be replicated by all.... hoping that gives the circuit diagram ...
You are more experts about me and decide if it would appropriate open a new tread on this new project called????
Quote from: Itsu on 2014.04.20, 10:15:24
Ok, but what does that mean? Does it mean the design of this circuit is wrong, or did i make a measurement/connection mistake somewhere?
I don't think you've made a wrong measurement. Maybe you made a relative current measurement error before, but not this time. You just can't go wrong with two CSRs sharing a common junction.
Quote from: Itsu on 2014.04.20, 10:15:24
The difference between a normal transformer and a flyback transformer is the diode in the secondary, right?
Yes, and...
- the direction of the secondary current is different relative to a primary current compared to a "normal" transformer.
- the magnetic coupling coefficient is lower than in a "normal" transformer. (usually accomplished by a gap in the core)
Quote from: Itsu on 2014.04.20, 10:15:24
Well we do have one in the form of D6, so why does it not work in flyback transformer mode?
Because the L2 connection is reversed...or diodes are backwards or conducting in reverse.
Tutorial_FE
I'm not sure you fully understand, this is not an XName design, it's an Akula design that he has shown to be free running, i would ask you to refrain from negative posts, we are here because we are trying to see if we can get it working, by throwing expertise at it and to try and uncover Akula's secret.
We work as a team here, there are no negative comments, only conclusions and suggestions, if you feel that way then stop working on it and work on something else, Free energy research comes with failure as part of the process of learning, most here are learning while enjoying what they are doing, who knows a breakthrough maybe gained in the process, and certainly everyone can carry the knowledge gained into future projects.
Now if you want me to start a thread on the device you have shown then i can do that, PM me with a title and i will open a thread and move your post into it.
Personally i think so far i have never seen so many people joining together to try and understand such a device, hopefully it has been enjoyable and a valuable experience for most so far.
@Peterea I'm not negative .... I was happy for new device...
Dear Peterae.
I agree with your sentiments 100%. I am learning quicker than I have ever done !! It is with thanks to the members here at OUR, their patience akin to that of Job himself !! :)
I am now going to tax them even further now !!
Dear Verpies, with reference to our conversation in chat last night. I have dismounted and checked each resistor in the potential divider, all are correct. I still am unable to lower the voltage.
Do you have any suggestions. Please be aware that R 6 is still lifted on one side whilst these tests were carried out !!
Cheers Grum.
Quote from: TutorialFE on 2014.04.20, 14:00:02
@Peterea I'm not negative .... I was happy for new device...
I'm interested in Akula's new device also. If he posts schematic, definitely start thread for it. O0
Quote from: wattsup on 2014.04.19, 21:04:25
@Happy
Thanks for trying the diode at R3. Listen man, it is too complicated for me to explain so while you guys work out the circuit end I will do it myself and prepare the tuning of the coils and set-up a procedure for testing and report back. This part, the hunting part is my specialty and I would have to be standing behind you to explain everything and even then the circuit frequency drive limitations would just drive me crazy. Better to work the output end with full control first, then to bring it into a stand alone. With the set-up below, if I hit a good effect I should be able to remove the battery and make it run even with the FG in place, the point would be made.
wattsup
@ Wottsup,
That's great man, thinking outside the box on the coil design might just do the trick. I look forward to your construction details for a self-running coil design. My setup is at the ready to try any suggested coil design, assuming
full constructional details are provided. However, it would not be wise at this stage to assume that there is any error in Akula's schematics insofar as what is necessary for self-running, given a fit for purpose coil design. Its only the coil design details that have not been given to us.
Happy :)
Quote from: 4Tesla on 2014.04.20, 15:29:37
I'm interested in Akula's new device also. If he posts schematic, definitely start thread for it. O0
Its what's inside that SMPS box that interests me. ;)
QuoteI'm interested in Akula's new device also. If he posts schematic, definitely start thread for it. Afro
No problem guys, just PM me a title and i will start one when we need to.
Tutorial_FE thanks for clarifying your position O0
Quote from: Grumage on 2014.04.20, 14:13:36
Dear Verpies, with reference to our conversation in chat last night. I have dismounted and checked each resistor in the potential divider, all are correct. I still am unable to lower the voltage. Do you have any suggestions.
Check that R14 and R15 are connected according to the diagram below and have resistances that are written next to them on this diagram.
If still no luck, than make a high resolution photo (or scan) of the components and traces marked in red (put R14, R15 and R6 back into the board for the photo)
Component designations are in accordance with Groundloop's PCB.
The Sunday Akula Review - hahaha
@Itsu (but also @Grum - @Happy)
About the spikes, where is your copper strip grounded? On the diagram it shows the copper strip grounded at C3. I don't know if that would make any difference when this strip will receive some static discharges from the the shock treatment of the Ecore. Maybe we need to add something between the copper strip and ground to kill or attenuate the spike before it hits the ground since in essence this is a floating ground system? Or send the copper strip connection to a real ground off the circuit to see if the spike will go away. Don't know.
Also, in most of Akulas' devices he has the transformer off and away from the board. Maybe you need to extend your 5 transformer wires to place it further away from the board since maybe its air static from L1/L2 towards the L1 choke then feeding into the VR1. Or as you are pulsing, move the transformer to see if there are any angles that influence the spike. Or place a metal plate between the transformer and the board to see if more shielding makes a difference just to see if the source is the transformer via air.
Also, for all builders, please remind us what type of winding you have, T1000 or standard mode. It may be good to eventually make up a build model number that builders can include in each of their posts (or temporarily in their signature) so we always know the device type and not have to ask ourselves this question after a few more pages. hahaha Or put a URL in your signature where you have posted a build summary.
Maybe something like.........
Four letter model designation: AK30 - AK01 - AK??
Characters 1 and 2 for circuit board type: GL = @Groundloop - SH = Shark-01 - FB = Free board - others
Characters 3 and 4 for diagram type: ???????? to be determined by @verpies.
Characters 5 and 6 for transformer type: EC = Ecore - CC = Ccore SC = Straight core - others
Characters 7 and 8 for wind type: T2 = T1000 mode with two layers secondary below the primary - S1 = Standard mode - others to be determined.
Characters 9 to 12 for primary and secondary turns: 1545 would mean 15 turns primary and 45 turns secondary.
So if a builder uses AK30-GL-??-EC-T2-1554 we will know right away your build type and not have to remember this or ask you the same thing. Or maybe just summarize your build in your signature any way you want.
- Coil Winding Methods and Logic -
By the way the Akula circuit is working, seems to me the L2 has to react after the L1 pulse not at the pulse, then the winds must be Standard Mode so L2 can work on its own half of the core out of phase to L1.
In the T2 mode I cannot see why you want to have the L1 pulse hit the L2 two layers under L1 right away as L1 hits its half of the core, you are in essence hitting the L2 at the same time and both fluxs leave their center gap and will cancel out on the ecore ends if you have no coils on the ecore ends to pick it up. Does not make sense. I would want the L2 to be fully reactive to the Ecore movements and not mix it up by also being on the L1 side.
If L2 is over the gap -
If L1 is left of gap and L2 is right of gap, at most, I would say go with a T1 mode, I would not wind more then one layer of L2 under L1 and wound from furthest left of L1 side to furthest right on L2 side then build up the rest of the L2 winds on the right. This way only one layer of L2 is exposed under the L1 pulse making it perfectly directional and not like it is now with two L2 layers pushing in two directions when L1 hits. The only other variable with one layer L2 from left is when you get to the gap, reverse direction and continue the L2 on the right side making it a first layer bucking mode that should keep the same direction even if moved from flux of two gap sides. That adds another variable to the build.
Just my opinion but I will be testing a standard tuning and wind method first.
@verpies
Thanks a heep. Now I know how it works. D6 R2 C4 is a voltage splitter (and virtual ground) sharing the output between L2 and the load. R3 is basically a recharge regulator. But I would still add a diode between R3/C11 and facing C11. I would not want any chance of R3 feeding into L2 in case D6 went into reverse threshold mode.
Also can R3 value be changed to pass more back to C11. Seems to me this should remain a major variable that needs to be tuned specifically to each device. If in Build 1 the L2 produces less recharge then in Build 2, then R3 should be able to let more pass to C11 in Build 1 then in Build 2.
The other variable I am seeing is this. D6 could also work if installed in reverse since with the Q1 short circuit, you can expect high voltages to forward pass D5 and possibly hitting the reverse threshold of D6 to fill L2 and (help) recharge C11 along with L2s regular output from L1 flux transfer.
@ALL
Reality Check:
Also, I took a close up shot of the original Akula30 circuit board to match his components to the diagram since I wanted to see this first before I set-up for my testing. I am very puzzled to see right away there is only one big diode there. How is that possible that on the circuit diagram there are two D5 and D6? I think it may be good to inspect the photo and try to trace the components to the diagram to see where the diagram is adding more components or diverging from the diagram.
Seems to me if Akula used his live board to make the self-runner and if the live board does not concord to the circuit diagram, and if Akula did not afterwards modify his live board to reflect the published diagram and then re-do his tests to make sure it is still self-running and all this done before publishing that diagram, then the diagram becomes an unproven variation of the original live board. That is not good because it means the diagram is not proven except by x_name41 who had to fake his demo when using that diagram. Just trying to cover all the bases. At this point I cannot trust Akula to be so forthcoming as to give exact details of his builds. He is probably getting a self-runner but when diagrams do not follow the actual build, we have to presume the inventor is playing a game of "I did it this way but you should try to do it that way and see if it works". I hate playing games when it comes to such matters, I would prefer to see exactly how Akula physically did it and not how he made his diagram. Diagrams do not produce OU.
You see, unfortunately we are caught in the middle of something that has varying interests. I am sure Akulas' main interest is to make some money with this, otherwise he would be more open and his diagrams would be 100% accurate and we would already have circuits coming out of China in container loads. But by the same token he needs the worldwide exposure to attract those money guys but he surely does not want to spill all the beans to keep control. So he shows an OU effect and puts down a diagram. Follow the diagram and good luck. If you need any direct assistance this increases demand for his direct intervention that could then be negotiated in his favor. OU forces melding with personal economic forces changes the game and puts the onus on us to try and see through the haze of interests. It is not easy when a guy sees this as a form of potential revenue for his family while trying to get exposure and not spill all the beans. The irony of all this is that our effort here is basically based on a non-proven diagram that we know had been faked by a second party. This for Akula is a good situation. He says use 1 ohm for R3 but he probably knows that a better value is required based on the recharge potential. But if builders do not wise up on those notions because we "stick to the diagram and values", we could be here for a long time before a self-runner is on the plate.
@Peterae
Truth is never negative. Better to be aware of the counter forces then to rely on blind faith. Sometimes I spend hours on SM videos close up trying to find a new lead for @GK or my FTPU tests, each lead adds to the others. In this case all we have is one real photo and a host of diagrams that do not reflect the real photo. So somewhere, somehow we have to see through this and realize, this is not a replication but 1/4 replication and 1/4 re-creation and 1/2 pure R&D. That's what Akula expects us to do to make it work. That's the reality.
But I have such a good feeling about this because at least he put down the basic out-of-the-box principle and this can be tried in many ways, not only as per the diagrams. I could even try a more passive variation of this in the FTPU or in any other build. So it's still good but "no free lunch'". Akula is playing a very delicate balancing act and he is doing it very well. He probably learned from TKs demise. Show you're open but don't say it all to give it away. As long as we are wise to the game to realize the diagram is not 100% and changes will be required, we should be OK.
Sorry for long post. I'll put myself on voluntary mute mode so I won't distract you guys any more then I have already until I get my testing up so, back to the bench.
wattsup
Quote from: verpies on 2014.04.20, 17:05:48
Check that R14 and R15 are connected according to the diagram below and have resistances that are written next to them on this diagram.
If still no luck, than make a high resolution photo (or scan) of the components and traces marked in red (put R14, R15 and R6 back into the board for the photo)
Component designations are in accordance with Groundloop's PCB.
Dear Verpies.
Thanks for the info. I removed the TL 494 and expected to see between pins 1 and 16 a variable resistance of 2K up to 6.7 K instead I got a slowly rising resistance, rather like charging a cap, this does not compute !! :) As I should see just the fixed 2K and the variable 4.7 K and nothing else. Am I correct in this assumption ?
Cheers Grum.
Quote from: Grumage on 2014.04.20, 18:37:11
Dear Verpies.
Thanks for the info. I removed the TL 494 and expected to see between pins 1 and 16 a variable resistance of 2K up to 6.7 K instead I got a slowly rising resistance, rather like charging a cap, this does not compute !! :) As I should see just the fixed 2K and the variable 4.7 K and nothing else. Am I correct in this assumption ?
Cheers Grum.
Grum,
Cold solder on socket pin 16?
GL.
Quote from: Grumage on 2014.04.20, 18:37:11
Thanks for the info. I removed the TL 494 and expected to see between pins 1 and 16 a variable resistance of 2K up to 6.7 K instead I got a slowly rising resistance, rather like charging a cap, this does not compute !! :) As I should see just the fixed 2K and the variable 4.7 K and nothing else. Am I correct in this assumption ?
You are correct.
There should not be any cap connected to pin 1 (even inadvertently)
Dear Verpies, Groundloop and all.
I think I have finally "Sussed" it !! O0
It seems there is an error in the overlay drawing with regards to the GL schematic. The resistors are labelled as R14 (Nearest to duty pot) and R 15 above (R14), on overlay drawing. In fact the 12K resistor should be fitted closest to the duty pot (R15) and the 2K above!!
The upshot is that now I can bring up the voltage to 12 V and shut off the pule train with the duty pot !! HOORAY !! :) However I am not seeing a diminishing of pulse width ?? But I had not reconnected R6 !!
It has been a very busy day for me, feeding the 5000 !! I will resume testing tomorrow, once again thanks for all the input. O0
Cheers Grum.
Quote from: wattsup on 2014.04.20, 17:59:01
The Sunday Akula Review - hahaha
@Itsu (but also @Grum - @Happy)
About the spikes, where is your copper strip grounded? On the diagram it shows the copper strip grounded at C3. I don't know if that would make any difference when this strip will receive some static discharges from the the shock treatment of the Ecore. Maybe we need to add something between the copper strip and ground to kill or attenuate the spike before it hits the ground since in essence this is a floating ground system? Or send the copper strip connection to a real ground off the circuit to see if the spike will go away. Don't know.
Also, in most of Akulas' devices he has the transformer off and away from the board. Maybe you need to extend your 5 transformer wires to place it further away from the board since maybe its air static from L1/L2 towards the L1 choke then feeding into the VR1. Or as you are pulsing, move the transformer to see if there are any angles that influence the spike. Or place a metal plate between the transformer and the board to see if more shielding makes a difference just to see if the source is the transformer via air.
Hi wattsup,
i was fighting these spikes this afternoon, but manipulating the copper strip does not seem to influence them.
I can remove the copper strip connection from the board and/or connect it to real ground, but the spikes remain.
Disconnecting 1 side of L1 (choke) clears these spikes.
I used some 0.1uF ceramic caps directly across pins11/12 to 7 of the TL494 and made a solid stranded wire directly from pin 7 to minus
Also a 0.1uF ceramic cap across the 7812 out and ground pins did not clear it up, so i guess i have some violations against the star ground concept.
This means i have to redo the whole ground system of my board, so i will leave it for the time being.
Hopefully the PCB's are better designed so you won't have these spikes.
Regards Itsu
Quote from: verpies on 2014.04.19, 23:25:19
You still have the L1 and L2 currents reaching maxima and zeros simultaneously. That's a characteristic current transfer in a "normal transformer". That transformer mode is wrong in my opinion.
For a "flyback transformer" mode (the correct one IMO), L2 should be reversed and these waveforms should look like this:
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12891)
Ch1 (yellow) : Gate voltage in reference to ground
Ch2 (blue) : Voltage across R1 (0.2 Ohm), with C2 removed (L1 Current)
Ch3 (green) : Voltage across R3 (1 Ohm) (L2 Current)
OK, i swapped L2 connections and redid the current measurements using R1 and R3 as CSR's (C2 disconnected).
Ch1 (yellow) gate signal, groundless as the other to probes grounds are at +12V level, so its somewhat distorted and just for reference.
Ch2 (blue) = voltage across R1, with C2 removed (L1 Current)
Ch4 (green) = voltage across R3 (L2 current)
Regards Itsu
Quote from: Grumage on 2014.04.20, 20:07:15
Dear Verpies, Groundloop and all.
I think I have finally "Sussed" it !! O0
It seems there is an error in the overlay drawing with regards to the GL schematic. The resistors are labelled as R14 (Nearest to duty pot) and R 15 above (R14), on overlay drawing. In fact the 12K resistor should be fitted closest to the duty pot (R15) and the 2K above!!
The upshot is that now I can bring up the voltage to 12 V and shut off the pule train with the duty pot !! HOORAY !! :) However I am not seeing a diminishing of pulse width ?? But I had not reconnected R6 !!
It has been a very busy day for me, feeding the 5000 !! I will resume testing tomorrow, once again thanks for all the input. O0
Cheers Grum.
Grum,
I have checked and there is no error in the component values.
R15 (12K) is closest to variable resistor. R14 (2K) above that.
The main reason that I did not include the component labeling at
the silk screen was to avoid just such problems. Sometimes EAGLE
display the component numbering all over the place and it is a lot
of work "smashing" the components and manually move the numbering
around for a better view. EAGLE CAD is a free downloadable program.
When you solder the board, then you use the program to see what
components to solder where. That was the main idea.
The symptoms you did discovered when using a Ohm meter at pin 1
on the socket to pin 16 was a "floating" value. So when you did re-solder
the resistors to the correct position, then you most probably did get
a good solder connection for the ground of R14.
But I'm glad that you solved your problem.
GL.
Quote from: Groundloop on 2014.04.20, 20:38:18
I have checked and there is no error in the component values. R15 (12K) is closest to variable resistor. R14 (2K) above that.
...but on the
image of the PCB (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12942), the designator R14 is closer to the 12k resistor ...and the designator R15 is closer to the 2k resistor.
Its not surprising that Grum (and one other person) had wrongly followed these designators according to their proximity to the resistor.
If 2 persons make the same mistake, then it is a good indicator that they were confused by the same issue during the assembly.
These designators need to be addressed in v2 of the PCB.
Quote from: Itsu on 2014.04.20, 20:22:59
OK, i swapped L2 connections and redid the current measurements using R1 and R3 as CSR's (C2 disconnected).
Ch1 (yellow) gate signal, groundless as the other to probes grounds are at +12V level, so its somewhat distorted and just for reference.
Ch2 (blue) = voltage across R1, with C2 removed (L1 Current)
Ch4 (green) = voltage across R3 (L2 current)
Now the waveforms look well. Note the flat tops and
rising ramps in L2 current waveform at the ends of the OFF period. They are significant.
The magnitude of L2 current and the length of an interval during which this current flows, is proportional to the energy that is being returned to C11 - treat it as your payback.
A video of these waveforms at different supply voltages and DTY pot settings with and
without the LED load would be nice at this point.
If you want to generate more raw experimental data as food for thought, you can try temporarily substituting an automotive bulb in lieu of the LED bank.
I know this circuit is not designed to operate with resistive loads, but it will not break.
In theory, you should see some narrow strong negative "magic" pulses in L2 current (during the same scoping arrangement as
here (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12940)) when the circuit is correctly tweaked with its potentiometers and its transformer is constructed correctly (the most difficult part IMO).
Quote from: Grumage on 2014.04.20, 20:07:15
I think I have finally "Sussed" it !!
Good, now you can have the satisfaction of solving the problem for yourself and for others. O0
Quote from: Grumage on 2014.04.20, 20:07:15
The upshot is that now I can bring up the voltage to 12 V and shut off the pule train with the duty pot !! HOORAY !! :) However I am not seeing a diminishing of pulse width ?? But I had not reconnected R6 !!
You can reconnect R6 and the MOSFET's gate resistor now, but it will not make the DTY pot control the duty cycle like you expect.
Before you proceed further make sure that your L2 winding polarity is according to
the dot convention in Fig.2 (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12954) or such that you get L1 & L2 current waveforms like Itsu
here (http://www.overunityresearch.com/index.php?topic=2358.msg37902#msg37902).
The DTY pot controls the sensitivity of MOSFET's
duty cycle reduction in response to the following events:
1) The C6 is overcharged by pulses coming from L2 via D6. Not when C6 is overcharged by your external power supply!!!, or...
2) The L2 generates a negative pulse with a very steep falling edge ("negative" refers to the "down direction" on
this (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12940) scope shot of Ch4 ), or...
3) The negative pulse from L2 is so strong that it completely discharges C10 to zero ...or charges it negatively.
(
Events described in pt.2 and pt.3 are brought into the TL494 by R6 ).
P.S.
On
Akula's diagrams (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12849) C6=C11 and C10=C3 and R6=R7.
The strong negative pulse generated by L2 is that "magic pulse" that is not predicted by conventional electronic engineering rules and transformer theory.
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12954)
Quote from: verpies on 2014.04.20, 22:57:33
...but on the image of the PCB (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12942), the designator R14 is closer to the 12k resistor ...and the designator R15 is closer to the 2k resistor.
Its not surprising that Grum (and one other person) had wrongly followed these designators according to their proximity to the resistor.
If 2 persons make the same mistake, then it is a good indicator that they were confused by the same issue during the assembly.
These designators need to be addressed in v2 of the PCB.
These designators are now addressed in v2 of the PCB.
Quote from: Itsu on 2014.04.19, 20:47:04
I did just that, but besides a cleaner drain voltage signal (nice square wave, so no more jump half way) no noticeable other effects are seen.
Regards Itsu
Good day Itsu,
Funny you should mention the change in the Drain waveform..... So, changing position of C11 to the *opposite* side of L3 (Choke number according to Verpies *new* schematic) diminished or eliminated the aberration that appeared about half-way thru the OFF pulse that was discussed by Verpies and I earlier?
Unfortunately, regarding resonant systems it is usually impossible to just throw one together with random values,(Inductive & Capacitive reactance) and expect it to be in resonance. So it is not surprising that you noticed nothing out of the ordinary......
It really makes me wonder why the photo on the 30 watt Akula schematic does NOT depict ALL the parts that are listed on the same schematic.......
take care, peace
lost_bro
Quote from: lost_bro on 2014.04.21, 00:32:41
It really makes me wonder why the photo on the 30 watt Akula schematic does NOT depict ALL the parts that are listed on the same schematic.......
take care, peace
lost_bro
Maybe to deepen the confusion. :-\
Quote from: Hoppy on 2014.04.21, 09:41:33
Maybe to deepen the confusion. :-\
Akula's last video with the separate power supply is the same way. When he takes the device outside and places it on the tree stump, he puts down a schematic first that clearly has three integrated circuits, but the board he is showing has only two. I went through that video frame by frame and found a hex inverter chip, a PLL and a half bridge driver with a center-tapped transformer. Maybe we will see a running demo of that device in the future.
Quote from: verpies on 2014.04.20, 23:25:15
Now the waveforms look well. Note the flat tops and rising ramps in L2 current waveform at the ends of the OFF period. They are significant.
The magnitude of L2 current and the length of an interval during which this current flows, is proportional to the energy that is being returned to C11 - treat it as your payback.
A video of these waveforms at different supply voltages and DTY pot settings with and without the LED load would be nice at this point.
If you want to generate more raw experimental data as food for thought, you can try temporarily substituting an automotive bulb in lieu of the LED bank.
I know this circuit is not designed to operate with resistive loads, but it will not break.
In theory, you should see some narrow strong negative "magic" pulses in L2 current (during the same scoping arrangement as here (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12940)) when the circuit is correctly tweaked with its potentiometers and its transformer is constructed correctly (the most difficult part IMO).
Ok, here some specifics on how i have the transformer constructed and connected, see picture and video:
https://www.youtube.com/watch?v=lJTxEPuNIlk&feature=youtu.be
Fine tuning still will be a problem as the high current prevents me from doing long ontime measurements.
yesterday i had it on to long while tuning the dty pot R11 (no change in the currents during that tuning), and thick smoke came up from C6 / VR1 area :D
Anyway, thanks for the guidance up till now, its much appreciated.
Regards Itsu
Quote from: lost_bro on 2014.04.21, 00:32:41
Good day Itsu,
Funny you should mention the change in the Drain waveform..... So, changing position of C11 to the *opposite* side of L3 (Choke number according to Verpies *new* schematic) diminished or eliminated the aberration that appeared about half-way thru the OFF pulse that was discussed by Verpies and I earlier?
Unfortunately, regarding resonant systems it is usually impossible to just throw one together with random values,(Inductive & Capacitive reactance) and expect it to be in resonance. So it is not surprising that you noticed nothing out of the ordinary......
It really makes me wonder why the photo on the 30 watt Akula schematic does NOT depict ALL the parts that are listed on the same schematic.......
take care, peace
lost_bro
Correct, the drain voltage now was as clean as the gate signal, nice square signal.
I did not take a screenshot, and i had to revert back to the normal C11 position to make these last current measurements, as with C11 at the other side of the choke,
the current measurements where skewed, probably because the ground leads also had to be moved to c11 (before the choke).
Regards itsu
I cannot access the latest Akula video's as they now seem to be "private". Or is it something wrong on my end?
Hope someone saved them.
Quote from: ION on 2014.04.21, 12:05:52
I cannot access the latest Akula video's as they now seem to be "private". Or is it something wrong on my end?
Hope someone saved them.
Here's a clone, err, remix of No 3.
http://www.youtube.com/watch?v=nb2sN2o4grc
QuoteThe strong negative pulse generated by L2 is that "magic pulse" that is not predicted by conventional electronic engineering rules and transformer theory
O0
But it is predicted using electron current flow :D
Hi Itsu,
Has you have 2 coils with one with a metallic foil, may I suggest you to try this little kind of experiment
with you own 'gears and stuffs' ? Perhaps something new?
(http://freenrg.info/4OUR_F/My-exp2.jpg)
Some results:
Input: square wave. About 15 volts PK-PK. 50% DTC. 3W/220Volts Leds Bulb.
Bulb not connected (no load): 90 volts PK-PK.
Across the 1 Ohm resistor : kinda sine wave. 240 mVolt PK-PK.
Bulb connected (with load): 80 volts PK-PK.
Across the 1 Ohm resistor : kinda similar sine wave. 250 mVolt PK-PK.
Needs a very precise tuning. Freq from about 125 KHZ without additional cap to about 30KHZ.
Does not work with cap > 20 nanoF.
Needs at least a double trace oscillo to make more precise measurements.
In this configuration, the circuit must be grounded (via the Grid).
I have also tested some other configurations with success.
Success = to get the Leds lighted (but not fully, indeed).
This CCT is the same as the one pictured in "principle_of_work_of_Kapanadze.pdf"
See the attached file.
One picture from this document:
(http://freenrg.info/4OUR_F/Kapan-Pic.jpg)
Cheers,
Jean
I dont have the electronic skills you guys have and it amazes me to watch you work.
A visual aid for those visually impaired,,,,,,,like me :D
http://makeagif.com/i/ySzksZ
@Verpies and GL Thanks you for reporting this situation as resistors reverse position ...
I had welded the R14 close to the trimmer now I will weld with the right location R15 close to the trimmer...
@Itsu, @Grumage, @Happy
This is a loop system where there cannot be any room for snags, delays, flux bounce, What ever L1 produces has to impart and return right away to C11 or its dead. Any delay will kill the effect. There is no other way because this is supposed to work in a loop meaning there is no time to play fancy flux games in the core. The circuitry just won't wait around for any delays.
First thing is the primary you guys are using of 15 turns. Way to many turns. Most flybacks I have used I would not put more then 5 turns primary to keep the full pulse as close to the core as possible instead of diluting the pulse through three layers of primary that also are on two layers of L2. You don't need any more then 5 turns plus if you need a little more inductance, 10 turns would be the maximum you would ever want on the L1 like Akulas' core photo that I mapped previously.
The other variable is the core type. There should be no center gap in this at all. If there is a gap you will transfer via a flux bounce but that will take too long for looping. L2 cannot sit around waiting for that bounce to occur because L1 is already getting its next pulse by then so there is waste and cancellation. There cannot be any more activity then pulse/receive in this system and everything is designed exactly for that.
The problem with gaped cores is you loose a good 80% of the Al value. The Al value is the only thing required to just keep looping and looping inside the core for this to work. Remove 80% of that and you just have a tight trampoline that likes bouncing people around but without any time to do that its just wasting energy.
Then, if we look at this scientifically, the T1000 mode has no basis in anything logical. No one can explain why such a wind is required or even explain what the T1000 wind is creating inside the cores besides what I tried to explain in my previous post. Everything right now in the Ecore is wrong. So why are we all working this thing with a T1000 wind. Of course, I got it, this wind is so exotic, so out-of-the-box, hmmmmmm, more like out of this universe in the "Wishful Thinking" galaxy.
The T1000 wind method is seen in air coils like the TK coil types, but not when you are using the added influence of an e core.
To prove this, just run the circuit, scope the L2 then stop the circuit, remove one turn of L1, then try it again to see if L2 is reduced, if not, remove another turn of L1, etc, so you need to work backward and remove as much of the L1 as possible. This will slowly increase the pulse applied to the first 5 turns of L1 that is nearest the core or in your case over two layers of L2 (which I think is premeditated stagnation). If you see no change in L2 by the time you are at the last 5 turns, then remove the 5th turn and so on. You will quickly realize that the problem all started when L2 went under L1. What a waste.
If I were you guys, I would strip the damn things and rewind them properly or find a new ecore to work with.
Start 5 turns for L1 starting top left going down and stop at the gap, leave enough length of L1 wire to make 5 more turns plus a length more to connect to the circuit. Wind 2 layers of L2 starting at bottom right going down and leave enough wire to make two more layers. The best will be if the L1 wire is thicker then the L2 wire (just like Akulas' coil). You basically want to be near the same weight of copper on each side. Once you have 5 turns and 10 turns, connect the core to the circuit even with the longer wires, or, if you want, make a slit in the non turned length to connect it closer to where the turns stop. Run the circuit, check L2 output.
Now add a few more turns to L2 and check it again, if it increases output, add 2-3 more turns to the L1 now on the second layer. Check L2 output. If it does not increase, remove the 2-3 turns. Add more turns to L2 and check again and as long as L2 increases with more turns, keep adding more turns to L2. You can always try 2-3 more turns on L1 and check again to see if L2 increases but I doubt it because after the first 5 turns L1, as soon as you start the second layer of L1 you start diluting the first layer of L1 which is not wanted at all. So you keep adding turns to L2 and check if it increases until you get to the point where adding any more makes no difference or little difference. Remove those turns and stop there. That is now your Ecore with winds. That's what this device wants as a transformer. Nothing exotic, just fast and straight transfer. If you only have a gaped core, use it but think about getting a non-gaped core as well because you will have to do this on that non-gapped core as well and interchange them both just to cover all the bases. It is possible that the gaped and non-gaped cores will require a different number of L2 turns.
I am sure once this is done, the system will start purring along near loop mode. The only other factor will be to find the right value for R3 to match the real output conditions of L2 so that C11 gets ONLY just the right return to keep the loop going.
For now, forget about spikes in the circuit. Once the coils are well balanced and running, that problem should wind up being eliminated with a smoother running ecore.
Just ask your self, with such a furty little core and very good strong driving pulse, why do I need so many primary turns? hehehe
I figured out the finer criteria of the circuit. The game is simple but the rules are unforgiving. Via L1 push the output to L2 high enough so D6 cannot add to L2. This way more of D5 goes to load and if L2 can go even higher on its own, create a reverse threshold effect to D6 so L2 also adds a very strong ripple to the D5 output to the load. There is no other way I can see this work. It is a great system to test with.
If L1 is right on the first layer of the former, when Q1 shorts it out and you scope L1, you should be seeing high spikes and not the wave forms seen so far. I think the stable waveform you are seeing is because L1 is just being (*&&&?%*(*&) mufflered or dampened by the distance is has to impart through two layers of L2 then to the core. Imagine when L2 is energized, now L1 has same two layers distance to pass plus the fact that the two layers are energized, this isolates L1 from the core even more so you are working the core in a very suffocated way.
I promised to stay on mute mode but I needed to convey this now cause I am ordering my e-cores ETD-29-16-10 N87 gaped and non-gaped plus a few others (bigger ones) to have on hand but while they arrive, I did not want you guys to waste any more time with the T1000 method because I know for sure it is a waste of time and it is creating bad effects on the circuit operation.
All the best.
wattsup
@wattsup
I agree L2 reacts very fast, you have to add turns to L2 or decrease L1
@ Wottsup,
I got straight onto your suggested coil wind but have not noticed any real change from my previous L1 - 15 turn primary and L2 - 28 turn secondary. It would be wise for you to build the circuit using the PCB (if anyone has any leftover spare), so we can compare results using the same coil config and circuit build - apples to apples.
Happy
Dear all.
Take a look a this before it disappears !!
https://www.youtube.com/watch?v=CKmW7HxXvas&feature=youtu.be
If anyone can save it, I suggest you do !!
Cheers Grum.
Good day Grum
OK, I am downloading the video to my computer as I write this message.
Looks like the same circuit board used with the SMPS in the last video, just looped it back directly w/o the SMPS......
Very interesting.
take care, peace.
lost_bro
EDIT so now we have to find different transformers (ie: flyback)......
Dear Wattsup.
I have two transformers wound, see attached.
The reasoning behind the T-1000 wind is as follows. If we take the original picture presented at the start of this thread and the turns ratio stated on the schematic. Then we would clearly see that one end would be a lot smaller than the other !! However the original picture clearly shows an almost even wind on both sides, suggesting that at least two layers of wire (first 30 turns) were laid end to end !!
Secondly the GeoFusion ring core also had it's secondary wound across the gap in the Ferrite. I am afraid non of us came close to a successful replication of that device either !!
This short You Tube of mine shows an interesting effect though !! https://www.youtube.com/watch?v=SPV89JWVdtY
With that said I am still looking forward with great interest to your results as I know that your famous "Diode Carousel" will play a big part in your investigations !! O0
Cheers Grum.
Quote from: lost_bro on 2014.04.21, 18:33:50
Good day Grum
OK, I am downloading the video to my computer as I write this message.
Looks like the same circuit board used with the SMPS in the last video, just looped it back directly w/o the SMPS......
Very interesting.
take care, peace.
lost_bro
EDIT so now we have to find different transformers (ie: flyback)......
Dear lost_bro.
Thanks !! O0
Cheers Grum.
Quote from: Itsu on 2014.04.21, 11:00:33
Ok, here some specifics on how i have the transformer constructed and connected, see picture and video:
https://www.youtube.com/watch?v=lJTxEPuNIlk&feature=youtu.be
Fine tuning still will be a problem as the high current prevents me from doing long ontime measurements.
I don't pretend to know how that transformer is supposed to be constructed, but all of the Akula's diagrams state that the copper shield should
not form a closed loop, and your shield appears to be shorted. Shorted shield will lead to high current when windings are pulsed.
Also
the video (http://youtu.be/bYEuV-ji4tk?t=0m45s) that shows the transformer construction of a
related device, shows two shields and neither one is shorted.
@ Grum thanks just in time has already locked but I managed to download it!
Quote from: TutorialFE on 2014.04.21, 19:39:58
@ Grum thanks just in time has already locked but I managed to download it!
I missed it, but fortunately T-1000 snagged a copy and left it on his TS3 server. (Thank you T !)
I think with this one he leaves little doubt as to his device being genuine. The flyback core is certainly very dark black--that is probably going to be the hardest part to replicate. Everything else looks relatively straightforward.
In this thread, a device claimed to be OU also uses two open foil shields for a high voltage bias.
http://www.overunityresearch.com/index.php?topic=2261.0
(he claims far more power than Akula)
Quote from: Matt Watts on 2014.04.21, 19:58:58
I missed it, but fortunately T-1000 snagged a copy and left it on his TS3 server. (Thank you T !)
I think with this one he leaves little doubt as to his device being genuine. The flyback core is certainly very dark black--that is probably going to be the hardest part to replicate. Everything else looks relatively straightforward.
What do you think of this? likely to be the same that uses Akula ....?
That ferrite in his video is too grey to be MnZn it's hard ferrite and looks like a line transformer from a TV as Tutorial FE posted to me.
Dear All.
A little bit of extra knowledge came my way today. It would appear that the successful Russian devices are using a Ferrite material that holds some magnetic Remanence !!
I have been looking through my stash of Ferrite rods that I stripped out of old radios when I was a young man. :) I have found one that after applying a magnet to it can influence a Compass needle for a number of minutes later. I tried the same experiment with the cores I am using. No effect was seen. This situation obviously shows that our modern materials are of a better make up than that of the past !!
So does anyone have an idea how this rod that I have could be used ??
Cheers Grum.
Quote from: verpies on 2014.04.21, 19:36:28
I don't pretend to know how that transformer is supposed to be constructed, but all of the Akula's diagrams state that the copper shield should not form a closed loop, and your shield appears to be shorted. Shorted shield will lead to high current when windings are pulsed.
Also the video (http://[b=http://youtu.be/bYEuV-ji4tk?t=0m45s) that shows the transformer construction of a related device, shows two shields and neither one is shorted.
Hmmm i thought it was clear in the pictures, but my copper shield is not shorted, see picture below.
Your video link is faulty, it should be: https://www.youtube.com/watch?v=bYEuV-ji4tk&feature=youtu.be&t=0m45s
Looking at it it seems that the first copper shield was in series with the first coil, where have i seen that before ???
Regards Itsu
Quote from: Grumage on 2014.04.21, 20:25:16
It would appear that the successful Russian devices are using a Ferrite material that holds some magnetic Remanence !!
http://magician.ucsd.edu/essentials/WebBookse41.html
QuoteWe can coax the magnetization of otherwise firmly entrenched particles to follow an applied field, if that field is larger than the coercivity. Exposing a particle to a large magnetic field, will allow magnetic particles whose coercivity is below that field to flip their magnetic moments to a direction at a more favorable angle to the applied field, resulting in a gain in magnetic remanence in that direction. This type of magnetic remanence is called an isothermal remanent magnetization or IRM
Sounds an awful lot like what T-1000 has been explaining to me.
Quote from: Itsu on 2014.04.21, 20:30:23
Hmmm i thought it was clear in the pictures, but my copper shield is not shorted, see picture below.
Your video link is faulty, it should be: https://www.youtube.com/watch?v=bYEuV-ji4tk&feature=youtu.be&t=0m45s
Sorry for the malformed link. Do you have a link to that
new Akula video (http://www.overunityresearch.com/index.php?topic=2358.msg37925#msg37925) that was suddenly made private?
The white wires under your copper shield looked liked they were soldered to the shield and were bridging the gap.
Quote from: verpies on 2014.04.21, 22:30:18
Do you have a link to that new Akula video (http://www.overunityresearch.com/index.php?topic=2358.msg37925#msg37925) that was suddenly made private?
Here it is, thanks to freeenergyinfo:
https://cloud.mail.ru/public/1404c889d895/%D0%B8%D1%82%D0%BE%D0%B3%20%D1%84%D0%BE%D0%BD%D0%B0%D1%80%D1%8C%20%E2%84%963.mp4
Quote from: verpies on 2014.04.21, 22:30:18
Sorry for the malformed link. Do you have a link to that that was suddenly made private?
I managed to download it and upload it here ....
https://www.youtube.com/watch?v=HMNHiCn46dc
https://www.youtube.com/watch?v=3IjQ37TopWE
I've devided in two parts and I hope It can be help to everyone ... O0
Thanks.
Gutted transformer & gutted capacitors - that's new.
Maybe there are not any button batteries embedded in these caps after all.
P.S.
I just realized that I don't know how supercaps appear inside,... practically.
Anyone?
When starting the circuit up with the 9 Volt Battery
the LEDs seem quite bright. As soon as the 9 Volt
Battery is disconnected it seems that the LEDs drop
somewhat in their intensity.
That could be a symptom of capacitor charging and
discharge. Is the 9 Volt Battery used to charge
capacitors and also provide Gate Charge to a switching
MOSFET which controls discharge of the capacitors to
the LEDs?
Once the circuit is activated how long will it continue to
run before it eventually dies out?
After close examination of the video at very slow speed
it appears that the LEDs are DC operated.
Supercapacitors could be the secret.
Can anyone work out the value of this cap which appears to marked 5V and is close to where the 9V battery is connected to the circuit. 1000 can be seen, so is this a 25V cap or a 2.5V / 5V cap at a higher capacitance?
The LED light level is very low and illuminated for very short periods of time, so I plump for a capacitor fed supply rail feeding a carefully tuned fly-back oscillator with series connected LED load. Note that a third 2,200uF capacitor and an additional TO220 device appears on the circuit board in part 3 when the device is demonstrated working disconnected from the SMPS. The switch has also been transferred to the board. Notice the difficulty he has latching / holding the LED's illuminated, suggesting to me current starvation to the oscillator.
Hoppy
Looks like a 100uF 25V cap to me, it's not unusual for part of the voltage to be missing as the sleeving is place on the cap during manufacture from a long printed tube and is spliced and heat shrunk onto the aluminium body of the cap.
Thanks for saving the videos Tutorial_FE O0
Dumped those led's are low Hz pumped if you look at Tutorial's first video above at 4:51 you can see the iris of the camera trying to react to the pulsed light, to see that look at the white background.
It is not good he keeps playing with the pot and making it stop, if i were showing OU i would set it up and let it run, OK you can show the pot adjustment is critical but why again and again and again, get's a bit annoying, watching the video I'm thinking no dont adjust it again LOL
Quote from: Peterae on 2014.04.22, 08:03:12
Looks like a 1000uF 25V cap to me, it's not unusual for part of the voltage to be missing as the sleeving is place on the cap during manufacture from a long printed tube and is spliced and heat shrunk onto the aluminium body of the cap.
Unfortunately, I cannot get a clear clip on the top fold-over of the cap label to confirm the voltage rating.
Quote from: Peterae on 2014.04.22, 08:03:12
Dumped those led's are low Hz pumped if you look at Tutorial's first video above at 4:51 you can see the iris of the camera trying to react to the pulsed light, to see that look at the white background.
It is not good he keeps playing with the pot and making it stop, if i were showing OU i would set it up and let it run, OK you can show the pot adjustment is critical but why again and again and again, get's a bit annoying, watching the video I'm thinking no dont adjust it again LOL
Ultra suspicious! Anyone care to design an ultra low current / low frequency LED driver working from a cap supply rail and capable of a few seconds operation after each application of a 9V battery.
Don't get me wrong, i still think it's impressive, but please show it running for at least 1 minuet :)
Maybe he did in another video, i have been in and out the last couple of days, and the video vanish faster than they appear :-\
Bearing in mind i can light 13 LED's using no battery's at all and for ever :-*
http://www.overunityresearch.com/index.php?topic=202.msg2130#msg2130
(http://www.overunityresearch.com/index.php?action=dlattach;topic=202.0;attach=1110)
While Akula No3 seems to be the best of the videos, superbright LEDs are not a good proof of anything unless as others have said it is allowed to run for a very long time.
Notice (Akula No3) the core gapping circular pieces of something...looks like copper to me.
I believe in the video he uses stock untampered electrolytic capacitors.
I cannot stress enough the need for precise gapping of the core for efficient operation of a flyback converter. The gap allows quick dissipation of the remnant stored magnetism, allowing the core to "reset" for the next pulse of energy. Without a gap, the core will quickly go into saturation and your FET will draw excessive current unless you employ cycle by cycle real time current limiting.
If you can measure the FET current you will see a sudden spike or rise in current at the end of the current ramp.
Other factors such as frequency too low for the given inductance can cause the same spike.
Properly designed, you should not be drawing high currents, the Watts in and Watts out should be very close.
If you are drawing 10 amps from your battery at 12 volts, you are supplying 120 Watts to a 30 watt circuit. This is indicative of the above mentioned problem.
There are other factors that can cause excess dissipation e.g. slew rate too low in the gate drive to the FET or operating frequency too high or both.
You want to strike a happy medium with all these factors for most efficient operation.
Where extremely compact lightweight design is not needed my preference is low frequency operation , high inductance in the core and of course for a flyback converter, precise gapping.
Quote from: Hoppy on 2014.04.21, 16:20:06
@ Wottsup, I got straight onto your suggested coil wind but have not noticed any real change from my previous L1 - 15 turn primary and L2 - 28 turn secondary. It would be wise for you to build the circuit using the PCB (if anyone has any leftover spare), so we can compare results using the same coil config and circuit build - apples to apples. Happy
@Happy
Sorry for delay in responding. Had some family stuff all Monday but managed to order my cores.
That is good since now you know you can get the same result with fewer turns. That is because when you reduce the number of primary you increase the pri/sec ratio at the same time. But what wind method are you using again?
What type of diodes are you using as D5 and D6?
@all
I am attaching a pdf on the D5 and D6 diode model 15SQ045 that was in the parts list. Is this the most appropriate diode you guys can find for this app? The question is, is it close to the specs of Akulas diode model. The 045 designation putts this in the VRRM 45 volts range which could be either to high or to low depending on when you would expect L2 to bleed into the load. Don't know enough about specs. Above my pay grade. hahahahahahahaha But this is another variable point.
The thing is I cannot see a second big diode on Akulas photo so is it possible that the D6 diode is a smaller type since D6 should not require the same "through-put" as D5 but may require a higher (or lower) VRRM? I know Akula put down the same model on his diagram but the photo says otherwise. The D5 and D6 diodes in the original diagram are marked as being MBR3545. Is this a Russian model number and who then indicated that the 15SQ045 should be used?
@Grumage
Man oh man, everything is coming together slowly but surely.
Remember when I had posted a mapping of the Akula 30 Ecore showing how it fits inside the former and explained how the former diameter that Akula must have had makes it impossible for the L1 side to have three layers of L1 (seen as thicker wire) plus two layers of L2.
Look at Akulas last Cap Gutting Video, the core has gaps but both coils run across so there is a balance and he only uses one layer per side across the gap. Do you see his primary has a second layer of a few more turns. Akula probably does not know it but that is because all pulsed coils suffer from what I call Half Coil Syndrome where you have 75% of the energy dissipation occurs in the first half of the coil, he puts extra turns on a second layer to push that 75% across more of the 1st layer wind length. He must have used his scope to wind and tune at the same time to then stop at that specific second layer turn count. hehehe
That is a big difference then the T1000. That is such great news as a wind confirmation of basic rule, if you pulse one side of gap, you pick up on the other side. Otherwise, if you pulse across the gap, you pick up across another gap, otherwise there are timing issues that the circuit will not wait around and needs immediate reaction, like Akula just showed.
Otherwise with the T1000 and one center gap, you may as well look at a tick-tack-toe game to figure out the flux paths. I am sure your L2 does not like it and does not know how to output it. You were getting a deep negative if I remember @verpies mentioning it. Or was it @Itsu.
The second best news about the last Akula video is exactly what I was talking about. He is using 4 pots, and we are using two pots, the two extra pots that would be good to use are at either R3 or R2 or R5. R3 (vary the return), R2 (vary the L2 bleed) and R5 (if varying R5 can vary the total load consumption), this will provide the level of flexibility required to tune the circuit to many coil types and available energy levels. That is a big big clue to tuning the circuit with pots or by replacing some resistors.
On a side issue, I am curious. If you scope across L1 with probe and ground, one way, then the other way, then only with the probe (no probe ground) one end then the other end, are the waveforms different? Is it the same on the L2 side? If the single probe waveforms are different, those eight waveforms (4xL1 and 4xL2) are the ones I would want to analyze and compare when knowing exactly where the probe and ground were placed. Like P-R1-L1 means the probe is between R1 and L1, G-L1-D5 means the ground is between L1 and D5. Hehehe
@verpies
I need to ask you a simple question about scopes, not because I don't know but because I need someone with some authority in the matter to explain how a scope probe and ground signal is calculating to map the waveform we see on our screen. If this is supposed to be a learning experience then this should be part of it otherwise I fear it will make my job explaining things a lot harder once I get my E-cores.
So here is a simple scenario. You scope across a coil where you already know the probe end is at +80 and the ground clip end it at -40. So how does the scope treat this information and where will the waveform be at that one specific instance in time?
wattsup
I would like to make some suggestions for future builds.
1) breadboard stage: this can be a PCB or vector board design, but parts should be spread out on the board physically as they are seen in the schematic as nearly as possible. This makes for very easy location of parts when you are probing, and if everyone can conform to this, youtube video's will be easier to interpret, as all breadboards will look more or less the same. I have often used this technique and it makes for fast troubleshooting against the schematic. I use pin sockets or solder posts for components that might be changed often. I raise the parts off the board a bit to make it easier to hook a probe to a leg. Also, I can snip out a part and solder a new one to the old leads or easily shunt or series a value. Add test points throughout the board at key points.
2) prototype PCB: this is post breadboard and is used to make sure the design has no issues with trace coupling, interlayer capacitance, parts fit etc. It is made compact for efficiency of material use.
3) production PCB: after all bugs are worked out of the prototype PCB the V1.0 production board is released.
I any any case, parts designators should agree with the schematic across all boards, 1,2,3.
Akula lantern 3 translated by Stivep1 - Wesley
итог фонарь №3
https://www.youtube.com/watch?v=1BNgo09N0yM
Quote from: wattsup on 2014.04.22, 13:12:50
@verpies
I need to ask you a simple question about scopes, not because I don't know but because I need someone with some authority in the matter to explain how a scope probe and ground signal is calculating to map the waveform we see on our screen.
Ideally like this:
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12983)
Non-ideally there are stray parallel capacitances, series cable capacitances, inductances, propagation delays, etc...
In some cheap scopes the inverting input of the front end differential amplifier is connected to the chassey of the scope which makes up a huge plate of a stray capacitor.
Quote from: wattsup on 2014.04.22, 13:12:50
So here is a simple scenario. You scope across a coil where you already know the probe end is at +80 and the ground clip end it at -40. So how does the scope treat this information
It takes the difference between the inverting and non-inverting input of the front end amplifier/attenuator and applies it to S/H + ADC.
Quote from: ION on 2014.04.22, 14:03:28
I would like to make some suggestions for future builds.
1) breadboard stage: this can be a PCB or vector board design, but parts should be spread out on the board physically as they are seen in the schematic as nearly as possible. This makes for very easy location of parts when you are probing, and if everyone can conform to this, youtube video's will be easier to interpret, as all breadboards will look more or less the same. I have often used this technique and it makes for fast troubleshooting against the schematic. I use pin sockets or solder posts for components that might be changed often. I raise the parts off the board a bit to make it easier to hook a probe to a leg. Also, I can snip out a part and solder a new one to the old leads or easily shunt or series a value. Add test points throughout the board at key points.
2) prototype PCB: this is post breadboard and is used to make sure the design has no issues with trace coupling, interlayer capacitance, parts fit etc. It is made compact for efficiency of material use.
3) production PCB: after all bugs are worked out of the prototype PCB the V1.0 production board is released.
I any any case, parts designators should agree with the schematic across all boards, 1,2,3.
Yes, this is good advice IMO. I'm happy with strip board but less experienced builders would benefit by using breadboard.
Hoppy
Quote from: wattsup on 2014.04.22, 13:12:50
@Happy
Sorry for delay in responding. Had some family stuff all Monday but managed to order my cores.
That is good since now you know you can get the same result with fewer turns. That is because when you reduce the number of primary you increase the pri/sec ratio at the same time. But what wind method are you using again?
What type of diodes are you using as D5 and D6?
wattsup
@ Wottsup,
I'm using the wind method you suggested in your previous post.
Diodes are exactly as specified by Akula (MBR3545), as this is exercise is to replicate his device as accurately as possible. I do not wish to alter the diodes, just try different coil winding configs as this was the only detail not given to us. The mosfet driver chip mod is IMO valid as there is a push-pull version of his circuit published, so the driver chip is not a significant deviation from this. I have built all three versions.
Happy :)
Dear All.
Well, another fruitless day !! I have rewound a transformer with 6 turns primary and 18 turns secondary. I have physically gapped the two halves with relation to Dr Andrey Melnichenko's suggestions. Creating two separate halves. Primary on one "triblet" secondary on the other.
At the moment we have a PWM with a fixed duty cycle. As far as I can see the "Duty" pot fitted to the boards we have is primarily a "high voltage cut off" ?? Correct?? So a simple question. can we reduce the pulse width ?? ATM I am pushing over 2 A @ 12V this can't be right ?? Ok, yes, that is only 24 Watts. But I think the LED's could be lit by a fraction of that by having a narrower pulse width !!
I am getting to the point of finding a suitable wall to bang my head against !! :D
Cheers Grum.
Evening Grum,
Before any attempt is made to change the design in order to reduce the duty cycle, I feel that we must decide on why the 'STOP' switch is fitted. My experiments do not show any real purpose for this switch. Yes, it does turn off the LED's (when closed) at lower supply voltage settings but always at the expense of higher current draw. The duty pot certainly sets the high voltage cut-off point which effectively cuts the PWM output. My hunch is that this circuit was an unsuccessful attempt by Akula to achieve desired self-running operation. The latest videos from him, suggest a more successful attempt, assuming of course they are not faked. A few more replications from others may shed some light on these issues.
Hoppy
From Grumage:
QuoteATM I am pushing over 2 A @ 12V this can't be right ?? Ok, yes, that is only 24 Watts. But I think the LED's could be lit by a fraction of that by having a narrower pulse width !!
You are right, however this is the problem with using high brightness LED strings....it is very difficult to know how much power is being used unless you compare the brightness using a light meter that is accurate at the frequency of the light output and comparing against a calibrated DC drive from your power supply and generating a multipoint look up table.
I would prefer a small incandescent lamps, as they don't play as many tricks on your eyes as pulsed LED's.
Save your head for now, there will be plenty of time for that later
Quote from: Grumage on 2014.04.22, 18:28:45
At the moment we have a PWM with a fixed duty cycle. As far as I can see the "Duty" pot fitted to the boards we have is primarily a "high voltage cut off" ?? Correct??
Only 17% correct.
The "high voltage cutoff' is just 1/3 of the feedback loop's function. The others are described in
this message (http://www.overunityresearch.com/index.php?topic=2358.msg37907#msg37907). The DTY pot controls the sensitivity of the feedback loop.
Quote from: Grumage on 2014.04.22, 18:28:45
So a simple question. can we reduce the pulse width ??
Yes, but you have to realize that a reaction to the event described in
pt.1 (http://www.overunityresearch.com/index.php?topic=2358.msg37907#msg37907) is very quick. Scope how the voltage across C6 varies. The feedback loop has to react to these small changes in voltage on pulse-by-pulse basis, but it can't because you are overriding these changes by applying a steady external voltage source to C6 and destroying this feedback.
Because your circuit is under unity at this time, you must help it with the external power supply. Let's say that the circuit is only 70% efficient. This means that you have to supply 30% of the power from an external power supply...but you are supplying 100% *.
Thus the feedback loop has nothing to do. If you delivered only 31% of the power then the feedback loop would start to react to variations across C6.
So put a high power ~100Ω rheostat/variable resistor in series with the supply line between your PS and the whole contraption, so the PS is not so stiff and acts more as a current source than a voltage source - only then the feedback loop will get a chance to regulate the pulse width.
Set the PS to 15V and tweak the DTY pot together with the rheostat. If you don't have a rheostat then stretch out a nichrome wire from some household heating element and clamp a crocodile clip on it in various places or try out a lot of different incandescent light bulbs in series (12V and/or 230V ones).
* When two voltage sources (PS and L2 in this circuit) are in parallel, then the source that provides more voltage supplies
all of the current.
The lower voltage source normally sinks all of the current if there is no diode in series between these sources to stop it or current sharing resistors to distribute it evenly.
Quote from: verpies on 2014.04.22, 20:09:49
Because your circuit is under unity at this time, you must help it with the external power supply. Let's say that the circuit is only 70% efficient. This means that you have to supply 30% of the power from an external power supply...but you are supplying 100% *.
Thus the feedback loop has nothing to do. If you delivered only 31% of the power then the feedback loop would start to react to variations across C6.
So put a high power ~100Ω rheostat/variable resistor in series with the supply line between your PS and the whole contraption, so the PS is not so stiff and acts more as a current source than a voltage source - only then the feedback loop will get a chance to regulate the pulse width.
Set the PS to 15V and tweak the DTY pot together with the rheostat. If you don't have a rheostat then stretch out a nichrome wire from some household heating element and clamp a crocodile clip on it in various places or try out a lot of different incandescent light bulbs in series (12V and/or 230V ones).
* When two voltage sources (PS and L2 in this circuit) are in parallel, then the source that provides more voltage supplies all of the current.
The lower voltage source normally sinks all of the current if there is no diode in series between these sources to stop it or current sharing resistors to distribute it evenly.
Good Day All:
That is an excellent idea, (Verpies , above)... I was thinking about this exact problem the other day. The incandescent light bulb is excellent because you can use it to regulate (measure) the amperage the circuit is consuming in real time. If the bulb is the automobile type dc, then the bulb will light up when it's current rating is reached.
@Verpies
Whats your take on using a SMPS to do the same thing, ie: variable pulsing the output of the SMPS to mimic the action of just momentarily touching a battery to the Vc & Neg. on the circuit board? Kind of like Akula using a 9volt battery to jump start his latest circuit. The circuit under question could be pulsed and then tuned between the pulses for best results.
Any suggestions for such an application?
take care, peace
lost_bro
Grum:
To calculate the power dissipation in the LED string measure the voltage across R5 the 1.0 Ohm (which represents the current through it and the LED string) and multiply by the voltage across the LED string. This should give you the power in Watts the LED string is drawing.
There should only be a small amount of HF ripple on C3 that would not adversely affect the measurement.
Quote from: Hoppy on 2014.04.22, 16:41:19
@ Wottsup,
I'm using the wind method you suggested in your previous post.
Diodes are exactly as specified by Akula (MBR3545), as this is exercise is to replicate his device as accurately as possible. I do not wish to alter the diodes, just try different coil winding configs as this was the only detail not given to us. The mosfet driver chip mod is IMO valid as there is a push-pull version of his circuit published, so the driver chip is not a significant deviation from this. I have built all three versions.Happy :)
@Happy
I don't understand why when I look up MBR3545 all I find in terms of specs is stud diodes. All of brands I could find are stud diodes. How is that possible that there is no specs on standard fat diode with this model? I am attaching a pdf of the MBR3545 stud type and see there are differences in the specs between that and the one listed on our parts list. But why is the MBR a stud type when in the Akula30 photo, there is no stud diode? So many inconsistencies.
@verpies
Thanks for your reply and I will post more on that later. I would have liked for you to put down the number the scope would have shown in my question, please.
I think guys should switch to a battery feed, not more. The system has to run with quick gain to achieve a total output. The time required to fill C11 and C3 caps and it should just cruise along. It should not require much to start, not more then a 12v battery an off switch and maybe a fuse.
@all
I made a synopsis of the Akula effort up till now that may help clarify things or help put them in perspective. I know you can get discouraged but things are not so bleak.
After this post I will post something about scoping that may help some of you. @Grumage, I asked you about your waveforms in my previous post and it would have been good if you could answer because your answer is tied in with what I am going to discuss with @verpies later on. You don't need to put up the scope shots, just let meknow if with the probe only the waveform is the same as with a ground.
I would say we are in very good shape. With his last Cap Massacre video, it's as if Akula is talking to us and trying to tell us some things. When he had trouble restarting it around 5:56 onward, how many of you thought, "Hey, you have to readjust the pot you just deregulated man" then he readjusts the pot to make it work again. I am sure he did that on purpose to show us the importance. Спасибо
Akula just hatched his last build and video in what, a few days. He did not spend months working his coils. On his last system, he could have removed those 4 nice fat pots if he really wanted to and replaced them with fixed value resistors after he found the best values with his pots and we would never have known it. Akula Methodology 101.
Akulas videos and builds are telling me "Hey man, the coils have to be 1 pri thicker then the 1 sec, don't overdo it on the winds, either both winds over or both off the gap, the general circuit is this, the looping logic is this (great logic indeed), the diodes can be one big fat guy or three smaller more responsive guys in parallel (like in his last video) and lastly, get control over your R variables. Adjust the Rs to set the loop. Once the effect is found, measure the pot resistance values and replace the pots with a good resistor nearest to those values." That's what he did to make the Akula 30. Instead of matching the coil to the circuit, he is matching the circuit to the coil. As long as the base principle is sound and you realize you need to find the R values, the effect he is showing is replicable in a million ways. hehehe
BUT DOES IT WORK???????
Stay tuned folks when Jed says to Eli-may "Eli-may, we'll get that varmint and loop it on home". hahahah
The 4 pot methodology would cut through the chase of getting coil winds perfected to fit a pre-set circuit. Instead of trying 100 coils, try one but with more variables to tune. Otherwise you could be winding coils for the next year. That's what he is trying to tell us.
Sooooooooooooooo, next move should be replace R1, R2, R3 and R5 with 4 comparable pots then find the best finer values. Some here will know more about choosing the right pots. That will give you guys a new way to play but more importantly it will bring us closer to matching the Akula Effect. What can I say any more then that would be redundant.
I am preparing some tests with a flyback ccore (while my ecores arrive) just like in his last video because the core, ecore, ccore should not make that much difference if you employ the base principle. The ecore will just move more flux around but in essence, his last video with the primary pulsing out from both ends for the secondary to catch right away from its ends, tells us the impress is at the first reflection. So that also tells me that in the ecore, each coil has to be on its side of the gap, exactly like in the photo and the ecore mapping I had shown.
I don't know about you guys but this is getting so interesting.
Can't wait to play around with this starting tonight. Can't wait to see one of you guys with the first loop.
wattsup
Quote from: wattsup on 2014.04.23, 12:01:23
@Happy
So many inconsistencies.
wattsup
@ Wottsup,
Yes, and the biggest one being how in his latest videos, he can hardly get a few small LED's to light and yet in his 30W version, he gets a powerful LED floodlight to shine brightly in self-running mode!! ??? C.C
Not so Happy :(
@Happy
Come on man, don't lose heart so soon. You are not getting the point here. He did not make that last video to show a Guinness Record for most output. He did it to show us the finer points in getting it to work and he even cuts up the caps to show us there are no batteries inside the circuit, then he takes apart the coil to show how they were wound and to show how the flux in the core/gap relation are simpler then we thought. He shows us three diodes in parallel for D5 and I am sure there is much more but I have not spent too much time on them to trace the circuit, etc., in order to confirm other similarities between this last device and the Akula 30.
Man oh man, if SM made a video like this instead of the Crapville we have to visit each time, we would already have OU.
wattsup
@ Wottsup,
I sense that he's won you over already.
A further inconsistency: why are there two pots on the circuit board but only one shown on the schematic?
Could the supply rail be 3V - get my train of thought??
Hoppy
Quote from: Hoppy on 2014.04.23, 14:13:37
@ Wottsup, I sense that he's won you over already.
A further inconsistency: why are there two pots on the circuit board but only one shown on the schematic?
Could the supply rail be 3V - get my train of thought?? Hoppy
@Happy
The schematic shows two pots R11 and R15. But those are punny things for the frequency side.
He has not won me over because I still am very conscious that he could have hidden button batteries in a few of those pots, but man oh man, there has to be a limit to the level of paranoia we can handle. Seems like we will never be 100% assured of anything until it is accomplished and repeated as a self-runner. All in good time my friend, just hang in there.
When this project was started, there was no guarantees, no set time table and no special expectations but our ability to sort out what and how we can to make it work. All we can do is our honest to goodness best efforts and have some level of trust that Akula is not taking us down the Yellow Brick Road only to find out the Wizard is a feeble old man. hahaha
wattsup
Hi,
One thing is missing in all these circuits - the Automatic Frequency Control for automatic tuning into resonant frequency.
If you guys will be able design simple and economic feedback based resonant frequency auto tuning circuit, you can replace frequency control with that circuit and have instant resonance in any core/coils configuration.
Please see book attached about that stuff - http://www.tubebooks.org/Books/afcs.pdf
Cheers!
Quote from: wattsup on 2014.04.23, 15:19:55
@Happy
The schematic shows two pots R11 and R15. But those are punny things for the frequency side.
He has not won me over because I still am very conscious that he could have hidden button batteries in a few of those pots, but man oh man, there has to be a limit to the level of paranoia we can handle. Seems like we will never be 100% assured of anything until it is accomplished and repeated as a self-runner. All in good time my friend, just hang in there.
When this project was started, there was no guarantees, no set time table and no special expectations but our ability to sort out what and how we can to make it work. All we can do is our honest to goodness best efforts and have some level of trust that Akula is not taking us down the Yellow Brick Road only to find out the Wizard is a feeble old man. hahaha
wattsup
@ Wottsup,
I'm referring to his the latest 3V flashlight video as translated by Wesley: -
https://www.youtube.com/watch?v=Lpx1SJdXnmo&feature=youtu.be
...but yes, the button cells could also apply to the previous 3-part video with the SMPS. Lots of pots to hide the batteries in maybe.
His skins too smooth to be a wrinkly. ;D
Hoppy
Quote from: Hoppy on 2014.04.23, 15:34:09
...but yes, the button cells could also apply to the previous 3-part video with the SMPS. Lots of pots to hide the batteries in maybe.
First PSU, then capacitors, now pots? Forgive me, but I smell trolling... ;)
P.S> Wesley was not supposed to put latest private video on internet without asking akula first by the way.
But as it is there already, it was all about showing off where Roman is with power output improvement, and with circuit diagram with what he got effect first sent to me by him in private separately.
Quote from: lost_bro on 2014.04.22, 21:49:58
Whats your take on using a SMPS to do the same thing, ie: variable pulsing the output of the SMPS to mimic the action of just momentarily touching a battery to the Vc & Neg. on the circuit board? Kind of like Akula using a 9volt battery to jump start his latest circuit. The circuit under question could be pulsed and then tuned between the pulses for best results.
I was thinking about using pins 8 & 9 to do something like that.
Quote from: T-1000 on 2014.04.23, 15:33:04
One thing is missing in all these circuits - the Automatic Frequency Control for automatic tuning into resonant frequency.
If you guys will be able design simple and economic feedback based resonant frequency auto tuning circuit, you can replace frequency control with that circuit and have instant resonance in any core/coils configuration.
Are you writing about an LC resonance or acoustic resonance?
Hasn't Akula already used the CD4046 PLL to do automatic frequency locking like that in his C-core perpetual flashlight video?
Quote from: T-1000 on 2014.04.23, 15:58:20
First PSU, then capacitors, now pots? Forgive me, but I smell trolling... ;)
P.S> Wesley was not supposed to put latest private video on internet without asking akula first by the way.
But as it is there already, it was all about showing off where Roman is with power output improvement, and with circuit diagram with what he got effect first sent to me by him in private separately.
@T1000,
I'm afraid I smell something different ;)
What power output improvement? I thought the two original 30W circuits were supposed to be the most powerful self-runners shown so far.
Hoppy
Quote from: Hoppy on 2014.04.23, 16:12:48
What power output improvement? I thought the two original 30W circuits were supposed to be the most powerful self-runners shown so far.
Nah, the most powerful replication with same effect was >100W - http://www.youtube.com/watch?v=Du8_moSa7KY This one was auto-adjusted by microcontroller.
And akula's more powerful designs are binded with offshore company(STEHO energy AG) and contract so he got handicapped and won't share at this time for sure. So LED flashligt has no value to him and to who he is working with and he can go until perfection of it.
P.S> Sorry, sometimes I cannot just share circuits without authors permissions and those two cases are inconvenient while knowing how they operate and trying to explain principles to people. It all leads back to Lithuanian Yoke experiment...
Dear All.
Over the last 24 hrs there has been a bombardment of info come in from all areas !! From what I can make of it we don't stand a "Cat in Hell's chance" of making this device work without the "Magic" bit !! Barium Ferrite cores !! But please feel free to disagree !! :)
Remember Floyd Sweet's VTA used Barium Ferrite magnets !!
With that said, who's up for a bit of Alchemy ?? I remember well the School experiment where Sulphur and Iron filings were placed in a test tube and heated creating the compound Iron sulphide. The question is can we combine Barium Sulphate which is readily available and create a compound with Iron?? I may well be barking up the wrong tree but it's an idea !! I have the furnace and Crucibles still so could easily cast some cores !!
As I posted earlier in the thread, I have to hand a Barium Ferrite rod. Today I placed a coil around it and had my GM counter looking at it. Very occasionally it would crackle (at certain frequency). How much energy is required to trigger a better response ??
Lots of questions !!
Cheers Grum.
Quote from: Grumage on 2014.04.22, 18:28:45
Dear All.
Well, another fruitless day !! I have rewound a transformer with 6 turns primary and 18 turns secondary. I have physically gapped the two halves with relation to Dr Andrey Melnichenko's suggestions. Creating two separate halves. Primary on one "triblet" secondary on the other.
At the moment we have a PWM with a fixed duty cycle. As far as I can see the "Duty" pot fitted to the boards we have is primarily a "high voltage cut off" ?? Correct?? So a simple question. can we reduce the pulse width ?? ATM I am pushing over 2 A @ 12V this can't be right ?? Ok, yes, that is only 24 Watts. But I think the LED's could be lit by a fraction of that by having a narrower pulse width !!
I am getting to the point of finding a suitable wall to bang my head against !! :D
Cheers Grum.
Hi Grum,
well as i had the same problem (14V @ 10A) i decided to cheat a little and using an (all positive) 5V square wave from a FG on pin 3 to switch the output of the TL494.
This way i can use the duty cycle and/or frequency of this square wave to reduce the on time of the gate signal to where the average current input is acceptable.
Perhaps its an option for you to to be able to make some longer measurements.
Video here: https://www.youtube.com/watch?v=QoUmdTF6mtQ&feature=youtu.be
Regards Itsu
Quote from: Itsu on 2014.04.23, 19:38:50
I decided to cheat a little and using an (all positive) 5V square wave from a FG on pin 3 to switch the output of the TL494.
If you don't apply the signal from the FG to pin3 through a small diode (
cathode (http://www.overunity.com/11142/background-discussion-of-tts-pyramid/dlattach/attach/117188/) to pin 3), then you risk damaging the TL494 permanently.
Without a diode, the original feedback loop will be disabled, too.
Feeding pin 3 with more than +3.5V is not advisable, either. Feeding it with more than +5V or less than -0.5V is damaging.
For those that do not have a signal generator and still want to cheat:
Applying 0V to 3.5V DC to pin 3 through a small diode (
cathode (http://www.overunity.com/11142/background-discussion-of-tts-pyramid/dlattach/attach/117188/) to pin 3) will decrease the duty cycle of
every output pulse that the TL494 generates.
The 0V to 3.7V DC can come from a 470Ω pot in a voltage divider like this:
(http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12988)
Quote from: Grumage on 2014.04.23, 17:55:37
Dear All.
Over the last 24 hrs there has been a bombardment of info come in from all areas !! From what I can make of it we don't stand a "Cat in Hell's chance" of making this device work without the "Magic" bit !! Barium Ferrite cores !! But please feel free to disagree !! :)
Remember Floyd Sweet's VTA used Barium Ferrite magnets !!
With that said, who's up for a bit of Alchemy ?? I remember well the School experiment where Sulphur and Iron filings were placed in a test tube and heated creating the compound Iron sulphide. The question is can we combine Barium Sulphate which is readily available and create a compound with Iron?? I may well be barking up the wrong tree but it's an idea !! I have the furnace and Crucibles still so could easily cast some cores !!
As I posted earlier in the thread, I have to hand a Barium Ferrite rod. Today I placed a coil around it and had my GM counter looking at it. Very occasionally it would crackle (at certain frequency). How much energy is required to trigger a better response ??
Lots of questions !!
Cheers Grum.
Interesting thought and something that JB made a strong point about, as being the secret to the Kromrey Converter and other overunity devices, so who knows. Barium ferrites were widely used back in the 1960's / 70's but is not used in today' ferrite products. The Chinese claim that they can supply barium ferrite products but I have yet to hear of anyone that has actually proved beyond doubt that the Chinese products are in fact as claimed.
Hoppy
Latest OU video from TK - enjoy.
http://www.youtube.com/watch?v=Frp03muquAo
Hoppy
Did you notice this in his vid
The only way to get neg bemf is to pulse a coil pos.
anyway
later
dave
One more then I'll shut up
Why would he even use the term negative bemf if he did not understand that a pos bemf and neg bemf can be extracted from a coil
Iv heard it said that even he doesnt know how his circuits work, I tell you now he knows exactly how they work
He's using charge separation,
Hi guys,
Shortly after I posted in the "Akula0083" thread at overunity.com, a member of this Forum (who knows my work from the past) made a request if I would be willing to assist in "steering you all in the right direction", as he put it.
I decided to 'take a break' from my current project and made a detailed analysis of the published circuit which all of you are trying to replicate.
In addition, I have re-drawn the diagram with my corrections added.
I released those files about 2 weeks ago but so far only a few members had access to them.
I have now decided to publish them here, hoping that perhaps some of you may benefit.
Since my analysis (report) is 7 pages long, it is written in a document format.
My circuit diagram is drawn in 'Protel' format but I have converted it to PDF for you to 'read'.
Both files are attached.
Further, I now have additional (important!) info about the transformer and the winding method.
In addition, may I also suggest you check out the following:
http://www.youtube.com/watch?v=F8c82ABs02M
http://www.ferromagnetic-energy.ru/core_of_the_physical_effect/sequence/
http://www.ferromagnetic-energy.ru/
http://www.ferromagnetic-energy.ru/core_of_the_physical_effect/
I have also attached files related to the links above.
Cheers,
Les Banki
Quote from: Les Banki on 2014.04.24, 04:05:22
I DISAGREE with the entire statement!
And I claim that my statements are correct and you are wrong.
I will present referential, logical and experimental evidence to demonstrate that you are in error.
You've made a mistake with me by confusing my analysis of the circuit (as drawn) with your expectations of how it is supposed to function.
Quote from: verpies link=http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg394475/#msg394475 date=1395850126
On this schematic, the voltage signal taken from R5 must become negative to activate the 2nd error amplifier of the TL494 (pins 15 & 16) via R7.
If R5 is non-inductive then this would mean that the current through these load LEDs needs to flow in reverse for this to occur.
Quote from: Les Banki on 2014.04.24, 04:05:22
First of all, I cannot see how the voltage across R5 can go negative!
For the purpose of evaluating the correctness of my statement, the origin of the negative signal is irrelevant. That signal on pin 15 must become negative to activate this TL494's comparator as shown on that schematic, period. The correctness of this schematic was not the subject of my statements. I analyzed that schematic as stated, and I am defending this analysis now.
There is even experimental evidence to prove my statement as correct.
Itsu please help me out here and show this guy that experiment you've made about negative signal on pin 15.
I could answer how that signal can become negative but that is a different issue. It's not an issue of my statement's correctness but it is possible.
Quote from: Les Banki on 2014.04.24, 04:05:22
Sure, it is possible to get some switching transients (spikes) with negative going portions
Not only that. The origin of these negative pulses is beyond the scope of proof dealing with the correctness of my statements.
Quote from: Les Banki on 2014.04.24, 04:05:22
...but those will be 'clipped' by the internal circuitry of the IC.
You are wrong again. The error amplifier/comparator input can legally accept input signals up to -0.3V below ground.
The TL494 datasheet clearly on the 1
st page that:
"The error amplifiers exhibit a common-mode voltage range from –0.3V to V
CC–2V"
Quote from: verpies link=http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg394475/#msg394475 date=1395850126
The 1st error amplifier of the TL494 (pins 1 & 2) is configured to react to quickly falling edges at R5 (via R7). The R11 pot determines how steep these falling edges need to be in order to trigger this 1st error amplifier.
Quote from: Les Banki on 2014.04.24, 04:05:22
I DISAGREE with the entire statement!
You are wrong again.
This video (http://youtu.be/exIFDDgSrng?t=1m32s) presents clear experimental evidence that the 1st error amplifier reacts to steep falling edges at R5 on that schematic.
BTW: When an error amplifier has any feedback (even AC feedback) between its output and inverting-input, then you cannot call it a "comparator" anymore.
I have successfully defended all of my statements.
My full analysis of the behavior of these error amplifiers is in this thread - not on overunity.com.
For example, in the 3 points listed in
Reply #844 (http://www.overunityresearch.com/index.php?topic=2358.msg37907#msg37907) and in many others.
Quote from: Dave45 on 2014.04.24, 01:57:16
Did you notice this in his vid
Yes, but that's the wrong thread for it.
This video is being discussed in another thread (http://www.overunity.com/14524/3v-ou-flashlight/msg398781/#msg398781).
Quote from: verpies on 2014.04.24, 10:14:38
Itsu please help me out here and show this guy that experiment you've made about negative signal on pin 15.
Right, well i never made any video's when putting a negative voltage to pin 15, so nothing to show.
It starts somewhere here in this thread some weeks ago:
http://www.overunityresearch.com/index.php?topic=2358.msg37153#msg37153
I put a variable negative voltage to pin 15 (between 0 and -5V) compared to ground, and the reaction was that around
-0.8V the output signal on pin 11 went instantaneously from a duty cycle of 44% to 0%.
Regards Itsu
@all
Was going to post this last night but fell asleep on my chair (shit) so I am posting this morning before I read @Les Bankis' post and stuff and while I nurse a back ache. hahahahahahahahah
Something we missed in the coils. Looking at Akulas Cap Massacre Video (ACMV) when he takes the coil apart but I did not notice this before but that ccore has three coils. I thought I was hallucinating because how late it was, but no.
One the square sided gap you have a primary of 11 1/2 turns going across the gap to the other end.
On top of that you have a second layer pick-up coil wound in the same direction as the 1st layer primary.
On the round sided gap you have a secondary of about 20 turns.
So you have 5 wires leaving the ccore. Hmmmmmmm. 5 wires just like in the Akula30 core photo.
So then looking again at the Akula 30 photo of the coil again, looking at the bottom left terminal there is what seems to be two wires going there. I had looked at that many times, saw there was something but could not officially put my finger on it until he put out the last ACMV video.
That would make the winds as follows.
Primary starts bottom left down and 5 1/2 turns of a thicker wire stop at gap.
On top of this start again bottom left down and 5 1/2 turns of thicker wire also what seems to be a totally different shade of beige.
Secondary of thinner wire stars bottom right down for X turns till stop at gap then continue with secondary layer from gap back to the right side.
This gives 5 wires coming out of the coil. The copper strip is not soldered, it's just there as a shield so his diagram is again not conveying the right information. Man oh man. The pains we go through.
Looks like everyone is in argue mode. Let's just all calm down and move forward.
The only way we will really know is not only with the AK30 diagram but matching those to the actual circuit which is kind of hard since only one photo is available and nothing else, no video of the AK30.
wattsup
Quote from: wattsup on 2014.04.24, 13:07:46
On top of this start again bottom left down and 5 1/2 turns of thicker wire also what seems to be a totally different shade of beige.
Secondary of thinner wire stars bottom right down for X turns till stop at gap then continue with secondary layer from gap back to the right side.
Bingo!
That would make much more sense. These Shark devices look to be based on the Melnichenko device.
Something else to keep in mind. The wire diameter may not be critical, but the insulation thickness could be--interlayer capacitance. If the magnetic fields move at some fixed rate, the distance from the source at which these fields cut through the conductors would create a time delay--so if the windings are packed too tight, the electronics may not be able to run at high enough frequencies to provoke the desired effect.
I have a bunch of cores that were salvaged from other devices.
Is there an easy way to determine the resonant freq of a core if I have no information about the material of the core?
I have an inductance meter if that can be of use.
Thanks for any help
DonL
Quote from: dllabarre on 2014.04.24, 13:51:03
I have a bunch of cores that were salvaged from other devices.
Is there an easy way to determine the resonant freq of a core if I have no information about the material of the core?
I have an inductance meter if that can be of use.
Thanks for any help
DonL
Dear Don.
There is a simple way of ascertaining the AL value of your cores. Just wind ten turns of wire around the core, measure the Inductance in Henries, milli or micro. And use this calculator to tell you it's AL.
http://powermagnetics.co.uk/calculator
You can also find, with the use of a permanent magnet, whether you have a Ferrite of low magnetic Remanence. (It will hold no lingering field when applied to a Compass after the magnets removal) Or the contrary, does hold some magnetism after the removal of the permanent magnet !!
Hope this helps ??
Cheers Grum.
Dear Wattsup.
Would you be kind enough to draw your findings schematically ?? I myself find it much easier to understand than a written description !! :)
Cheers Grum.
Quote from: Grumage on 2014.04.24, 14:37:08
Would you be kind enough to draw your findings schematically ?? I myself find it much easier to understand than a written description !! :)
Grum, I'm not entirely certain this is what he is alluding to or if this is just a minor improvement using multiple windings. It's interesting to ponder regardless.
I was thinking because the three primary windings in the second drawing are all in series and the dot notation for two of them are the same, one could deduce those two to be a single coil. This would give the mechanical appearance of just two primary windings--one twice the size of the other.
Then of course we have a slightly larger variation. Overall, the schema looks similar.
Quote from: Matt Watts on 2014.04.24, 15:34:42
Then of course we have a slightly larger variation. Overall, the schema looks similar.
This one has only one diode (not counting (LEDs) no transformer (just two separate inductors) and 2 switching power transistors.
I'd say, it's very different.
Quote from: Grumage on 2014.04.24, 14:37:08
Dear Wattsup.
Would you be kind enough to draw your findings schematically ?? I myself find it much easier to understand than a written description !! :) Cheers Grum.
@Grumage
For you, I would even draw a Mona Lisa but cannot guarantee she will have the same smile. hahaha
Meanwhile I made a small avi on the wind method here: http://purco.qc.ca/ftp/Wattsups%27%20stuff/misc/
Give it some time to download. I did not want to put this video in my youtube channel. hahaha
We have to look closer to the actual board photo and not only the schematic.
Why do you think he scratched out the IC number on the original board then identifies it in his schematic? hahahahh
wattsup
Quote from: verpies on 2014.04.24, 18:16:19
This one has only one diode (not counting (LEDs) no transformer (just two separate inductors) and 2 switching power transistors.
I'd say, it's very different.
Yes, in retrospect they are different. Can we separate the differences from the similarities to find a common theme though? Or a way to differentiate what is critical and what is not?
I suspect these schematics are Roman's way of mentally prototyping his ideas; I doubt very much he has posted any schematics that exactly match devices he has demonstrated as working units. I say that based on the Steho connection.
Quote from: wattsup on 2014.04.24, 19:40:11
Meanwhile I made a small avi on the wind method here: http://purco.qc.ca/ftp/Wattsups%27%20stuff/misc/
Give it some time to download. I did not want to put this video in my youtube channel. hahaha
Other than your vice moving around and the wires not staying put, this is a very helpful video. Thanks for making it.
So attached, I have a shot of the final wind with the wires enumerated. Can you describe how they hook-up or simplify down to just four wires? Are #2 and #3 hidden?
Quote from: Matt Watts on 2014.04.25, 01:23:32
Other than your vice moving around and the wires not staying put, this is a very helpful video. Thanks for making it.
So attached, I have a shot of the final wind with the wires enumerated. Can you describe how they hook-up or simplify down to just four wires? Are #2 and #3 hidden?
@Matt Watts
No they are visible and together. There are 5 connections just like in his last video except that in the last video he is pulsing the two ends each with one mosfet and the center tap is going somewhere that I have not mapped but the center tap needs constant + so the 2 end mosfets can switch in and out, either together or at a slight delay.
@all
The new wind method now explains to me why in the AK30 the top 5 /12 turns coil wire color was slightly beige compared to the wire underneath it. It's because they were two winds in bucking mode with center tap. That was not in the schematic.
The AK30 board shows a SemiSouth type mosfet so why did he indicate IRF3205. I have a few of these on hand. If I remember correctly these types do not have an internal diode. There is no resemblance whatsoever to the IRF3205.
My main question is when you guys started looking at the schematic to plan out these circuit boards, didn't anyone look at the actual board or did you guys just rely on the AK30 schematic? I kind of tuned out during the very demanding pages of circuit elaboration (was and still am on the FTPU) but maybe I should have taken more steps to stop you guys and make sure the board coincides with the schematic way before a circuit was designed. Don't get me wrong because I am not blaming anyone but myself but man oh man, the things we have to go through to make a few extra watts. hahaha
If Akula schematics cannot be trusted, then when do we know what to trust and what not to trust? Things like the TL4941N may not even be used on his board because when you look at the peripherals and what is going to the pins, they do not match the schematic. Plus why did he scratch out the IC number and then tell us it is a TL4941N in the schematic?
@T1000 indicated that Akula had a contract and was held to an NDC so this is why he is just playing around with these other designs for what? What is the use of showing designs, providing inconsistent schematics and all the rest of this whole thing if he will not take 10 minutes to talk in direct terms about the builds.
If the schematic has so many inconsistencies compared to the board, then we must also consider that Akula could have provided the schematic with values that would ensure builders are at 300KHz or below and that duty cannot be controlled as well. That ensures him the publicity he is obviously looking for while we would never get it since in my experience, such devices can "resonate" at lower frequencies, yes, but only to a small percentage of its potential. At only 5 1/2 or 11 turns L1 puling most probably in the 1-5 MHz range will this device really come to life, but we would never know it. So we have to chuck out the schematic, produce the Ecore as we now know, use the general schematic for the ecore side and run it via a mosfet and an FG that can get much higher switching in order to find the proper self-running frequency.
But please guys, let's not lose heart and keep concentrated on the build and figure this to together because I do not have the talent to do it alone but I do have the analytical skills to figure out the functionality.
Like this next point: Look at C11 the R1 0.22 resistor and the C2 100n cap. Why the hell is that C2 cap there? Seems to me with C11 already there, there is no need for C2. What can C2 do more then C11? But if between R1/C2 and L1, there was a PNP mosfet switching that front end of L1, maybe C2 would then have a use but since Akula took that out of the schematic, he had to patch up that part as he did. Does this make any sense to the more proficient OUers here?
So next damn damn damn damn step is to map the official circuit board to the schematic and see where Akula deviated from reality so we would just be thrown off the real effect.
wattsup
Wattsup said:
QuoteLike this next point: Look at C11 the R1 0.22 resistor and the C2 100n cap. Why the hell is that C2 cap there? Seems to me with C11 already there, there is no need for C2. What can C2 do more then C11? But if between R1/C2 and L1, there was a PNP mosfet switching that front end of L1, maybe C2 would then have a use but since Akula took that out of the schematic, he had to patch up that part as he did. Does this make any sense to the more proficient OUers here?
C11 will have a rather high esr, the film type C2 will be more effective at snubbing the HF transients, having a very low esr. For high frequency work it is common practice to shunt a large electrolytic with a small low esr cap such that the two present a very low impedance in a circuit. Value in uF is not the whole story with capacitors, one needs to consider esr and effective inductance.
Since the C2 is on the other side of the 0.22 ohm, it forms a first filter for the HF transients from L1. It may also produce a high Q resonant behavior with L1 being somewhat decoupled from C11 by the 0.22 ohm. Additionally, L2 transients enter the C11 node directly via a 1 ohm resistor, C2 and R1and R3 form a small filter that might prevent the L2 transient from falsely retriggering the FET due to high dV/dt.
There are many not so obvious subtleties to Akula's circuits. We need to unravel them all. From my experience in switchmode designs, Akula seems to pay attention to the subtleties. e.g. R2, C4 across D6 and C5 across R5. Some of what he does also puzzles me.
I agree that we need more precise definition of components against what we see in the videos.
Thanks for your winding clarification video. I use a small dab of superglue to bind the first turn to the core, this makes subsequent turns easier to tightly apply. A dab on the last turn holds it all tightly together while tape is applied. Small plastic spring clamps keep your fingers out of the glue while it sets up.
Quote Matt Watts ....If Akula schematics cannot be trusted, then when do we know what to trust and what not to trust? Things like the TL4941N may not even be used on his board because when you look at the peripherals and what is going to the pins, they do not match the schematic. Plus why did he scratch out the IC number and then tell us it is a TL4941N in the schematic?
I have noticed that there are too many coincidences in the video just made and the old photo posted ... something not quite right because if Akula should make another video why he is so careful to place each thing in the same old position?
Find the differences....
Quote from: ION on 2014.04.25, 13:04:00
C11 will have a rather high esr, the film type C2 will be more effective at snubbing the HF transients, having a very low esr. For high frequency work it is common practice to shunt a large electrolytic with a small low esr cap such that the two present a very low impedance in a circuit. Value in uF is not the whole story with capacitors, one needs to consider esr and effective inductance.
Yes
I would again like to thank Groundloop and Peterae for making the PCB available FOC.
I am at the moment very tied up with work projects but will continue to monitor the forums for any new developments from the other members who are populating and testing their PCB's and especially those looking into Akula's 3V OU flashlight device. I'm fairly convinced that Akula used a hidden power source, possibly small button cells hidden in the pots for these devices but I'm hoping to be proved wrong.
Hoppy
Anyway a good actor could win an Oscar on Youtube.... O0
https://www.youtube.com/watch?v=DmeUuoxyt_E
The only magnetic field working for me for now ... with 3 w is this joule thief "Lasersaber" lights 54w light 6 leds enought (each 9w led)
As i was going nowhere with my current draining circuit, i tried Wattsup his transformer configuration idea.
As i understand it, the transformer configuration will be like in the transformer layout shown below.
To fill the coreformer, i ended up with 18 turns L2 (2*9) and 2*6 turns on both L1's
I tried the 4 possibilities to hook up this new transformer but all show an even higher current consumption.
The old transformer pulled 14V / 10A without activating the bench PS current limiter, but all the 4 possible connections on the new transformer layout is causing the bench PS current limiter to be activated (10A).
So i had to reduce the duty cycle to stay within acceptable current consumption, so i tried Black_Birds solution mentioned some weeks ago (3.9K resistor from pin2 to ground and a 20K pot between pins 2 and 3).
This did not work then (inverted output signal due to the use of the transistor driver), but works ok now with the non-inverting MOSFET driver.
I can now regulate the duty cycle between 0 and 44%.
Setting the duty cycle to 10% reduces the current to about 1A which is not causing any heating up.
New measurements are taken (current through R1 (C2 removed) and R3 = L2 and both L1's) and shown below.
I tried to use the new transformer in "flyback" mode and there was 1 combination out of the 4 which shows this (see screenshot).
Anyway, no abnormalities seen during a long evening testing.......
Video here: https://www.youtube.com/watch?v=1RIiL8T90qs&feature=youtu.be
Regards Itsu
@All
regarding my previous comment ... I've thinking a lot and the only reason why Akula has provided everything exactly as in the past can be a form of visual perspective or pleasant from a distance so that everything looks so very cool ...
or a form of superstition on the continuous operation of his invention also remember that is was afraid to stay close with the camera because there was a big electromagnetic field..
or the magician does the trick as when preparing the show ...
Meanwhile, I'm comforted because I've understood what were meant when were talked about the gyroscopic effect of TPU by Steven Mark ... with these other videos .. :)
http://www.youtube.com/watch?v=y8NBNkSXtOg
Quote from: Itsu on 2014.04.26, 10:53:19
So i had to reduce the duty cycle to stay within acceptable current consumption, so i tried Black_Birds solution mentioned some weeks ago (3.9K resistor from pin2 to ground and a 20K pot between pins 2 and 3).
This method of duty cycle reduction affects the operation of the feedback loop involving pin2.
The method of limiting the duty cycle by
injecting (http://www.overunityresearch.com/index.php?action=dlattach;topic=2358.0;attach=12988) 0 - 3.5VDC voltage into pin3 through a small diode - does not affect the operation of the feedback loop.
v3 screenshots front/back.. back3.jpg flipped to match front.
Akula0083 - Lantern No 3 - Disassembly - Full Length Video
https://www.youtube.com/watch?v=E9Tw66wrW0A (https://www.youtube.com/watch?v=E9Tw66wrW0A)
Akula0083 - Lantern No 3 - Demonstration (first version with power supply)
https://www.youtube.com/watch?v=ZmP_Y0Le6Wg (https://www.youtube.com/watch?v=ZmP_Y0Le6Wg)
Wesley Translate's Akula video #15: Akula's TPU part 4
https://www.youtube.com/watch?v=1BNgo09N0yM (https://www.youtube.com/watch?v=1BNgo09N0yM)
Quote from: 4Tesla on 2014.04.27, 00:17:51
v3 screenshots front/back.. back3.jpg flipped to match front.
That's a good start. Any chance you can create an overlay of the two views aligned and scaled?
I suspect we could determine what the chips must be if we fully reverse engineer these images and draw possible schematics.
Quote from: Matt Watts on 2014.04.27, 05:40:00
That's a good start. Any chance you can create an overlay of the two views aligned and scaled?
I suspect we could determine what the chips must be if we fully reverse engineer these images and draw possible schematics.
I tried to overlay, but couldn't. Maybe someone better at photoshop. Maybe we should invite magpwr from over on ou.com to see if he would be interested in making schematic.. He is very good at reverse engineering, but hand draws schematics. But we can see if verpies or someone can make digital version. If you all interested I can send a pm to magpwr.
Edit: I sent a PM to magpwr.
4Tesla
If he wants to join, i need an email address O0
Thanks
Peter
I will look at the layout today and see if i can work a diagram out although i think it's going to be hard to see where the tracks are cut if those image above are the best we have O0
Quote from: Peterae on 2014.04.27, 06:35:24
4Tesla
If he wants to join, i need an email address O0
Thanks
Peter
I will look at the layout today and see if i can work a diagram out although i think it's going to be hard to see where the tracks are cut if those image above are the best we have O0
Thanks Peter! Perhaps someone else can get higher res screen capture. I'm just using the print screen key on my pc.
Quote from: Itsu on 2014.04.26, 10:53:19
I tried to use the new transformer in "flyback" mode and there was 1 combination out of the 4 which shows this (see screenshot).
That does not look like the "flyback" mode, because absolute values of currents in |L1| & |L2|
both increase at the same time.
In a "flyback" mode the absolute value of |L1|'s current should start to increase with the rising edge of the gate pulse (it does) and the absolute value of |L2|'s current should start to increase with the falling edge of the gate pulse (it doesn't).
P.S.
Increasing absolute value of a
negative current, looks like a falling edge on a scope (albeit going below zero). In "flyback" mode L2's current is always negative in reference to L1's current as "seen" by your CSRs, thus leaving L2's current channel non-inverted, helps with an intuitive visualization on the scope.
I've been reviewing the videos and although i think it's certainly possible to draw a circuit, it looks like a lot of work to me, i think i am going to pass that onto someone else, but i did reach a conclusion, there's 2 identical circuits on the board, each TL494 drives a separate winding from what i can see assuming it is a 494 which seems probable, one immediate difference is there a much larger cap near pins 5-7 on one chip than the other., i suppose everyone worked this out anyway O0
Hello everyone
today I wanted to see if there 'a chance to hope that these replicas is real ... unfortunately I have bad news ....
I've seen the video Xname41 Ruslan Kulabuhov and even here there' a good question mark on the validity of the real operation...
look what I've find...
This the video
http://www.youtube.com/watch?v=01Sf1pZuHn0
he says that there nothing hide under the table and when the lifts it to show that there is anything under appear the black box (box of emergency) but does not show what there is in the box.... ah ah ah ;D
Conclusion he has an inveter connect to the hidden battery... :P
Quote from: TutorialFE on 2014.04.27, 11:37:02
Hello everyone
today I wanted to see if there 'a chance to hope that these replicas is real ... unfortunately I have bad news ....
I've seen the video Xname41 Ruslan Kulabuhov and even here there' a good question mark on the validity of the real operation...
look what I've find...
You might create different topic about it as there was series of videos and also one of them was in Riga city where this guy picked working generator from the ground to show no hidden wires...
Hi there,
Here is yesterdays conference recording with Roman (akula) ,myself and the rest of people - http://webfile.ru/63bd3cd22caa0cb0343e3c0c52158678
Please save it and try to translate, it contains important information.
Cheers!
Sorry for this long post but I think there is much to cover.
@Happy
If you want I will buy your circuit, plus labor and transport costs to me. I am sorry to here you will not be talking part for now.
@ION
Thank you very much for your kind explanation as well as confirmation that HF filtering is suitable, but maybe I can add, especially if your plan also included a mosfet before L1.
If I was just returning L2 via R3 into C11, transients or not, I would not care and let them enter L1 again. I would not want to filter anything at all since all change in the primary would be reflected to the secondary. But if I wanted to place a mosfet before L1, maybe some filtering to feed the mosfet drain is suitable and that is the point here. Why put that small cap if not because you had a mosfet there but did not include it in your schematic. With all the work involved in making the schematic in the first place, it would be very easy for Akula to have forgotten to remove the C2 from the plan when he removed the mosfet.
@all
The only confirmation we have is in two builds, the winds are as discussed in my previous post. I am so sorry I could not find that sooner. I had always though that front left wire was not at the right angle and color but only once the v3 video came out that this became more then evident.
We know the wind method (L1/L3 - L2), we know the pulse method (mos at ends of L1/L3), we may know the truth about the return method (D5/D6/L2/R3), but we do not know how to drive it. At what frequency? I am now convinced it is above the 300K that Akula pegged (or precrippled) this in the schematic while probably confiding in a friend "Don't worry, as long as they keep using that IC below 300k, they'll never get it!".
In the AK30 circuit board photo we see one IC, one SemiSouth type Jfet and a regular mosfet. There are not three mosfets like in the schematic.
On the schematic, why is the IC pin 5 going to a cap of 3nf when even the graphs for this IC do not show anything higher then 1nf. Why 3? How does this impact operation parameters? Everything has to be looked at guys.
Also on pin 6, why is R15 1K and R16 a 10k Pot when this line may be better in the 500k range? Please see image (CT/RT) below to look at this question and advise.
D6 on the actual board is a 630 volt or more 100nf capacitor. Why such a high rating?
The feedback loop is taken at a point that never made sense to me either. Why would you take that signal after the load where voltage would be zero or near zero, then pass it through additional resistors before sending it to the IC? Look, I am not well versed in these matters that I usually leave entirely to others to work out with better experience in circuitry.
But in a nutshell, this basically leaves us to our own devices to figure out so all component values around the IC would need to be re-calculated and/or re-confirmed as if we were designing the circuit from scratch while including a mosfet before the L1.
The two mosfets can pulse off the same signal to both gates to start then we can work on producing a very slight delay between them.
Change in a coil occurs as soon as you connect the first side to any potential, that potential permeates the complete length of the coil. So if you connect one side, let it permeate to the end, then disconnect it while applying the other side, now the other side will permeate the complete length, etc. This produces change in the complete length of the coil wound on the core which is exactly what is required to maximize and/or reflect those changes to the secondary at full force. This will use the coil to its fullest extreme.
@Itsu
Thanks for showing your tests with the variation. However, I must explain to you that with the recent findings the ecore has 5 wires, not four. That means the 5th wire requires some new considerations and the L1 most probably requires a mosfet before the L1.
I attach a diagram of Akulas yoke circuit that has the two mosfets on each end of a two wind primary. I think this yoke device seems to be the one Akula had licensed out and then silenced but it makes so much more sense to me.
In short, pulsing a coil DC from only one side of a coil will always produce the Half Coil Syndrome (HCS). Telsa then invented AC but this did not fix the problem completely. Instead of having one half of a coil undergo the same pulse change while the other side stays mainly static, AC is basically alternating that one side but the other half is still static. I am sure Telsa was not 100% satisfied with AC even if it did great good for society he must have known there was something much better. The only real way to overcome HCS is to pulse the coil from both sides. This way the complete length of the coil undergoes a complete change at pulse on and at pulse off and not only half the coil. This increases the dynamic changes in flux that impart more to the secondary. I can go deeper into the HCS later but it involves new notions of how coils work which is not important here. Akula probably does not know this as well and only sees a working effect that he tries to explain in so many ways.
To see this actually happen, take a variac and put a pulse generator across each end. Then put your scope probe (no ground) on the sweeper blade and turn the handle. This will allow you to see how the pulse energy is dispersed inside the variac coil where you will soon realize that 75% of the pulse energy was dissipated in the first half of the pulsing side of the coil. Funny though the higher in frequency you go while sweeping, you will now see nodal peaks while sweeping the handle but the signal will be more homogeneous throughout the coil.
We all like to think that the pulse enters the coil at a power of 10 and exits at a power of 10. No way, every coil has HCS so imagine your 10 turn one layer primary is pulsing across a gap located between the 5th and 6th turn. By the time the pulse gets to the gap, the remaining 5 turns will only dissipate 25% (or less) of the impulse. So what do you think this does to your pulsing plans? Why do you think Akulas' V3 just lights some leds. hahaha
We ain't out of the woods, but we ain't as lost as it would seem. Let's keep our chins up and keep at it.
I'll be on my bench all day today. Hallelujah.
wattsup
Quote from: verpies on 2014.04.27, 09:06:17
That does not look like the "flyback" mode, because absolute values of currents in |L1| & |L2| both increase at the same time.
In a "flyback" mode the absolute value of |L1|'s current should start to increase with the rising edge of the gate pulse (it does) and the absolute value of |L2|'s current should start to increase with the falling edge of the gate pulse (it doesn't).
P.S.
Increasing absolute value of a negative current, looks like a falling edge on a scope (albeit going below zero). In "flyback" mode L2's current is always negative in reference to L1's current as "seen" by your CSRs, thus leaving L2's current channel non-inverted, helps with an intuitive visualization on the scope.
Ok, thats clear.
I used your 5V circuit presented
here (http://www.overunityresearch.com/index.php?topic=2358.msg37989#msg37989), and it works even better then my earlier (Black_Bird) solution plus it does not interfere with the pin 3 function, great.
I continued to build yet another transformer layout, now consisting of:
L1 = 27 turns (1 layer) AWG 18 magnet wire (across the whole former)
L2 = 81 turns (3 layers of 27 turns) AWG 18 (1mm diam) magnet wire
So keeping the 1:3 ratio
Current consumption is still the same 14V @ 10A with max (44%) duty cycle (no current limiter kicks in however).
Reducing the Duty Cycle to 30% pulls still 5A, and only when reduced to 12% its acceptable at 1A, see the both screenshots below for the currents through L1 and L2 (flyback mode)
The extra in the coils is that i have put in copper strips at the beginning of L2 (inside) and the end of L1 (outside).
Both copper tape strips are almost 1 turn, with a gap of about 3mm.
Regards Itsu
Thanks T1000 i have uploaded the file below to store a copy
Quote from: T-1000 on 2014.04.27, 12:24:31
Hi there,
Here is yesterdays conference recording with Roman (akula) ,myself and the rest of people - http://webfile.ru/63bd3cd22caa0cb0343e3c0c52158678
Please save it and try to translate, it contains important information.
Cheers!
Quote from: wattsup on 2014.04.27, 13:40:00
Sorry for this long post but I think there is much to cover.
@Happy
If you want I will buy your circuit, plus labor and transport costs to me. I am sorry to here you will not be talking part for now.
wattsup
@ Wottsup,
I'm afraid its too beaten-up now to be much good for experimenting; cut tracks & damaged pads etc. I have removed all the primary components and will return to this project should any developments suggest that Akula really does have something concrete to offer in the form of a self-runner. At this point in time, I'm fairly convinced he is playing games with us. Best for you to build on breadboard or strip-board for experimentation purposes as I did originally before I populated the PCB. A PCB is not a good medium for experimentation.
Hoppy
Quote from: 4Tesla on 2014.04.27, 05:57:23
I tried to overlay, but couldn't. Maybe someone better at photoshop. Maybe we should invite magpwr from over on ou.com to see if he would be interested in making schematic.. He is very good at reverse engineering, but hand draws schematics. But we can see if verpies or someone can make digital version. If you all interested I can send a pm to magpwr.
Edit: I sent a PM to magpwr.
How does this look ( Ref. http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg400046/#msg400046 )
Dear All and particularly Verpies.
I am presenting a video update and a few questions.
https://www.youtube.com/watch?v=3y3Z6Bu23KE
I think I might have something not quite right ?? There is a large DC voltage level associated with L2!! Transformer is wound T-1000 style.
Cheers Grum.
Hi Grum,
both R17 (R1) and R19 (R3) are at the DC level of C11 meaning around the supply voltage.
Here a scopeshot of my DC levels on R1 (R17) blue and R3 (R19) green (could be either side of these resistors) compared to ground:
Regards Itsu
Well here it is!! :) 8)
Both chips are TL494
Started new thread for this device:
http://www.overunityresearch.com/index.php?topic=2420.msg38106;topicseen#msg38106 (http://www.overunityresearch.com/index.php?topic=2420.msg38106;topicseen#msg38106)
The GIF is corrupt.. I can't open on my PC. Can anyone open and save as a jpg?
Quote from: e2matrix on 2014.03.03, 18:26:29
We all know Wesley right? The guy who went to visit Kapanadze in person. Akula is one of his friends (or fellow experimenter) who Wesley has worked with a lot. In my book Akula has high credibility and I don't believe what wattsup is presenting here is any proof at all of fakery. Where wattsup sees wires and holes seems like imagination getting carried away to me. Sorry wattsup but I am not buying this hidden wire conspiracy this time either. Could we ask Akula to do a better video? Sure but then someone would come along and say he edited out the hidden wires. Not hard at all to do with a video editor. I'm not against asking for another better video though ....
I'll add a couple more thoughts too. Most people are terrible at doing video's. Most move the camera around too fast and too much. Also don't you think if someone was going to go to all the trouble to build that circuit and then to fake a self runner they would have done a much better job of hiding wires and holes if those were really part of faking a self runner? I know I wouldn't want to get caught making a fake and stand losing all credibility .... only makes sense such a person would cover all obvious sources of fakery if they were faking it. My opinion anyway ....
Beyond the shadow of the doubt, this circuit is "REAL" I went to YouTube and searched for Free energy - 1w LED lamp - akula 0083 the video there is there very clear. He even put's running circuit in the microwave oven and the device kept running! I am totally agreement with: e2matrix. No further video is required.
Now the question is: Who is "Wattsup" ? Hiding behind this user name? Does anybody knows who this guy is?
Quote from: Grumage on 2014.04.27, 18:42:15
I am presenting a video update and a few questions.
Could you post a closeup screenshot, so I can see the zero reference levels for each channel.
Please trigger on the gate waveform and watch out for that ground clip on the Ch3 (gate channel) when you are sensing across these two CSRs with ground clips of Ch1 & Ch2 connected to the positive terminal of C11.
Two hires scopeshots with a large and small duty cycle would be nice.
P.S.
Also, why is your duty cycle limiter on pin 3 so jumpy? Didn't you use a 470K pot instead of a 470Ω pot?
@Hy all
about the fact that there were hidden button batteries inside the potentiometers ... in the last Akula little device....
....Akula move two potentiometers so remain the other potentiometers where would be two batteries
if two batteries connected in series swich on the LEDs but not connected in parallel ...
who can understand the connections behind can understand if this possibility or not of the hidden battery..
Quote from: 4Tesla on 2014.04.27, 21:35:10
Well here it is!! :) 8)
The GIF is corrupt.. I can't open on my PC. Can anyone open and save as a jpg?
I've converted the picture...in JPG
Woooouuuuu.....
Я благодарю Акула0083
Quote from: TutorialFE on 2014.04.27, 22:42:47
I've converted the picture...in JPG
Woooouuuuu.....
Cool!! I just patched together a version.. I'll add yours to the OP as it is better than mine. O0
Quote from: TutorialFE on 2014.04.27, 22:28:37
@Hy all
who can understand the connections behind can understand if this possibility or not of the hidden battery..
And connections on top. ;)
Quote from: Mehmet on 2014.04.27, 22:18:15
Beyond the shadow of the doubt, this circuit is "REAL" I went to YouTube and searched for Free energy - 1w LED lamp - akula 0083 the video there is there very clear. He even put's running circuit in the microwave oven and the device kept running! I am totally agreement with: e2matrix. No further video is required.
Agreed, I doubt that further video's from Akula would be useful. We already have more than enough videos from him to confuse and baffle!
Hoppy
Good day All:
This just in from MenofFather @ O.U.com,
Quote from: MenofFather on 2014.04.28, 06:56:25
In Akula0083 30 Watt Self Running Generator is pach of pulses. Duty cycle is 50 precents of this pach/box. Someething like in Stanly Mayer divice. So if you not use back pach from R5 and C5 you not get selfrunning!!! And R11 must be ajusted in that possition, that be pach/boxes of pulses, or is or not is.
And how Akula say, this divice work on feroresonance.
So if you not using back pach, then you must made generator, who sends boxes of pulses. And valuses of C3 and C11 must be like in schematic, not that that more is better, more not better, can not be one, let say 2200 uF other 5000 uF, must be like in schematic.
Pulses have frenquency about 400-460 kiloherc and its duty cycle is about 10-40 precents.
Wery important wery fast close mosfet.
pach / boxes of pulses = gated pulse train?
Wery important wery fast close mosfet once again fast switching & fast OFF time (falling edge)....
OK, for those not familiar with the Stanley Meyers gated pulse wave form: = modulated carrier wave....
Modulated by frequency of
Ferroresonance or Dimensional resonance?
take care, peace
lost_bro
Here that I mean boxes of pulses.
Quote from: verpies on 2014.04.27, 22:21:02
Could you post a closeup screenshot, so I can see the zero reference levels for each channel.
Please trigger on the gate waveform and watch out for that ground clip on the Ch3 (gate channel) when you are sensing across these two CSRs with ground clips of Ch1 & Ch2 connected to the positive terminal of C11.
Two hires scopeshots with a large and small duty cycle would be nice.
P.S.
Also, why is your duty cycle limiter on pin 3 so jumpy? Didn't you use a 470K pot instead of a 470Ω pot?
Dear Verpies.
As requested.
Cheers Grum.
Sorry I just re read your post. !! :-[
Cheers Grum.
Addendum.
Ch2 and Ch3 probes set at X10.
Gerbiami MenofFather.
Sveiki atvykę į O.U.R. forumas. Jaučiu jums bus didelis privalumas komandai. O0
Džiaugsmo šūksniai Grum.
Note that the schematic fanar3 cxema.jpg, the upper TL494 has both emitter outputs, pins 9 and 10 tied together driving the FET. This would indicate the FET is mostly on except for the dead time. Could this be an error?
In the lower TL 494 part of the schematic only one of the emitters, pin 10 drives the FET.
In both cases the dead time is being varied.
Edit: Modified based on GL's input.
Quote from: ION on 2014.04.28, 18:58:41
Note that the schematic fanar3 cxema.jpg, the upper Tl494 has both emitter outputs pins 9 and 10 tied together driving the FET. This would indicate the FET is mostly on minus the duty cycle. Could this be an error?
In the lower TL 494 part of the schematic only one of the emitters pin 10 drives the FET.
ION,
The TL494 have two modes of operation based on where you tie the the CTRL pin. If you tie the CTRL pin to
VREF then one internal transistors will be on and the other off, and opposite. If you tie the CTRL pin to ground, then
the two transistors switch on/off at the same time.
GL.
Quote from: Groundloop on 2014.04.28, 19:06:59
ION,
The TL494 have two modes of operation based on where you tie the the CTRL pin. If you tie the CTRL pin to
VREF then one internal transistors will be on and the other off, and opposite. If you tie the CTRL pin to ground, then
the two transistors switch on/off at the same time.
GL.
Ok just read that in the data sheet, pin 13 output control vs pin 4, dead time control. Thanks GL
Note that VD3 and VD6 "catch diodes", if garden variety silicon or schottky diodes, will circulate nearly all of the flyback energy in the 7 and 13 turn coils upon opening of the FET's. This is interesting since this should quench or absorb all energy that would normally be flowing into the secondary 19 turn as part of the flyback pulses in these windings.
If the oscillators are exciting to ferro resonance, then that could account for energy appearing on the 19 turn, not flyback energy.
If VD3 and VD6 are Zener diodes or transient absorbing diodes, then flyback energy will appear on the 19 turn, amount depending on their voltage clamp rating.
How need wound windings? Akula first wound like in picture, one windings on one side, oher on other for him he get selfrunning, then rewound like in standart transformer, first one windings, then other and he not see any big diference, also get selfrunning.
Cuper tape is not shorted one turn after all windings. Akula on load put many LEDs. He say, something like that, that we can try use as load not LEDs, but with it, with incandesel lamps it can not good work, because diodes fast on and off, incandesel bulb not do that. And first he use wery small load and if with samall load he get selfrunning, then he make, biger, biger load. O0
Dear All.
Well I did say I thought there was something wrong !! There was. Both 2W 1 ohm resistors R7 and R 19 (GL PCB) were open circuit !!
So please find attached scope shot of voltages across R19 and R17. Gnd ref C 11 +. Ch 1 and Ch 2 probes are at X1.
Cheers Grum.
Quote from: Grumage on 2014.04.29, 18:05:28
Dear All.
Well I did say I thought there was something wrong !! There was. Both 2W 1 ohm resistors R7 and R 19 (GL PCB) were open circuit !!
So please find attached scope shot of voltages across R19 and R17. Gnd ref C 11 +. Ch 1 and Ch 2 probes are at X1.
Cheers Grum.
Ok thanks for info....
@All
Just received my ecores and I am very disappointed at the size of these little critters. The ETD 29 fits the board but I am so afraid it is way to small to push 30 watts or more of transfer to recharge the system and supply a load. How did we wind up with this core size? The former only has 7 terminals per side wheres the AK30 former has 10. That is a big difference in size. Dammit, I also received some slightly larger ETD 34's but they did not ship the formers. But I am sure those are too small as well. If the core is not the right size, then pumping all this juice into them is useless because you will never get it back.
I think I am going to order real ecores that have a 10 terminal former otherwise we cannot compare this to Akula's device.
wattsup
Quote from: NerzhDishual on 2014.04.21, 14:12:26
Hi Itsu,
Has you have 2 coils with one with a metallic foil, may I suggest you to try this little kind of experiment
with you own 'gears and stuffs' ? Perhaps something new?
(http://freenrg.info/4OUR_F/My-exp2.jpg)
Some results:
Input: square wave. About 15 volts PK-PK. 50% DTC. 3W/220Volts Leds Bulb.
Bulb not connected (no load): 90 volts PK-PK.
Across the 1 Ohm resistor : kinda sine wave. 240 mVolt PK-PK.
Bulb connected (with load): 80 volts PK-PK.
Across the 1 Ohm resistor : kinda similar sine wave. 250 mVolt PK-PK.
Needs a very precise tuning. Freq from about 125 KHZ without additional cap to about 30KHZ.
Does not work with cap > 20 nanoF.
Needs at least a double trace oscillo to make more precise measurements.
In this configuration, the circuit must be grounded (via the Grid).
I have also tested some other configurations with success.
Success = to get the Leds lighted (but not fully, indeed).
This CCT is the same as the one pictured in "principle_of_work_of_Kapanadze.pdf"
See the attached file.
One picture from this document:
(http://freenrg.info/4OUR_F/Kapan-Pic.jpg)
Cheers,
Jean
Hi Jean,
sorry for the late reply here, allthough we already had some PM's exchanged.
I can see the resemblance with what we are doing with the Akula 30 circuit being the coil with the foil (capcoil)
I still am looking for this specific (old) 220V/3W led bulb, so thats one of the reasons i did not start.
What i did do is try to setup the experiment you mention on your website (see video and picture), so perhaps you can mention this website in a reply in case other wants to join in.
I used an old 30mH coil which i had and wrapped a copper tape around it with a 2mm gap.
Then i setup my FG and scope the way you describe and scanned for resonance.
I found something around 810KHz in which my 20Vpp sine wave put in on the copper tape, outputs some 40V pp on the scope (unloaded).
The thing i have problems with is the fact that i can not monitor the input (FG 20Vpp) because the FG ground is kind of floating and connecting the scopes ground lead will destroy this floating.
Please confirm that i hooked up the FG and scope the way you describe on the website.
Also confirm if the 30mH coil is ok, or if i need the 90mH mentioned.
Video here: https://www.youtube.com/watch?v=YTSW9wsDE_8&feature=youtu.be
Thanks, regards Itsu
Quote from: wattsup on 2014.04.29, 23:58:21
@All
Just received my ecores and I am very disappointed at the size of these little critters. The ETD 29 fits the board but I am so afraid it is way to small to push 30 watts or more of transfer to recharge the system and supply a load. How did we wind up with this core size? The former only has 7 terminals per side wheres the AK30 former has 10. That is a big difference in size. Dammit, I also received some slightly larger ETD 34's but they did not ship the formers. But I am sure those are too small as well. If the core is not the right size, then pumping all this juice into them is useless because you will never get it back.
I think I am going to order real ecores that have a 10 terminal former otherwise we cannot compare this to Akula's device.
wattsup
Size of the core is usually inversely proportional to the operating frequency. You can push 30 watts through a very tiny core if the operating frequency is high enough. This is the trend in modern switchmode converters. Whereas early designs used 20 kHz and fairly large cores, modern designs approaching a megahertz use very tiny cores, all other factors being equal e.g. amount of power thruput.
The problem with HF operation is core loss and switching loss goes up unless one is an experienced designer and pays careful attention to core material type and reducing switching loss to a minimum.
One can actually do away with the core altogether and use just air or vacuum for coupling in the RF range, as is the practice in large and small radio transmitters.
For FE research, I prefer operating my switchers at lower frequencies, between 20 kHz and 100 kHz where switching loss and core loss are easily minimized, such that possible OU operation in a looped back converter does not have to overcome these sometimes formidable losses.
Naturally when replicating a design it is best to stick with the size core specified by the inventor.
ION, how did you become so practical? I like that in an engineer. We could use more of them. :)
ION
Do you have practical way to determine the max power the choosen core can pass depending on frequency ? Something which could help with unknown core ?
After spending some time with Grumage I did find out stuff about TL494 which is not working fully as per supposed function. The error amplifier inputs stay always on DC level which is not good at least for Grumage.
For easy fix:
The pot R11 which is supposed to adjust generator ON and OFF modes has to be fed back from feedback on C5/R5. So you can arrange switch off level as soon as LEDs get lit and wait until C3 gets discharged then generator will kick in for another cycle.
Note: if there will be not enough voltage to make generator shut down when LEDs are on you might get voltage reference right from capacitor C3.
Please see modified circuit diagram attached:
Quote from: T-1000 on 2014.04.30, 21:10:00
After spending some time with Grumage I did find out stuff about TL494 which is not working fully as per supposed function. The error amplifier inputs stay always on DC level which is not good at least for Grumage.
For easy fix:
The pot R11 which is supposed to adjust generator ON and OFF modes has to be fed back from feedback on C5/R5. So you can arrange switch off level as soon as LEDs get lit and wait until C3 gets discharged then generator will kick in for another cycle.
Note: if there will be not enough voltage to make generator shut down when LEDs are on you might get voltage reference right from capacitor C3.
Please see modified circuit diagram attached:
Thanks for that info!!!Yes, it makes perfect sense in as much that the Vc of the Circuit is a constant voltage and that was used as the pos. reference for the old position of the voltage divider.... now it's pos. reference is the cycle by cycle variable charge on C5......
So, do you think this error was an oversight on Akula's part ??? or something intentional to throw a wrench in the works......
take care, peace
lost_bro
Hi Itsu,
Thanks+++ to having taken your time to make this experiment and this vid.
I have to apologize. My circuit was not accurate enough.
A more accurate version is:
(http://freenrg.info/4OUR_F/Exp-proposed_m.jpg)
I gave some precisions in: New_index2.html
I have figured out that adding this f' "Optional Cap" is kinda cheating!
The right circuit to be tested is:
(http://freenrg.info/4OUR_F/Exp-proposed-Mod3.JPG)
You can put your scope probes where you want! It could merely change the resonant freq.
I have noticed only KHZ ranges (30-120) and 20 Volts PK-PK input versus about, saying, 120/150 volts
output, IE, enough to lighting this bulb. This, with, also, only few mH (single) coil.
Now, if you PM me your snail mail address, I would be very pleased to dispatch you one of my
(old) not 'unpackaged' 1 watts/ 220vols (round shaped) leds bulb!
OK! I now guess that I will have to make a vid! My first (NRG) one. In English, preferably. No?
Cheers,
Jean
Hi Itsu,
Sorry, I did not fully read your post! Shame on me. :-[
Quote from: Itsu on 2014.04.30, 10:34:07
.............
so perhaps you can mention this website in a reply in case other wants to join in.
..............
Of course I can: http://freenrg.info/4I/New_index-3.html (http://freenrg.info/4I/New_index-3.html)
I have, right now, figured out that the best way to be understood(?) - at least listened (to(?))?
but also copiously mocked is to design a vid, preferably commented in English.
Why not?
En Français, cependant: Meilleurs Sentiments,
Jean
Quote from: Matt Watts on 2014.04.30, 14:33:36
ION, how did you become so practical? I like that in an engineer. We could use more of them. :)
Too many (50+) years on the bench. ;)
forest asked:
QuoteDo you have practical way to determine the max power the choosen core can pass depending on frequency ? Something which could help with unknown core ?
You have to determine how much power you are willing to waste in heating the core, then the answer is easy.
To clarify, it's all about your energy budget and your starting point e.g if you are told to design a power supply that must be 90% efficient, then how you spend your 10% must be practically divided among core loss, winding loss, switching loss, rectifier loss, etc.
You must also take into account many other factors too numerous to mention e.g. expected operating temperature range of the complete unit, size, weight, cooling method and so on.
But for a practical answer, if you can keep your finger on the core, you are probably not too close to the limit.
ION
Thank you.
What I asked was the common situation for me. I get ferrite cores from scrap yard.In case of EI cores dismantled from computer PSU or DC-DC converters I can find TDK pdf and by measuing dimentions I can find what type of core it is and then by carefully underestimate magnetic flux density for saturation this core become usable and I have build successfully some converters using it. The magic is described wonderfully in this article : http://tahmidmc.blogspot.com/2012/12/ferrite-transformer-turns-calculation.html
However , in case of flyback ferrite or yoke ferrite I can only find some rough documents about material types and flux density, but how to compute effective cross-sectional area ? In other words how to know what current I can pass through primary to avoid core saturation .... and so on.... We need a practical methods to use such cores without burning a lot of mosfets. I'm sure it is perfectly possible , yet tricky maybe!
Quote from: NerzhDishual on 2014.04.30, 23:23:05
Hi Itsu,
Sorry, I did not fully read your post! Shame on me. :-[
Of course I can: http://freenrg.info/4I/New_index-3.html (http://freenrg.info/4I/New_index-3.html)
I have, right now, figured out that the best way to be understood(?) - at least listened (to(?))?
but also copiously mocked is to design a vid, preferably commented in English.
Why not?
En Français, cependant: Meilleurs Sentiments,
Jean
Hi Jean, ok, i got the picture, its basically how i was using the circuit.
I got me 2 led bulbs see picture (6.5W and 3W), but only the 3W bulb keeps the resonating frequency of about 40Vpp at 810KHz.
When connecting the bigger 6.5W, the resonance drops.
Neither led will turn on :(
I will send you a PM with my snailmail address
Yes, a video will be great, just show me the circuit with its connections to the FG, scope etc, you then can make some written comments here.
Thanks, regards Itsu
forest said:
QuoteHowever , in case of flyback ferrite or yoke ferrite I can only find some rough documents about material types and flux density, but how to compute effective cross-sectional area ? In other words how to know what current I can pass through primary to avoid core saturation .... and so on.... We need a practical methods to use such cores without burning a lot of mosfets. I'm sure it is perfectly possible , yet tricky maybe!
You will never burn up your mosfets if you use a few tricks of the trade, like always observe current through the inductor using a small 0.1 ohm current shunt resistor. For a given core and inductance, if there is a sudden upturn at the end of the ramp, you are entering saturation and will need to either decrease the duty cycle or increase the frequency. The importance of a gap in a flyback converter is important in flushing out remnant magnetism, which is also a cause of saturation. It is always a balance between required inductance, gap size and # of turns.
Best also to current limit your power supply while testing until you gain confidence with the design. I also sometimes add a small incandescent lamp in series with the power supply and avoid large onboard caps which can take out a fet from instantaneous current during a hiccup.
@ION
Thanks for your reply but let me articulate my point with this photo below.
Here you see the actual size of ETD-29 that fits both circuit boards compared to an ETD-34, which is still very small and also fits the circuit boards. Consider that the ETD-29 ecore is being asked to dish out a good 30 something watts plus looping energy and compare this to the ccore used in the v3 device that's only putting out a few watts. There is no logic possible to explain any more clearly.
@All
I have ordered some ETD-54 - Type 87 non-gapped to do my tests. If you are using the ETD-29, maybe consider changing your load to not more then 2-3 watts to start and then there may be some small changes required in the circuit to lower the overall throughput. Maybe put just enough load so when the circuit is energized (large caps full) and when the power is removed, the leds would take at least 2-3 seconds to turn off. That would be one way to choose maximum load with the ETD-29.
Come to think of it, this one core size comparison, that I did not realize earlier until I had an ETD29 in my hands, now confirms to me that the x-named41 video was faked 2000%. That little ecore that he held between his fingers, like I have in my hand right now, could never do the job he is purporting. Wrong core size. So he just spiked his circuit board from under the panel and faked it, like I had already shown.
It takes time to figure things out. On the way, we discover new aspects that then build on others. Call it a process. If you are running the AK30 board, you will require more core if yo want to produce the design output.
wattsup
Maybe I can clarify by using an analogy.
The size of the inductor and it's inductance represents the size of the shovel moving coal from the coal bin to the steam boiler / engine. For a certain horsepower output, you need to shovel coal at a certain rate (frequency).
If you use a very large shovel, (large inductance) you can go at a slow rate (low frequency) to satisfy required coal input.
If you are using something the size of a spoon (tiny inductance), you need to shovel very quickly (high frequency) to satisfy the input coal requirement.
The coal represents the magnetic energy being transferred through the inductor on each cycle.
Each bite of coal is the amount of charge put into the inductor during the current ramp up cycle.
Dumping it into the steam boiler represents the discharge (flyback) portion of the inductor cycle.
The work performed by the operator and spillage in moving the coal represents the switching loss, core loss, winding loss.
You can imagine that trying to move the coal very rapidly will result in increased spillage (above mentioned losses)
Inductance is increased with the use of a core, but a core is not necessarily required, it just makes for a more compact design by concentrating flux into a preferred path.
The beauty of using a core is that it reduces the amount of windings required for a certain inductance, thereby reducing resistive winding losses.
In my opinion, if you are trying for an OU device, relatively low frequency, large cores and minimal winding ohmic losses would make sense to increase the efficiency of conversion.
If there is a "magic frequency" and expected power thruput, this should be the starting point for the inductor design.
Quote from: Grumage on 2014.04.28, 17:53:13
Dear Verpies.
As requested.
I'm back.
Is sth wrong with your components?
Where did you connect the scope probes and their groundclips?
Quote from: verpies on 2014.05.01, 14:14:03
I'm back.
Is sth wrong with your components?
Where did you connect the scope probes and their groundclips?
Dear Verpies.
Yes drastically wrong !! I had lost both 1 ohm 2 W resistors !!
I posted an updated reply here. http://www.overunityresearch.com/index.php?topic=2358.msg38168#msg38168
Cheers Grum.
Quote from: Grumage on 2014.05.01, 16:22:18
Yes drastically wrong !! I had lost both 1 ohm 2 W resistors !!
Oh! that explains it.
So if the scopeshot below shows currents flowing in L1 and L2 then it appears that you have the connection to L2 reversed because the absolute values (http://en.wikipedia.org/wiki/Absolute_value) of both currents increase at the same time (between the red and yellow line) like in a normal transformer.
So please reverse the connections of L2, set Ch2 to DC-coupling and invert the display of Ch1 on your scope so you don't have negative going peaks ...and show scopeshots of what's happening at the same duty cycle as well as a higher duty cycle.
Dear Verpies.
How are these ??
Please let me know if I have the connections to the transformer correct ?
Cheers Grum.
Quote from: Grumage on 2014.05.01, 18:48:22
Dear Verpies.
How are these ??
Please let me know if I have the connections to the transformer correct ?
Cheers Grum.
You must get something like boxes of pulses going from TL494, I think.
Quote from: Grumage on 2014.05.01, 18:48:22
How are these ??
Still indicative of a normal transformer mode.
Did you reverse the connection of
only L2 ?
Are your diodes polarized correctly and do they work?
P.S.
You can use AC coupling on gate voltage signal (Ch3) because it is riding on a large DC component.
Quote from: verpies on 2014.05.02, 10:18:37
Still indicative of a normal transformer mode.
Did you reverse the connection of only L2 ?
Are your diodes polarized correctly and do they work?
P.S. You can use AC coupling on gate voltage signal (Ch3) because it is riding on a large DC component.
@verpies and @grumage
I was going to ask the same thing because both waveforms flipped by only reversing the L2. This is the crazy stuff I was afraid of because don't forget that @grumage is using the T1000 wind method. So his L1 is two layers away from the core (weaker impress), and the L2 is creating potential cancellation hence very lazy peaks plus if the core is an ETD-29, it is very small to start with.
To confirm this, I would prefer to see 2 screen shots without using any ground reference, so only the probe at....
Ch1 - L1 - Feed Side
Ch2 - L1 - Pulsed Side
and
Ch1 - L2 - Return Side
Ch2 - L2 - Diode Side
You will notice in some of Akulas videos he is using his probe only. I only use my probe as well. I never use the ground unless I want to see a differential output, but in all other points you are better off working without a reference. Reference becomes universal so it is applicable to all points.
@all
Ordered yesterday bigger cores from Digi-Key and they arrived today. Got extra formers for all of these to simplify different wind tests without having to unwind them afterwards. I finally received .22 ohm resistors (had no luck finding them here) so should be ready to set-up some tests.
Photo below shows all four ecore sizes relative to a standard flyback core. I will start with the ETD-54. I will save the bigger one for when I know a little more about this system.
I will not use the circuits. Way too much hassle and it will not answer my questions but just create more distraction for me.
I am only interested in what's going on from the mosfet and from the battery feed through the coils and looping and load. If I can remove the battery and keep the looping in run mode with my FG controlling the mosfet base to better find the sweet spots, that is what I want to know for now. Once that is accomplished, then I will have found the true pulse parameters I need to maintain with a very simple pulse circuit. This would also tell me the feedback that is available for looping back to the mosfet gate.
wattsup
Dear all.
Your thoughts would be very welcome ?? !!
https://www.youtube.com/watch?v=7qUXZB8tom0
Dear Wattsup.
I will endeavour to try your request later. Unfortunately the weekly shop has to come first !! :)
Cheers Grum.
Hi Itsu,
I have sent you a PM.
I have noticed that these big round shaped LEDS and the small square shaped ones
did not work with Joule Thief or Super Joule Ringer (SJR) circuits.
On the contray, with a personal SJR I can light small round shaped 220V leds with
very few current (less than 10 mA) and a mere 1.5 volt AA or AAA bat.
Not in full brightness with only about 10 mA but enough to read a book during the night...
Regards,
Jean
Quote from: wattsup on 2014.05.02, 12:31:47
I was going to ask the same thing because both waveforms flipped by only reversing the L2.
L2's current flipped because its connections were reversed.
L1's current flipped because Grum inverted the display of Ch1 on his scope (right Grum?)
Absv of both of these currents are still increasing
at the same time which is indicative of a normal transformer mode. :(
See the scopeshots below for comparison ( Ch1: Gate voltage, Ch2: L1's current, Ch4: L2's current. The display of Ch2 and Ch4 is inverted )
Quote from: wattsup on 2014.05.02, 12:31:47
...his L1 is two layers away from the core (weaker impress), and the L2 is creating potential cancellation hence very lazy peaks plus if the core is an ETD-29, it is very small to start with.
Why would a winding method affect the L1 & L2 currents like this ?
Quote from: wattsup on 2014.05.02, 12:31:47
To confirm this, I would prefer to see 2 screen shots without using any ground reference,
That will not show us currents flowing in L1 and L2.
Quote from: Grumage on 2014.05.02, 13:24:04
Dear all.
Your thoughts would be very welcome ?? !!
https://www.youtube.com/watch?v=7qUXZB8tom0
Dear Wattsup.
I will endeavour to try your request later. Unfortunately the weekly shop has to come first !! :)
Cheers Grum.
Hi grum,
that noisy signal from the transformer sounds familiar to me as i also did had it for a while.
It turned out i had a dry solder joint at the 1 Ohm resistor (R3) coming from L2.
It was unsoldered, but still making irregular contact producing this noise.
Regards Itsu
Quote from: verpies on 2014.05.02, 14:09:10
L2's current flipped because its connections were reversed.
L1's current flipped because Grum inverted the display of Ch1 on his scope (right Grum?)
Absv of both of these currents are still increasing at the same time which is indicative of a normal transformer mode. :(
See the scopeshots below for comparison ( Ch1: Gate voltage, Ch2: L1's current, Ch4: L2's current. The display of Ch2 and Ch4 is inverted )
Why would a winding method affect the L1 & L2 currents like this ?
That will not show us currents flowing in L1 and L2.
Dear Itsu.
No. My scope tells me with a little arrow if I have deliberately inverted the signal. And it would appear that I can only do it on Ch 2 !!
I have also looked for dry joints, non found so far !!
Cheers Grum.
Knowing it's a fake,i have to know where the power supply is comeing from for such a heavy load.The only possibility can be the transformer core itself?. Those automotive head light bulbs draw far more than any battery could provide that may be inside the cap's them self.So the only place left that seems large enough ,is the transformer core itself.
I know this thread is about another device,but it seems like the place to pose the question-how has he done this?.
https://www.youtube.com/watch?v=OBGJnG-00Gc
@all
Should be running tests soon. Here is what I will try in my first test using the ETD-54 and Magwire coils as per the latest winding findings for the original Akula 30 ecore.
This will be my base line winding method that I can compare to other winds types.
The circuit will be the original AK30 schematic at mosfet and feed cap to loop and to load.
wattsup
Dear All.
I have just put this statement in Russian through Goggle translate.
Quote.
Question : Regarding the 30 watt flashlight , there you reguliruesh low frequency oscillator frequency ?
Akula say: No, guys , there rf pulses packs RF pulses , they ... well, how do you explain that? should be included , the fact is that when you have a transformer when the low-frequency part of the amplitude , you already there will be generated automatically burst , that is because there is a current feedback , ie , when it is fed a signal LEDs perenakalilis TL generator , the generator is off TL , respectively overvoltage occurs , there is a binary reduction of porosity , ie, switching off completely. Especially, if you examine the datasheet , I got off on the scheme itself smooth control duty cycle generator , it either is or is not present, so there is formed with a frequency burst ferorezonansa . And he is automatically generated , since we , for example, capacitor is charged to the required voltage that is held fixed burst , TL generator , he turned off at this time of the transistor until the capacitor is not charged . This cycle is formed by the low-frequency amplitude common transformer, which has at this moment is in resonance .
Arunas : Well, actually about that on our forum oveyuuniti long tormented , trying to understand how this feedback works .
Shark : Well there you see what nonsense it turns out, the TL -ku posing outside smooth control of porosity , huh? feedback , and a sharp cut-off , or is, or there is no signal and turns , T -ka then charged , the capacitor sees recharged , it shuts down and it will happen in cycles. Since the torus transformer already formed the amplitude and the cycle starts it automatically link , ie , here TL -ka , it in fact it is not , such as a generator. It already does everything that happens in the transformer. That means that the transformer starts to tear cycle the bursts and off for TL -ke , thereby supplying current to the LEDs , cutting off feedback , huh? ie komaratoru , uh already dozaryazhaya second capacitor . That is giving a signal to the shutdown. That is, there TL Single dependent, as such, it does not generate anything , it is the duty cycle zadaetsya it - equal feromagnitnomu resonance and the pulse repetition rate , and their cyclical already formed by the transformer .
Arunas there about the following two points . you have one ( nerazslyshanoe word) as the frequency, and the second is the principle of where you find it for feedback as I understand it , yes ?
Shark : Yeah, I mean look at the current, when we have more current LEDs begin to miss , we go off on a team full TL -ki . The same goes reinsurance voltage .
Arunas : Well, I never understood , thank you.
Shark : Now look : I have a third leg is involved . Yes ? there are over 22 Nanofarad is what has been done? Kapratar and fifteenth comparator directly strong (maybe not such a word ) is the voltage divider with the first and second key , so it does what? This blocks the duty cycle , there no smooth adjustment of porosity , she's either there or ... So we there TL -ka, must be understood. it works in batches .
...
Question : Roman , say , the capacitor C3 cumulative 2000 microfarads , you can put it more volume ?
Shark : In principle, it 's a capacitor is formed and that the main Periodicity enable, disable, TL -ki . Well, while the secondary is current , right? ie . this stress on the LEDs. That is changing this capacitor you change the cycle on and off TL -ki , as it will look on your charge control capacitor. ... When you turn on a system , you will see output tlki that ? there are tutus , here , specifically the duration you can rebuild their frequency ferorezonansa . Following them you'll be looking at the power , yes , the cost to the output , ie the number of signals in a pack . And you can see the cyclical , that actually , it is cyclic , it is almost 50 to 50 poluchaetsya running off, running off, it is precisely the control capacitor working . TL -ka sees capacitor re-charge , she turned off , no signals . Came the charge , it is included here , and it starts to go in cycles . And this cycle will automatically be matched with low frequency transformer already formed and they begin to auto control all this turns out. That is, it must be done to understand.
Unquote.
The Russian text was kindly posted by MenofFather at OU.com.
I think this might have a bearing on our replication efforts??
Cheers Grum.
Grum, this is somewhat helpful and would be more so if we knew what these terms were:
TL or TL generator
TL -ku
TL -ka
TL -ke
TL -ki
Can anyone assign electrical engineering descriptions to these?
Aaaaa, replace "porosity" with "duty cycle" everywhere in this translation
TL, TL -ku, TL -ka, TL -ke, TL -ki
...are all declensions (http://en.wikipedia.org/wiki/Declension) of the TL494 Integrated Circuit.
I think what he is basically saying is that the ripple on the large storage capacitor causes an on/off operation of one of the oscillators such that it puts out bursts of high frequency pulses.
The on off action occurs around the preset value of the voltage sense potentiometer. The burst settles at about 50% duty cycle i.e. burst on to burst off is around 50% but this may vary depending on loading.
Since he talks about packs of RF pulses, this seems to imply that the burst oscillator is adjusted to a high frequency, low duty cycle to produce the "picket fence" bursts he talked about elsewhere.
Quote from: Grumage on 2014.05.04, 13:17:53
I think this might have a bearing on our replication efforts??
It does because according to it the feedback loop is wired incorrectly in all of the schematic diagrams released to date, e.g. re. current limit in the LED bank.
A native Russian speaker should clean up this translation. I will not do it on translate.google.com because I'm not native and that translator stopped highlighting words in the source text, a while ago (no on mouseover events anymore >:( ).
Is there another translator out there that highlights words?
@Grumage
Well the only thing I have found was last night, late, I saw this effect but did not pay enough attention to it then made other small changes, so I will retry again this afternoon.
I was pulsing and was anywhere between 850KHz and 1.5MHZ. The LEDS where mildly lit but I could notice a secondary fluctuation in the brightness I'd say about 2-7 Hz. As if there was a cycling ramp up, ramp down, ramp up, etc.
Someone at OU said that Akula confirmed that the AK30 schematic was incomplete. Does anyone know more if this is a valid comment or hearsay. Someone should ask Akula why he does not show the AK30 again man. And if the schematic is or is not correct, and if not, where, otherwise we are stuck playing Pin the Tail on the AK30.
But for sure this diagram is incomplete. There is not enough change in the primary to impart enough change in the secondary. Everything is being stressed by the positive "freezing" of the transformer. There is positive hitting the coils from all sides and all there is available to make any change in this positive loop, is this mosfet stuck in the middle creating this minute short that is not doing anything. I am sure the schematic needs to include a mosfet between R1/C2 and L1 so that the primary is disengaged from the feed positive from both sides of the L1 coil. Only this will create enough change to impart anything to the secondary. But I will first try with a diode pointing to C11 after R3 with or without the D6 diode. I know @verpies says it won't make any difference, but it will inhibit the positive from entering the L2 and bleeding to the load via R2.
What I am noticing is that the simple fact the mosfet is pulsing is creating its own feed by leaking to the load. We have all seen this before where the pulse generator itself is the feed source via leaking. @TK referred to this as well at OU. One way to show this is if you pulse the circuit and light your leds and while they are lit disconnect C11 from the ground side. If the leds keep on lighting, it is the feed coming from the frequency drive circuit to the mosfet gate and leaking to the load. Maybe this is only typical with my setup because I am directly using my FG, but regardless FG or pulse circuit should impart the same leaking to load. So in reality, the output to load is never coming from L2 or even the pulse of the L1 to L2 does not influence the load.
If I have anything like that first LED fluctuating effect, this time I will video it, but as it stands so far, the is nothing with this thing as it is shown on the schematic.
I am pulsing the mosfet gate directly with my FG positive on the gate and negative to the circuit ground. Best duty was from 9-21%. Running off a 12vdc 10a battery that is charged via a power supply.
@Grumage and @Itsu, I realize that with your circuit boards, it is hard to try some variations, but with my open set-up, this is very easy to change things around, add diodes or pots here and there to eventually find the right mix.
wattsup
Anyone that has experimented with LED's using a function generator to power them, will know that they they can be illuminated at very low current levels when driven with square waveforms, or any waveform with sharp leading edges, when driven at high frequency in the Mhz range. Lasersaber has shown how a simple, albeit unconventional circuit can achieve this with maintained illumination when working from a low capacity charged cap with the initiating power supply removed. Akula's devices are operating on a similar principle to Lasersaber's IMO. The duration of illumination is dependent upon capacity, taking into the effect of leakage, which comes into play, particularly when using electrolytic caps. IMO, Akula will never be able to demonstrate level illumination whilst self-running over extended periods of time considerably longer than he has shown in his videos to date, using the capacitors shown. Button cell batteries or super caps are clearly not essential to at least get the appearance of self-running, having seen Lasersaber's video demos.
Hoppy
Dear All.
I have just come back from an exhaustive 2 hour session with T-1000 looking at the resonance of various Ferrites and even Transformer Steel cores. Based upon the PDF attached.
http://realstrannik.ru/media/kunena/attachments/6159/opredelenie_MR.pdf
I have dllabarre to thank for providing the "Google" translation.
Quote.
When conducted experiments on nuclear magnetic resonance, the resonance
frequency ferrite sought so (more options I have not found).
Pickup coil must be arranged in other geometrical axes.
Moreover, the number of revolutions does not affect resonance.
When a short pulse (0.5-1us) 12-20 volts...
immediately see the frequency of the ringing...
Appearance generator short pulses
Unquote.
We have concluded that all the Ferrites have a resonance in the 10 to 30 MHz region the lowest resonance was found from a piece of 316 Stainless Steel @ 4.2 MHz . I am hoping to make a short video tomorrow to show others how to find the resonant point of a material.
What is interesting is that Stainless Steel contains Nickel and I am sure somebody posted over on OU.com that there are Nickel based ferrite combinations available. The upshot?? The lower the resonant frequency of the material the easier it will be to manipulate the device.
Cheers Grum.
Jusr watched this one on his 1kW system. It's getting harder to spot as to where the power could be coming from.
https://www.youtube.com/watch?v=aUWAzmy-61o
That black wire looks like it goes into a hole in the pipe. Look closely and you will see it slightly to the right of where the wire enters the pipe.There is another hole about the same size a few inches below. Notice the welded horizontal support to the other pipe also a few inches below. This tells me the pipe with the black wire does not go that deep into the ground and is supported by the horizontal welded bar and the other pipe. It would be easy therefore to snake a wire up into that pipe and out the hole, a wire which is buried in shallow earth and goes back to a power source.
I totally distrust all these videos that need a ground wire, especially if it is pre-arranged ground and you do not see the actual hammering into the ground of the metal pipe or attachment of the wire. We do not see him strip this wire and attach it to the bolt.Look closely at the attached picture.
Of course this is my opinion only.
Quote from: ION on 2014.05.05, 11:12:35
That black wire looks like it goes into a hole in the pipe. Look closely and you will see it slightly to the right of where the wire enters the pipe.There is another hole about the same size a few inches below. Notice the welded horizontal support to the other pipe also a few inches below. This tells me the pipe with the black wire does not go that deep into the ground and is supported by the horizontal welded bar and the other pipe. It would be easy therefore to snake a wire up into that pipe and out the hole, a wire which is buried in shallow earth and goes back to a power source.
I totally distrust all these videos that need a ground wire, especially if it is pre-arranged ground and you do not see the actual hammering into the ground of the metal pipe or attachment of the wire. We do not see him strip this wire and attach it to the bolt.Look closely at the attached picture.
Of course this is my opinion only.
Yes, I agree. Look closely at the so-called cable termination in the clips below that I posted way back as signs of faking on OU.com. As ION says, the cable appears to run into the pipe as the sheath does not appear to be stripped back, revealing at least a sign of the bare inner conductors, as I would be expect. The termination looks too neat. He has failed to show us the removal and re-termination of this cable throughout his videos.
It dose look that way.Watching it again in full screen,it seems the dirt around the pipe is a dark grey color,while the rest seems to be a rusty brown color-freshly dug perhaps?..Problem with all this deception,is when something real comes along,we'll all have the yea yea attitude.It's all starting to sound like the boy who cried wolf.
I decided to go ahead,and order some pot core's and other parts anyway,as i need a break from my other project ATM.
TK does much the same thing in his Aqua2 video IMO. The difference is that he threads a thin wire up the tubular earth braiding and connects this to a 'live' mains supply under the concrete lid outside the house. I had a 2KW fire running on 5 metres of just one of these heat resistant sheathed wires, using the braid as the neutral.
I think the skepticism is because no one still managed to have resonance but I see that progress is not missing from all sides ...... Akula would have to be fixed in order to show how well everything like Kapanadze the wire on the tap..
By the way seems the dreams not stopping...
https://www.youtube.com/watch?v=KLIX6nPAg9k
@all
Sunday afternoon I installed a second mosfet so I am running with two IRF640s. With the second moffet located between R1/C2 and L1.
Since I have a two coil primary the mosfets are on the beginning of each coil L1 and L3 and the other ends are connected together and fed B+ by R1/C2. hahaha
The difference without any fine tuning yet is night and day. Much more power is being moved and when I remove the battery, the LEDS go off very slowly, not like the immediate off they would do before. So I think this is going in the right direction. Now I need to replace a few of the resistors with pots and start playing around with more or less load and see what happens. But the LEDs are still going off but you can see the system wants to keep them alive much longer then what those caps can do on their own.
All I do now is simply hit the positive with the battery B+ for a few seconds then remove it while I hunt the settings.
Never go below 5khz or the mosfets will start cooking.
This Monday morning I changed out one of the 640s because it was an older one and I saw a difference in the pulse strength between the drains of one mosfet and the other. I also reinstalled half an ecore to which I created a center gap by sanding it down slightly. Anyways, I was around 5.6 MHz, applied the B+, LEDs lit up, removed the B+, Leds slowly dimmed until they were very low light but it stayed there. I then removed the FG as well and the LED is still dimly lit. Remove all 3 scope probes and the LED is still dimly lit again since a 5 minutes. Weird stuff indeed.
Will investigate more after work and do a small video.
wattsup
PS: Question about scopes.
I put both scope probes A and B on the same point and at same volt/div and both are at 10x. But the ChA is reading 68 volts while ChB is reading around 6 volts. Changed both probes and put in news ones but still the same thing. Don't know why that is but it is definitely not normal. Any ideas.
Added:
a) I forgot to mention that I also added a diode pointing to R3 after L1. I think this is an important addition because it prevent B+ from entering the top of L2.
b) The photo of the dim leds was taken at about 20 minutes.
@TutorialFE,
Ref your annotated video clip below. Look carefully at the tight bending radius of the cable as it enters the hole and appears to turn in a downwards direction in my clip below yours. Now look into the second hole below the cable entry hole. Do you see what I think I see? Despite the bolt on the other side, there would be enough room for the cable to pass through the pipe.
@all
I would advise to all those doing bench work to check out their scopes. With all the pulsing these scopes take you could develop some difference in waveforms thus not providing the true readouts that you may be looking for.
Use an FG or any other pulse source and put both probes on the same + and - points and check if the scope is showing two identical waveforms under the same scope settings. If you have an FG start sweeping up in frequency and adjust your time/div to make sure both channels are giving the same waveforms.
I have one channel B on my HP 54645D that is not giving the right readings but only as the frequency rises. That would be very hard to notice if you are using your scope as usual. I put 4 images of my waveforms at 1, 5, 50 and 500KHz here;
http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg401398/#msg401398
My older Hitachi V-650F is still OK. The volt/div knobs were starting to act up so I just opened the hood and sprayed in some tuner cleaner and now they are perfectly ok, so at least I have that scope to work with plus Ch A on the HP. What a bummer. I wonder if it would be hard to fix my HP.
wattsup
I finally finished my Akula 30W board and did some first tests.
I built the board with verpies Mosfet driver Mod, i used the MIC4422BM. Gate pulses are clean, frequency range is from 14 to 108kHz. Duty cycle cannot be adjusted with pot, i will look into some of the here proposed mods and try them later on.
With L2 feedback path disconnected, current draw at 12volt is around 3,3 Amps, LEDs are bright.
When i hit the switch, lights go pretty dim and my PSU restricts down to 8,4V, 3,5 Amps (limit set by me).
The 0,22 resistor gets really hot while operation, I'm glad i used a 5W instead of a 2W.
For now I'm using copper wire for both coils, but i plan to use iron wire for the L2 coil as has been suggested. The first i ordered was not flexible enough to wind, hope the new one will be in the mail today.
kind regards and peace,
From other Planet
Dear All.
As discussed in chat earlier, please find attached Les Banki's PDF of the Akula 0083 30W device.
This circuit has been fully tested by the Author and works correctly to conventional electronic engineering.
From this basis it is suggested that we can then move on to looking at the more unconventional aspects. Particularly the feedback from the LED's !!
Cheers Grum.
Quote from: From Other Planet on 2014.05.07, 14:49:14
I built the board with verpies Mosfet driver Mod, i used the MIC4422BM. Gate pulses are clean, frequency range is from 14 to 108kHz. Duty cycle cannot be adjusted with pot,
Did you use this duty cycle limiter on pin #3 ?
Quote from: verpies on 2014.05.07, 22:59:47
Did you use this duty cycle limiter on pin #3 ?
Nope, but thank you for posting it again. Think i will try it if i find a 470k pot in my workshop :)
Quote from: From Other Planet on 2014.05.08, 01:23:14
Nope, but thank you for posting it again. Think i will try it if i find a 470k pot in my workshop :)
But there is not a 470k pot in that schematic. Only 470Ω pot
Quote from: wattsup on 2014.05.03, 16:15:39
@all
Should be running tests soon. Here is what I will try in my first test using the ETD-54 and Magwire coils as per the latest winding findings for the original Akula 30 ecore.
This will be my base line winding method that I can compare to other winds types.
The circuit will be the original AK30 schematic at mosfet and feed cap to loop and to load.
wattsup
Good work! Now put from generator any to mosfet with driver packs of pulses with frenuency 350-460 kiloherc. Were packs frenquency must be individual for any transformer and about 20-40 kiloherc. And packs duty cycle 50 precents. :)
@all
I had prepared a rather long post again about the AK30 but there is no point. After so many days on the bench experimenting with the AK30 circuit as provided by Akula, using two wind methods, gapped, ungapped, this circuit is just a straight line to load with a token pulse feature that does nothing. Resonance shmesonance, there is no point to producing resonance if the feed is being forced straight to load, what is the point? No point, no advantage, just heat.
The AK30 as he shows it in his schematic is a dud. This explains why there is no video ever done showing this device but it does not explain why Akula would not have come forward way before to explain that this schematic should not be taken as a functional guide. It is just a big joke man.
We will always have trouble with Akula designs because he is already sold out so whatever he says, does, explains, shows, purports, confirms, etc., etc., is just his way of trying to steer the world away from his biggest money maker (for him) and that is his yoke device. Once he sold that out, he probably was asked to create some duds to keep the public eye as far away from the yoke system as possible since they know, he had made some videos of the yoke before he licensed it away. That one fact for the money guys would be a valid deal breaker unless Akula could swing people away from the yoke. That is my gut feeling on the business side of things.
I think this is a great learning experience for us all. Instead of hastily moving forward to produce a drive circuit, first we all should ascertain if the main function principle and effect can be replicated in a more controlled manner using our regular bench equipment. Only once such preliminary tests show promise should we then blitz to make a circuit board, etc. My opinion and my future modus operandi. No more replicating a diagram if the base principle cannot be shown to work on the bench. I mean, why else do we have all these great pieces of equipment on our benches if not to cut through the fat of the matter.
Anyways, sorry for this bad news. It's bad for the AK30 but I cannot say anything else about the other devices he has shown except that I know now that all of them are just a calculated distraction (producing only peanut watts) from his main yoke device. Yes, he can say the yoke core lost its ability (bullshit), or any other excuse about the yoke not being of the right material (more bullshit), but that is all bull man. If you take a simple flyback ccore and measure the surface area of the contact points, then compare that surface area to that of a yoke, the yoke will win about 3 to 1. The yoke provides at least double or closer to triple the gap surface area and this must play into how the yoke halves can be used to transfer energy far better then a counteracting ecore.
Then it dawned on me to ask, besides the AK30, which Akula devices, shown to date, use an ecore? hmmmmmmmm. I wonder why?
So what else will be exported from Akulastan? Hmmmmmmm.
@GK
I'm back.
I just added a pdf to the ftp site on Tesla in case you never saw this one. It is so great to read his talks and there are so many great secrets in there when you read between the lines. The pdf was to big to attach here.
So I put it here;
http://purco.qc.ca/ftp/Inventors/Nikola-Tesla/General/
The pdf that starts with Nikola Tesla - On his work................
wattsup
Quote from: wattsup on 2014.05.12, 18:25:14
I think this is a great learning experience for us all. Instead of hastily moving forward to produce a drive circuit, first we all should ascertain if the main function principle and effect can be replicated in a more controlled manner using our regular bench equipment. Only once such preliminary tests show promise should we then blitz to make a circuit board, etc. My opinion and my future modus operandi. No more replicating a diagram if the base principle cannot be shown to work on the bench. I mean, why else do we have all these great pieces of equipment on our benches if not to cut through the fat of the matter.
No need to be sad, I said that in very beginning about understanding principles first just got ignored.. Anyway, the latest Teamspeak3 recording I posted few weeks ago has all explanations you need just find Russian speaking radio ham engineer for translation... ;)
Quote from: wattsup on 2014.05.12, 18:25:14
I think this is a great learning experience for us all. Instead of hastily moving forward to produce a drive circuit, first we all should ascertain if the main function principle and effect can be replicated in a more controlled manner using our regular bench equipment. Only once such preliminary tests show promise should we then blitz to make a circuit board, etc. My opinion and my future modus operandi. No more replicating a diagram if the base principle cannot be shown to work on the bench. I mean, why else do we have all these great pieces of equipment on our benches if not to cut through the fat of the matter.
wattsup
Before hastily moving forward, I suggest that more time is spent analysing Akula's videos, as they invariably give clues as to how he fakes his devices. As soon as I realised that his garden contraption was faked with the power down the so called earth wire, I immediately suspected every device that came after that was faked in some way. Personally, I will not waste time on replicating any more of his LED non self-runners. I think Lasersaber has a genuine long runner with the 'SJR Simple Looper' and Akula possibly discovered this effect but decided to mask it with more skulduggery in order drive bigger LED loads, so the principle of operation would be difficult to work out.
Hoppy
Quote from: T-1000 on 2014.05.12, 22:10:43
No need to be sad, I said that in very beginning about understanding principles first just got ignored.. Anyway, the latest Teamspeak3 recording I posted few weeks ago has all explanations you need just find Russian speaking radio ham engineer for translation... ;)
I don't think that any principles have been intentionally ignored. Clearly nobody to my mind has produced a convincing and well documented self-runner to date, so IMO its more likely that any principles involved in producing a true self-runner remain misunderstood, rather than ignored. Too many people on the fora are speaking from the position of an authority on the subject, as if they have 'done it and got the tea shirt', without producing any convincing evidence.
Hoppy
Quote from: verpies on 2014.05.08, 05:29:36
But there is not a 470k pot in that schematic. Only 470Ω pot
Yep sorry was tipping error. Works really good to override the Duty cycle, in lowest setting my Board is now consuming only around 30mA, but sadly not yet selfrunning.
Closing the feedback Loop still increases Input current....
Next i will try different winding types and wire materials for coils.
In output TL must be packs of pulses, if you not get from TL494 packs of pulses, then this divice guaranty, that for you not works as selfrunner. Here ajusting is wery fine and hard.
Stefan started a new topic
here
http://www.overunity.com/14643/successful-akula-30-watts-free-energy-device-solid-state-replication/msg402937/#new
hartiberlin
Administrator
Hero Member
ne)
Successful Akula 30 Watts Free Energy device Solid State Replication
« on: Today at 03:17:25 PM »
Quote
Some Russians have mastered the "Akula" LED Lamp circuit and now can light 30 Watts LED light lamps for free !
Part 1 http://m.youtube.com/watch?v=1_iutxqAz_M&feature=youtu.be
Part 2 http://m.youtube.com/watch?v=YleTYEVZaxs&feature=youtu.be
Dear Chet and all.
Where has Stefan been hiding ??
I watched those two videos probably 6 to 8 weeks ago !!
Nothing new here !!
Cheers Grum.
Yep, nothing new...but some tips by the man on this video may be useful. He used 9V voltage stabiliser and choke placed in other place.
Personally I just built two small DC-DC converters : one using flyback and one using push-pull topology. I'm not good at it it was my fist attempt. Theoretically flyback should work fine because it was modeled using Texas Instruments online tool, but in fact only push-pull works and it is solely because I used some tips from experienced electronic man and placed proper choke between secondary transformer output and storage/filter cap. The same may be for Akula device.
Hi there,
I managed to fix akula 30W circuit http://www.overunityresearch.com/index.php?topic=2358.msg36611#msg36611 with minor modification in way it is supposed to work:
1) The existing feedback loop needs to be removed and capacitor C5 + resistor R5 bottom right need to be replaced by one more layer of LEDs (I have 7 in series).
2) The connecting resistor R7 from TL494IN to that feedback loop needs to be grounded instead.
3) The connecting resistor R9 to plus side then pot and PIN 1 of TL494 needs to be disconnected from positive power input side and reconnected into positive side of charging capacitor C3.
In this way you will have pot R11 setting maximum level of capacitor C3 charge and regulating when TL494IN generator should stop after reaching voltage level. There is very sensitive level when it starts oscillating with ON/OFF modes on low frequency and that was intended to create secondary low frequency for transformer.
Also I did find out one issue there as well:
As soon as you connect L2 into circuit or just bulb back from positive side of capacitor C3 to positive side of capacitor C11 - the circuit starts to consume amperes instead if few milliamperes of current. This explains why ITSU and others have so much current draw. Without feedback loop LEDs get lit just fine. The feedback loop gets charge up to 20.5V in my case and I will need some another pulser/counter circuit to recharge primary capacitor C11 when the TL494IN is in OFF state perhaps.
P.S> I will repost my fixed version of circuit later when will have chance to access my computer, or Grumage/others who received copy might put it here....
Also I winded transformer with 1mm iron wire from http://www.ebay.co.uk/itm/GREEN-WIRE-FENCING-NETTING-PLANT-IVY-TWIST-TIE-TREE-SUPPORT-TYING-FIXING-GARDEN-/360611320881?ssPageName=ADME:L:OC:GB:3160 on ferrite E-core ETD49-CF139 from Power Magnetics http://powermagnetics.co.uk/pace-components/ferrite-cores/etd-cores/etd49-cf139 on which I made gap with isolation tape on all 3 junctions. The turns on secondary was 35+17 CW and on primary the another half with 17 turns CCW as per my suggestion in beginning of this thread.
@T1000,
Thanks for your circuit modifications.
You have created a grounded output Mr Bidini style energiser. If this is how it is supposed to work, can we now safely assume that Akula got it so badly wrong that he never in fact had a self-runner? Have you got this version self-running?
Hoppy
Quote from: Hoppy on 2014.05.24, 10:42:33
@T1000,
Thanks for your circuit modifications.
You have created a grounded output Mr Bidini style energiser. If this is how it is supposed to work, can we now safely assume that Akula got it so badly wrong that he never in fact had a self-runner? Have you got this version self-running?
Hoppy
Partially it is Be dini style energiser, yes. And partially it is supposed to fetch additional energy from NMR for powering up 30W+ and not just one or 5 LEDs - http://www.youtube.com/watch?v=4n22MNKrusA (the partial circuit is in second image at http://realstrannik.ru/forum/temy-atom/134569-qgenerator-na-nelinejnoj-induktivnostiq-rekonstrukcziya-sxemy-i-pechatnoj-platy.html#171345 )
Also until I will find a way to recharge primary power source capacitor C11 without drawing any additional amperage from power source - no way for self run, please reread the issue description I just posted.
Dear All.
T-1000 revised circuit diagram. O0
Cheers Grum.
Quote from: Grumage on 2014.05.24, 21:11:42
Dear All.
T-1000 revised circuit diagram. O0
Cheers Grum.
There was one more mistake left over and I fixed it as well :)
Also later I will see what I can do with variable duty cycle on PIN 16, probably it will go over resistor+pot + small capacitor and another feedback loop from PIN 10.
Good Day T-1000
One more detail, Pin #10 is an Emitter and in the *new* configuration as the last schematic indicates, is now Sourcing current.
The detail is that R15 connected to the Positive power rail from output of Voltage regulator is also Sourcing Current to the base of VT2.
Probably need to remove R15 from the Schematic also.
take care, peace
lost_bro
Quote from: lost_bro on 2014.05.25, 01:04:22
Good Day T-1000
One more detail, Pin #10 is an Emitter and in the *new* configuration as the last schematic indicates, is now Sourcing current.
The detail is that R15 connected to the Positive power rail from output of Voltage regulator is also Sourcing Current to the base of VT2.
Probably need to remove R15 from the Schematic also.
take care, peace
lost_bro
The R15 is connected to RT PIN 6 and is regulating frequency, so it is correct. The PINs 15/16 are grounded for a moment so the secondary error amplifier is disabled. I am thinking to use it for limiting maximum duty cycle as by default in my soldered circuit it gives 40-50% DC when PIN 1 is on logic "0" and <10% when PIN 1 is getting logic "1" after C3 gets charged with BEMF to minimum adjusted level.
Please see internal circuit of TL494 attached from http://www.ti.com/lit/ds/symlink/tl494.pdf :
Thank you T-1000
Quote from: T-1000 on 2014.05.25, 08:36:58
The R15 is connected to RT PIN 6 and is regulating frequency, so it is correct. The PINs 15/16 are grounded for a moment so the secondary error amplifier is disabled. I am thinking to use it for limiting maximum duty cycle as by default in my soldered circuit it gives 40-50% DC when PIN 1 is on logic "0" and <10% when PIN 1 is getting logic "1" after C3 gets charged with BEMF to minimum adjusted level.
Please see internal circuit of TL494 attached from http://www.ti.com/lit/ds/symlink/tl494.pdf :
My Bad....
can't see without my glasses,,,, should have wrote R16 *not* R15. sorry for the confusion......
It is R16 from the power rail 12 Volt reg. to base of Vt2.
take care, peace
lost_bro
HI,
I just made my final fixed version including maximum duty cycle control.
With this you will be able to control maximum ON state of coils for adjusting minimal power draw with LEDs on optimal brightness :)
The rest of circuit now needs to be researched for banging scalar waves inside of ferrite core on transition from ON to OFF state in coils then trying to see if NMR condition occurs...
@all
I have been working on some other configurations but I need some circuit help. I am putting down how I am using the AK30 with my FG and battery feed in general.
All I do is connect my FG positive to the mosfet gate. I would like to know if I need to add some resistor from the gate to ground and if yes what values should I look at, or, is there a small circuit I should put before the gate so I can get better pulsing effect or control or better on/off action through the mosfet.
Seems to me the pulsing as shown is not causing the battery to be pulsed enough even thought I see the pulse on the scope off the coils, but this may be only the FG pulsing through the mosfet and the complete circuit. So probably what I need is a little pre-circuit that I can connect to only my FG positive, then to the gate and to the circuit ground so I can pulse the AK30 to the frequencies I need or for any other future system where I need to drive a mosfet up to 20MHz with more bang.
Any help on this would be very appreciated.
@T-1000
Good ideas about those changes. Maybe consider adding a diode after R3 and pointing to C11 so you do not bleed any B+ into L2.
wattsup
Quote from: wattsup on 2014.05.26, 21:02:07
@all
I have been working on some other configurations but I need some circuit help. I am putting down how I am using the AK30 with my FG and battery feed in general.
All I do is connect my FG positive to the mosfet gate. I would like to know if I need to add some resistor from the gate to ground and if yes what values should I look at, or, is there a small circuit I should put before the gate so I can get better pulsing effect or control or better on/off action through the mosfet.
Seems to me the pulsing as shown is not causing the battery to be pulsed enough even thought I see the pulse on the scope off the coils, but this may be only the FG pulsing through the mosfet and the complete circuit. So probably what I need is a little pre-circuit that I can connect to only my FG positive, then to the gate and to the circuit ground so I can pulse the AK30 to the frequencies I need or for any other future system where I need to drive a mosfet up to 20MHz with more bang.
Any help on this would be very appreciated.
@T-1000
Good ideas about those changes. Maybe consider adding a diode after R3 and pointing to C11 so you do not bleed any B+ into L2.
wattsup
\
Good day Wattsup
In regard to the above question:
I usually use an opto-Mosfet driver chip (many types available) that can provide the Amperage necessary to charge rapidly the MOSFET gate.......but most importantly the dedicated opto- MOSFET driver IC will isolate and protect the FG from voltage spikes, BEMF, etc. Don't know about finding one that can reach 20Mhz.....
take care, peace
lost_bro
I was told you couldnt get resonance between two coils with the diode between - moved diodes.
Correct! The diodes would shut off the echoing back and forth.
Quote from: T-1000 on 2014.05.26, 15:48:53
HI,
I just made my final fixed version including maximum duty cycle control.
With this you will be able to control maximum ON state of coils for adjusting minimal power draw with LEDs on optimal brightness :)
The rest of circuit now needs to be researched for banging scalar waves inside of ferrite core on transition from ON to OFF state in coils then trying to see if NMR condition occurs...
Good evening T-1000
This is from O.U.com
Quote from: wayne49s on 2014.05.27, 21:11:00
Hi,
I was reading the same topic on overunityresearch and T-1000 is trying to close the OU loop. This message is for him if he has run out of ideas. I've been studying the schematic, and I believe that the bottom of L2 (with the diode) needs to be connected to ground instead of the C3+ line. Otherwise I don't see how it can charge the C11 with the BEMF.
Maybe someone with access to that forum can pass the message.
take care, peace
lost_bro
Quote from: wattsup on 2014.05.26, 21:02:07
@all
I have been working on some other configurations but I need some circuit help. I am putting down how I am using the AK30 with my FG and battery feed in general.
All I do is connect my FG positive to the mosfet gate. I would like to know if I need to add some resistor from the gate to ground and if yes what values should I look at, or, is there a small circuit I should put before the gate so I can get better pulsing effect or control or better on/off action through the mosfet.
Seems to me the pulsing as shown is not causing the battery to be pulsed enough even thought I see the pulse on the scope off the coils, but this may be only the FG pulsing through the mosfet and the complete circuit. So probably what I need is a little pre-circuit that I can connect to only my FG positive, then to the gate and to the circuit ground so I can pulse the AK30 to the frequencies I need or for any other future system where I need to drive a mosfet up to 20MHz with more bang.
Any help on this would be very appreciated.
@T-1000
Good ideas about those changes. Maybe consider adding a diode after R3 and pointing to C11 so you do not bleed any B+ into L2.
wattsup
basically you shouldn't need anything on the function generator; depending on the actual drive of the generator... if you took something like 100nF cap and connected the signal generator to it you could see if it pulls to ground itself. A square wave probably is a solid sharp rise and return to ground; if that's the case, the mosfet gate will be quite happy getting filled to its capaictance at a voltage and being reverted to a ground state...
the gate capacitance isn't much; probably 1 or 2 nF... maybe more... but 10uF cap to test would be more generous... but you might start see a leading /falling slope increase...
Quote from: giantkiller on 2014.05.28, 01:02:53
Correct! The diodes would shut off the echoing back and forth.
The "Echo Effect" perfect you nailed it ;)
If you just want to tune the transformer.
I think that connection between the drain and C3 needs to be broken or a catch diiode inserted or the capacitor will destroy the FET on "first turn on".
Quote from: Dave45 on 2014.05.28, 12:42:05
If you just want to tune the transformer.
In this connection type you will need to use inverter circuit to pulse L2 . But also as soon as you discharge C3 there will be much more current consumed on L1 for C3 recharge.
You can try to solder circuit as I provided with fixes (just the driving transistor might be connected in different way) and you will see.
Also to isolate MOSFET - the dual/triple shotky diodes in series between coil and transistor solves problem.
@Dave45
For mosfet pulse testing I have tried something much easier as explained below.
@all
Thanks for your posts. This problem of mine should be present in many mosfet pulsing systems. Right now I removed everything from the mosfet and I am just using it via a 12vdc battery and some leds with the FG around 1-2 Hz and 10% duty and 4-5 volts amplitude. Back to basics. The problem is at the 90% off time, the LEDS are still slightly lit which is a major no no if you are looking to pulse the L1 from nothing to full impulse thus creating the maximum change in the core that would then be imparted to L2. I made a small diagram showing this simple set-up that does not produce clear off times on the LED. Without clear off times, you get weak pulsing and low change in the core.
I know I am a dumb ass when it comes to formal EE but I remember very clearly at the top of this thread where I explained great concerns in the diagram. Seems like we are coming full circle here with guys having the device long enough on the bench with the AK30 circuit as originally shown and now looking to make changes that reflect a more logical circuit function.
At the time I tried explaining that since B+ enters L1, exits L1 through D5 and then heads to the load, this means the load is receiving a positive B+ at all times. This means at the base of L1 where the mosfet is, that line is always an active B+ (always passing B+) and not the true negative side of a normal coil pulsing scheme where the B+ is dead ended until it is connected to ground via the mosfet. These are two totally different effects.
Now when the mosfet closes, the ground is hitting the B+ that is exiting the L1 but is still continuing on to the load as B+. So what does that mosfet pulse accomplish? A major mix up or a token pulse riding above straight DC to load.
Then you have L2 and we all are wondering what the hell the polarities are in that coil when is becomes energized via the core flux. Having only R3 between the B+ and top of L2 means B+ is entering the top of L2 which means the bottom of L2 should be negative if the coil is to function in any form of directionality. But the bottom of L2 has a D6 diode pushing the B+ into L2 and that comes from D5 and that is also going to the load. So you have L2 that is biased to B+ from both ends at all times, then how can you expect L2 to produce a clear polarity of its ends to then push some B+ back into C11 to initiate some level of looping? You cannot. L2 is just being held in, hmmmmmm, call it "stasis" since that L2 is simply being frozen in place. L2 has no possible way of anchoring itself to a ground on one side, (lowest potential) to then produce a high positive on the other side to initiate any output. It cannot. That's why I mention to @T-1000 to consider putting a diode between R3 and C11, to keep B+ out of L2 and let that new diode recharge and hold C11.
Then the D6-R2-C4 is doing what for L2. We are hoping it will produce a floating ground so L2 can generate some form of clear polarity but I am sure this combo is doing absolutely nothing in that regard because the only thing circulating in that combo is straight B+ which is much more stronger and consistent then any change that can occur via flux transfer to a frozen L2 . If that combo generated any form of B-, it would be shorted out continuously thus creating a major waste of energy.
The final point is how to make sure the mosfet pulsing is causing a true on/off of the B+ going through L1 because your scope can see and show the pulse but it cannot confirm if the pulse is going through L1 or is the scope just showing the mosfet pulse that hits the gate and is simply reverberating throughout the circuit or system. How do you know your mosfet is pulsing the L1?
Hahaha. Always questions.
wattsup
PS: I am adding the following pdf on MagAmps because the circuits shown are so close to Akula, TK devices. I think the MagAmp is exactly what Akula made with the yoke device. The yoke device he wants us to forget. He used the yoke to pulse the main coils. I had tried this during the yoke builds we did with Wesley and had found that when I pulsed one primary and applied DC to the other primary, the secondary voltage dropped from around 650 to around 250 volts. This means the yoke could be used as a huge mosfet.
Quote from: ION on 2014.05.28, 13:16:53
I think that connection between the drain and C3 needs to be broken or a catch diiode inserted or the capacitor will destroy the FET on "first turn on".
Quote from: T-1000 on 2014.05.24, 09:57:57
Hi there,
I managed to fix akula 30W circuit http://www.overunityresearch.com/index.php?topic=2358.msg36611#msg36611 with minor modification in way it is supposed to work:
1) The existing feedback loop needs to be removed and capacitor C5 + resistor R5 bottom right need to be replaced by one more layer of LEDs (I have 7 in series).
2) The connecting resistor R7 from TL494IN to that feedback loop needs to be grounded instead.
3) The connecting resistor R9 to plus side then pot and PIN 1 of TL494 needs to be disconnected from positive power input side and reconnected into positive side of charging capacitor C3.
In this way you will have pot R11 setting maximum level of capacitor C3 charge and regulating when TL494IN generator should stop after reaching voltage level. There is very sensitive level when it starts oscillating with ON/OFF modes on low frequency and that was intended to create secondary low frequency for transformer.
Also I did find out one issue there as well:
As soon as you connect L2 into circuit or just bulb back from positive side of capacitor C3 to positive side of capacitor C11 - the circuit starts to consume amperes instead if few milliamperes of current. This explains why ITSU and others have so much current draw. Without feedback loop LEDs get lit just fine. The feedback loop gets charge up to 20.5V in my case and I will need some another pulser/counter circuit to recharge primary capacitor C11 when the TL494IN is in OFF state perhaps.
P.S> I will repost my fixed version of circuit later when will have chance to access my computer, or Grumage/others who received copy might put it here....
Also I winded transformer with 1mm iron wire from http://www.ebay.co.uk/itm/GREEN-WIRE-FENCING-NETTING-PLANT-IVY-TWIST-TIE-TREE-SUPPORT-TYING-FIXING-GARDEN-/360611320881?ssPageName=ADME:L:OC:GB:3160 on ferrite E-core ETD49-CF139 from Power Magnetics http://powermagnetics.co.uk/pace-components/ferrite-cores/etd-cores/etd49-cf139 on which I made gap with isolation tape on all 3 junctions. The turns on secondary was 35+17 CW and on primary the another half with 17 turns CCW as per my suggestion in beginning of this thread.
Hi T1000,
thanks for the input, i made the modifications as shown on bold above (without the extra leds) , so without the duty cycle control for which is still use the "verpies" method, see diagram below.
Would that pose any problem?
I too used the iron wire with similar coil setup (only 45 sec / 15 prim), and the thing is much more alive ;D
It screems and squeals and has ever changing scope traces (oscillating?).
Anyway, i will make some video tonight if i can get it somewhat stable, here a first scope capture:
Ch1 (yellow) gate signal, groundless as the other to probes grounds are at +12V level, so its somewhat distorted and just for reference.
Ch2 (blue) = voltage across R1, with C2 NOT removed (L1 Current)
Ch4 (green) = voltage across R3 (L2 current)
Regards Itsu
As it is a Holiday overhere, but also a rainy day, i could already make the video of the modified Akula 30W device.
With the 3 potmeters i now have many tuning posibilities :D
Anyway, here the video: https://www.youtube.com/watch?v=rXd2oMt9EsM&feature=youtu.be
Regards Itsu
Quote from: Itsu on 2014.05.29, 12:34:26
As it is a Holiday overhere, but also a rainy day, i could already make the video of the modified Akula 30W device.
With the 3 potmeters i now have many tuning posibilities :D
Anyway, here the video: https://www.youtube.com/watch?v=rXd2oMt9EsM&feature=youtu.be
Regards Itsu
I see it is still drawing loads of current.
Please try to add flip switch to L2 and try to run circuit again with L2 in disconnected state. That will radically drop down voltage consumption.
Also the next level modification attempt in my side will be with making additional coil in transformer for secondary sine wave impulse input (transformer action) and grounding L2 with output to additional electrolytic capacitor over paralel diodes and another diode back to the input of power supply as akula shown me another generic circuit (need to ask his permission to publish it)...
Cheers!
Quote from: T-1000 on 2014.05.29, 18:52:02
I see it is still drawing loads of current.
Please try to add flip switch to L2 and try to run circuit again with L2 in disconnected state. That will radically drop down voltage consumption.
I have the stop switch in the L2 circuit as on the diagram, but flipping it does not really make much change in current consumption.
Why would it, as the main current path (start - L1 choke - R1 - L1 prim. - Drain/source MOSFET - ground) is still there?
Regards Itsu
Quote from: Itsu on 2014.05.29, 19:02:48
I have the stop switch in the L2 circuit as on the diagram, but flipping it does not really make much change in current consumption.
Why would it, as the main current path (start - L1 choke - R1 - L1 prim. - Drain/source MOSFET - ground) is still there?
Regards Itsu
With L2 not connected in my coil the current consumption with LEDs still lit can go down to ~60mA. Off course, in my case there are only 7 LEDs in series, not 2 strings of them. ;)
I can say, that Topruslan write, that this lantern 30 W schematic is in general completed. He try make again this divice and not make. Hi say something, that ferite in this lantern is general, with not needed ferite core this lantern not work. Or maybe he something not good make, so not he get succes then replicate another time this divice, but this my opiom. Akula also, I remember, say, something like this that not all ferites good and that you can loose ferite futures then you test it. Topruslan say, that for him just look then he first time make this lantern 30 W. And write, that ferite destroing, he tired regluing ferite core in working lantern. Maybe he say true, maybe he say not true, is for you chose he say true or not true.
This is the circuit that should have been tested, using separate chokes, transformers, flybacks have been tested extensively.
Its the resonant effect between two coils that needs to be investigated or the echo effect.
Also V8carlos's bifilar capacitance experiments need further investigation.
And yet another one to hopefully glean some more insight from:
Quote from: Matt Watts on 2014.06.02, 05:43:15
And yet another one to hopefully glean some more insight from:
Thanks for sharing v4 video. I hope we see a schematic soon.
I think Avalon succesfully replicated Akula small device.
https://www.youtube.com/watch?v=vgYLVyswgeQ&feature=youtu.be
Quote from: Matt Watts on 2014.06.02, 05:43:15
And yet another one to hopefully glean some more insight from:
Ok, here is circuit attached from akula himself. :)
The idea behind is - after first pulse you let BEMF to reach maximum and hit with current from transistor again on its peak time so the magnetic domains when coming back wil be supported by current and voltage from BEMF WILL ADD with current from transistor so C28 cap will charge from much higher voltage on positive half-cycle. The LEDs are powered from BEMF. After this pulse you need to let ferrite's magnetic domains to return into first position. This is why there is low frequency in akula's ferrite case.
EDIT: "только поесни что вней добавлин повторитиль чтобы не мучитса с управлением транзисторов! ато упруца в то что генератор не сможет управлть транзисторами так кака надо." - the main transistor driver correction added in second circuit edition.
Cheers!
Quote from: T-1000 on 2014.06.02, 19:35:17
Ok, here is circuit attached from akula himself. :)
The idea behind is - after first pulse you let BEMF to reach maximum and hit with current from transistor again on its peak time so the magnetic domains when coming back wil be supported by current and voltage from BEMF WILL ADD with current from transistor so C28 cap will charge from much higher voltage on positive half-cycle. The LEDs are powered from BEMF. After this pulse you need to let ferrite's magnetic domains to return into first position. This is why there is low frequency in akula's ferrite case.
Cheers!
Thanks a million T-1000....
Do you know the value for T-1 is this like a TL431 or an SCR ?
T1 is switching Ground for R58 & C40.....
Thanks again, peace
lost_bro
@T-1000 thanks thanks thanks
Good evening T-1000
Looking over the Akula schematic I see that once again the parts count on the schematic does NOT match the parts count in the video or the picture on the same schematic.
What is your opinion on this? The schematic shows twice as many diodes, capacitors and transistors than the video or photo on the schematic.
So, all the parts are NOT needed to make this work?
take care, peace
lost_bro
Yeah, at least I see choke in good position. :D
The generator in left is said to be 1:1 by akula, the correction generator in the right was not in video.
Also that generator is supposed to reset circuit by recharging power input capacitor when parameters change (due temperature/humidity/etc).
So it is up to you to replicate effect now... :)
Cheers!
Quote from: lost_bro on 2014.06.02, 19:58:06
Thanks a million T-1000....
Do you know the value for T-1 is this like a TL431 or an SCR ?
T1 is switching Ground for R58 & C40.....
Thanks again, peace
lost_bro
T-1 is a PUT, "programmable unijunction transistor", but I think there must be an error in the schematics because it will be instantly destroyed upon turn on as both C28 and C40 have nothing to limit the current that will circulate in those devices upon turn on of T-1. The polarity of operation appears correct.
I think the idea was to use the extra winding to create a high voltage charge on C40, then discharge at the correct time through the PUT and into C28 to pump it up, but I think it will destroy the PUT unless some surge limiting resistor is inserted in the loop. A small inductor would be even better as it won't dissipate the energy.
PUT's can only take so much current, but a PUT can be fabricated from two complementary transistors, similar to an SCR. A PUT just brings out the unused terminal. See also "silicon controlled switch" a four terminal device where both the SCR and the PUT "gate" connections are available.
Akula is a careful designer with good attention to detail, however like most proprietary recipe's the chef leaves out some ingredients or adds others so no one can exactly duplicate his soup.
peace bro
Hi there,
There is one mistake in circuit diagram I just noticed - the VT14 has wrong arrow, it is supposed to be PNP transistor KT361A - http://alltransistors.com/pdfdatasheet_russia/kt361a-b-v-g-d-e.pdf
Hopefully that will not stop from trying circuit as everyone gone quiet here...
Quote from: ION on 2014.06.03, 20:55:04
T-1 is a PUT, "programmable unijunction transistor", but I think there must be an error in the schematics because it will be instantly destroyed upon turn on as both C28 and C40 have nothing to limit the current that will circulate in those devices upon turn on of T-1. The polarity of operation appears correct.
I think the idea was to use the extra winding to create a high voltage charge on C40, then discharge at the correct time through the PUT and into C28 to pump it up, but I think it will destroy the PUT unless some surge limiting resistor is inserted in the loop. A small inductor would be even better as it won't dissipate the energy.
PUT's can only take so much current, but a PUT can be fabricated from two complementary transistors, similar to an SCR. A PUT just brings out the unused terminal. See also "silicon controlled switch" a four terminal device where both the SCR and the PUT "gate" connections are available.
Akula is a careful designer with good attention to detail, however like most proprietary recipe's the chef leaves out some ingredients or adds others so no one can exactly duplicate his soup.
peace bro
Good evening ION
Another discrepancy is the fact that the KT361A is a PNP and his schematic shows a NPN symbol.......
Doesn't the Cathode go to ground on the PUT in standard configuration?
Akula has his upside down with the Anode on ground.
take care, peace
lost_bro
Quote from: lost_bro on 2014.06.04, 00:21:32
Good evening ION
Another discrepancy is the fact that the KT361A is a PNP and his schematic shows a NPN symbol.......
Doesn't the Cathode go to ground on the PUT in standard configuration?
Akula has his upside down with the Anode on ground.
take care, peace
lost_bro
Hi lost bro
The polarity is correct for the PUT, as the negative end of C40 will charge to below ground, making the anode positive respectively.
I see that PUT's such as the 2N6027 is rated at 50 mA, 300mW max.
The transistor I agree is the wrong type, should be PNP.
The rest of the circuit is fairly easy to understand, nice to see Akula can design with standard parts and discrete transistors.
Although the PUT is claimed to be old, I used to design with them back in the day and they are nice parts. (guess that makes me old) ;)
Some types include: 2N6027 2N6116 2N6119 MPU131 MPU231 SK3628
peace, ION
Quote from: ION on 2014.06.04, 00:48:34
The transistor I agree is the wrong type, should be PNP.
The KT361 is PNP as per specification.
I see confusion so fixed arrow in transistor marking on circuit.
P.S> The thyristor T1 should be blocking capacitor C40 discharge by default unless there is signal from VT14 to pass through.
Here I remove not nessary part, maybe for some be needed.
I like the placement of a choke there, that at least provides a good feed from positive to ground for recycling purposes...
I've been considering akula's screenshots. 3ns rise time... but with a 10ohm resistor before the gate in the KT805, it's only going to be only just barely on... enough that when Primary conducts, secondary also reacts, and causes a raise in the ground voltage, which closes the KT805 (base-emitter not high enough now).
The pulses that result from such a condition are increasing frequency, and staggered at a 'fixed' distance... it's not really 'ferroresonance' more like 'ferro-feedback' ... but then the pulse from 7414 goes off, so only a couple of these spikes happen.
which means primary feed of secondary must be from VD4,VD3,VD(unlabeled and choke) side to ground ... or from ground to these?
The issue would be, as drawn, all grounds are the same (which in theory they are), but relative placement of the parts is going to matter also... for instance if L2 is tied directly to the emitter of KT805, and then by a junction to another, because there is more space, the junction is slightly lagged... so maybe the mass ground of 1000u's and 100u's scattered all over don't apply as much....?
But the impedance of choke is going to make L2 go to a high voltage against Low MSR3545 and VD4 as capacitances...
So for that to happen, current 1 is to ground, then the high voltage and ground cause a sharp spike through L2 which is seen in the collector of KT805 from L1... partially becuase of ground distrtion partially just the pulse in the coil...
which will pull on unlabled, unparametered choke, which should feed from positive to ground... actually the capcitance of VD3,VD4 causes the coil to revese quickly and conduct from ground with a very high voltage, which is the distortion on the feedback....
if the pulse width was wider it would be more spikes with a decaying time and magnitude
I dunno it's what the pulses look like to me, from playing with lots of inductances and chained transistors....
the transistor must be off for a high voltage to develop, and that's a forward EMF that drives that (continues sucking electrons)... in a diminishing current, which should reverse the other coil also which is already wanting to itself reverse conduct... ... I dunno I'm losing the tought...
One can make similar pulses just adding a small inductance to the emitter to ground.
Although... that doesn't explain the huge dip in voltage when (turning up the variable cap on the driver transistor?)
"which means primary feed of secondary must be from VD4,VD3,VD(unlabeled and choke) side to ground ... or from ground to these? "
Only from VD3 getting energy back. From VD4 going to LED's.
Quote from: MenofFather on 2014.06.06, 12:46:11
"which means primary feed of secondary must be from VD4,VD3,VD(unlabeled and choke) side to ground ... or from ground to these? "
Only from VD3 getting energy back. From VD4 going to LED's.
Thanks MenofFather and d2xor for your analysis.
Since there are no phasing dots on the transformer we can only guess the following:
VD3 supplies energy from the flyback pulse in standard fashion from the flyback pulse to charge C23, VD4 supplies energy to the LED's from the forward drive pulse (while the inductor is charging, forward di/dt) so the LED's seem to be supplied directly by the forward pulse, not the BEMF or flyback.
Of course, this could be in reverse of the above statement if one of the windings is reversed.
The other part of the circuit is a high voltage rectifier multiplier circuit whose charge buildup on C40 is timely discharged through the PUT T-1 into C23.
Akula misses giving a very important piece of information when he does not supply the phasing dots for the coupled inductors.
Quote from: ION on 2014.06.06, 13:14:40
Thanks MenofFather and d2xor for your analysis.
Since there are no phasing dots on the transformer we can only guess the following:
VD3 supplies energy from the flyback pulse in standard fashion from the flyback pulse to charge C23, VD4 supplies energy to the LED's from the forward drive pulse (while the inductor is charging, forward di/dt) so the LED's seem to be supplied directly by the forward pulse, not the BEMF or flyback.
Of course, this could be in reverse of the above statement if one of the windings is reversed.
The other part of the circuit is a high voltage rectifier multiplier circuit whose charge buildup on C40 is timely discharged through the PUT T-1 into C23.
Akula misses giving a very important piece of information when he does not supply the phasing dots for the coupled inductors.
Well, I translated his explanation already - the capacitor C28 charge is on positive cycle half and the LEDs are on BEMF spikes in negative cycle half.
When BEMF reach maximum value he turns on transistor for positive cycle half which gives big pulse to capacitor C28 charge when mixed with BEMF. So he will be charging cap not from 9V but from 30-60V spike + current from primary coil.
Also the second circtuit part is to recharge primary power input cap over T1 when circuit -stops- running, this is attempt to reboot circuit after parameters change...
Quote from: T-1000 on 2014.06.06, 16:11:20
Well, I translated his explanation already - the capacitor C28 charge is on positive cycle half and the LEDs are on BEMF spikes in negative cycle half.
When BEMF reach maximum value he turns on transistor for positive cycle half which gives big pulse to capacitor C28 charge when mixed with BEMF. So he will be charging cap not from 9V but from 30-60V spike + current from primary coil.
Also the second circtuit part is to recharge primary power input cap over T1 when circuit -stops- running, this is attempt to reboot circuit after parameters change...
I don't see a C28: Edit( nevermind I found C28; it was a different font so I glanced over it not at it);
Edit: the simulation works, until I cause current to flow in 2 different directions to a capacitor... that is a curernt flowing into the positive plate and a current flowing into the negative plate (overall gaining in postiive) but if there's 'A' displacmeent current... can that be simultaneously bidirectional? ... like usually there A in and A out, not in and in....
the idea is to charge C+ using I(unlabled) as a ... choke is a bad word... the idea wold be to induce a DC current in I(unlableled) which is forced from positive to ground, thereby replenishing ground, and drawing postiive back to high... I stopped by the local store, but chokes there are like 200mA; and I think I want more current... since I can generate a trapped high between VD4, the choke (it's a pulse I'm giving it, so the impedance is high; if there is still a remaining current in the choke, the impedance will be slightly higher) and the power input pins... this is a tiny capacitance though, so it is only going to pull the choke for a very short time, but at a high current rate... but then the current should keep flowing back through VD3/VD4 and the LEDs to ground...
charging C38 for the secondary's operation and or C37 ... which actually C37 should probably be included on the schematic without the secondary block, which would increase the capacitance of that BEMF voltage net... 37 and 40?
----------------
maybe scratch most of that; C28 ....
-----
Okay so resetting c28; gotit... that would be an issue if that was deleted; ie the choke wouldn't transfer much current before balance was satisfied...
not sure the voltage multiplier is nessciary ( it's a current divider) oh... then to trigger the zener and allow a discharge of ground to lower c28...
man why's it gotta be so complex
----
so what's the voltage multiplier thing for? And sin't it missing a diode between c37 and 39? That rail becomes just a high negative potential... which would want c39 side toward the diodes want to be more positive... so it would have less effecitve capacitance... C38 diode side is just going to go positive once....
Were writen "вторичный блок (не обязательный)" in circuit of lantern 4, that mean, secondary block, not nessary. So circuit work and without it. I not see reason use that secondary block if work without it. It only stibilise.
C28, seems, is better not remove.
:)
This schematic probarly can be little simplifed. D100 looks also can be removed, but to not make mistake, I it not remove.
:)
Quote from: MenofFather on 2014.06.07, 18:54:43
This schematic probarly can be little simplifed. D100 looks also can be removed, but to not make mistake, I it not remove.
:)
I think D100 could be removed... works better without the additional voltage drop (so the C28 and C+ can balance closer to 0)
I agree with restoring C28; which is going to allow the choke to flow between two similar potentials... and the L2 is the return from C28 to ground...
REmoving the parallel resistor to the choke is probably OK also, otherwise you need another diode, which then makes the resistor not be used anyway... what I had in the simulator was that the current through the coil ended up just going through the resistor in a short loop.
This looks like it will be a very long runtime...
The simulator does make the difference across C28 grow with pulses
highest net on the right of the choke gets charged pretty high, power net C+ charges back up to original voltage
The sim... (http://www.falstad.com/circuit/#%24+1+5.0E-9+7.308818067910767+24+5.0+50%0AR+112+112+64+112+0+0+40.0+5.0+0.0+0.0+0.5%0Ag+160+368+128+368+0%0As+160+368+208+368+0+0+false%0As+112+112+160+112+0+0+false%0Ac+256+224+256+288+0+0.001+7.568371482764231%0Aw+256+288+256+368+0%0Aw+256+368+208+368+0%0AT+352+224+400+288+0+0.002+2.0+0.13286733956786034+4.116095213594576E-7+0.999%0As+352+288+304+288+0+1+false%0Ax+305+279+357+285+0+24+SW1%0Ad+432+240+432+288+1+0.805904783%0Ad+432+288+432+320+1+0.805904783%0Aw+432+240+432+224+0%0Aw+432+224+496+224+0%0Aw+256+416+256+368+0%0Al+400+176+352+176+0+9.999999999999999E-5+-1.6732761324439114%0Aw+400+176+400+144+0%0Aw+256+144+256+112+0%0Aw+432+320+432+352+0%0Aw+432+352+464+352+0%0Aw+464+352+464+176+0%0Ad+256+112+320+112+1+0.805904783%0Aw+320+112+400+112+0%0Aw+400+112+400+144+0%0Aw+176+112+160+112+0%0Ad+256+144+256+192+1+0.805904783%0Aw+256+224+256+192+0%0A162+496+416+496+368+1+2.1024259+1.0+0.0+0.0%0A162+496+320+496+272+1+2.1024259+1.0+0.0+0.0%0A162+496+272+496+224+1+2.1024259+1.0+0.0+0.0%0Ac+176+48+176+112+0+1.0E-10+0.0%0Aw+176+48+112+48+0%0Aw+112+48+112+112+0%0Ac+448+256+448+288+0+1.0E-10+-5.07700713724135%0Aw+448+288+432+288+0%0Aw+448+256+448+240+0%0Aw+448+240+432+240+0%0Ac+352+320+304+320+0+1.0E-10+5.013286732563271%0Aw+352+320+352+288+0%0Aw+304+320+304+288+0%0Ar+352+224+304+224+0+0.1%0Aw+304+224+256+224+0%0Ar+416+224+416+272+0+1.0%0Aw+432+288+400+272+0%0Ar+464+176+432+176+0+1.0%0Aw+432+176+400+176+0%0Aw+256+288+160+288+0%0Ac+160+288+160+352+0+1.0E-10+0.0%0Aw+160+352+160+368+0%0Ar+448+416+480+416+0+1.0%0Aw+496+416+480+416+0%0Aw+400+368+416+368+0%0Aw+416+368+416+416+0%0Ac+304+176+336+176+0+1.0E-10+0.8950964260319587%0Aw+304+176+304+144+0%0Aw+352+320+352+336+0%0Aw+352+336+224+336+0%0Aw+224+256+112+256+0%0As+112+256+48+256+0+0+false%0Aw+112+112+112+144+0%0Ar+224+336+224+304+0+0.1%0Aw+224+304+224+256+0%0Aw+400+272+416+272+0%0As+128+288+128+320+0+0+false%0Aw+160+368+128+320+0%0Ar+160+288+128+288+0+1.0%0Aw+304+144+288+144+0%0Aw+288+192+256+192+0%0Aw+176+112+256+112+0%0Aw+112+144+48+256+0%0Ad+496+368+496+320+1+0.805904783%0Aw+416+416+256+416+0%0Ax+132+140+177+145+0+20+SWP%0Ax+84+244+129+249+0+20+SWP%0Ax+67+315+115+320+0+20+SWG%0Ax+159+396+207+401+0+20+SWG%0Ar+256+288+304+288+0+0.01%0Aw+416+224+384+208+0%0Aw+704+368+704+384+0%0Ac+496+160+496+112+0+1.0E-6+0.0015078007137225083%0Aw+496+112+544+112+0%0Aw+544+112+544+352+0%0As+496+224+496+160+0+0+false%0Aw+544+352+544+416+0%0Aw+416+416+432+416+0%0Aw+432+416+448+416+0%0Ac+432+368+432+400+0+0.001+5.0%0Aw+432+400+432+416+0%0Aw+432+368+432+352+0%0Aw+112+224+48+256+0%0As+112+224+176+224+0+0+false%0Aw+464+352+592+288+0%0Aw+592+288+592+64+0%0Aw+592+64+224+96+0%0Aw+224+96+176+224+0%0Ar+544+416+496+416+0+4.0%0Aw+384+208+400+224+0%0Aw+352+176+304+144+0%0Aw+336+176+352+192+0%0Aw+352+192+400+176+0%0Aw+288+192+288+144+0%0Aw+400+288+400+368+0%0Ao+41+64+0+291+8.749002899132048+0.17498005798264096+0+-1%0Ao+18+64+0+291+8.749002899132048+8.543948143683641E-5+1+-1%0Ao+86+64+0+35+10.0+3355443.2+2+-1%0Ao+13+64+0+291+0.009765625+9.765625E-5+3+-1%0Ao+71+64+0+291+7.62939453125E-5+3355443.2+4+-1%0Ao+11+64+0+291+1.25+9.765625E-5+5+-1%0A)
So essentially a pulse generator with a short duty cycle, a few coils and diodes ... it doesn't look like anything special about the (7414 thing).. and some capacitance on the base of the driver transistor for extra sharp pulses.
might add a cap parallel to the LEDs .. won't get as much spike, but will get a longer continuous light...
Used my TL494 board to drive this... tried to put some capacitance in to cause the double-spiking, but it didn't work; I have a 5V regulator for the voltage output so I think it's got lots of current... tried tiny and large caps to no avail...
Reversing the secondary coil from the schematic allows slightly lower amp draw.
The C28 cap is too large at 1000uF ... but too small and the choke starves and develops its own back low EMF spikes. (diode to LED load to recover)
It is better to keep the diode between the choke and primary power. (which itself causes the back emf spike) even if the back emf isn't used, the diode reduces current draw.
It was lower amp reading on my power supply when I isolated the whole thing behind diodes.
Most optimal I could get was 30mA to get reasonable light on LEDs (indicator level at least)
I used my most recent pot core coil as the trasnformer, and another for the choke...
Was also better to add a diode between the LED load and ground, saved 10-20mA...
The TL494 board I have idles at 10mA when it's not generating any signal; and I'm using large darlington drive transistors...
did get a second or two of idle time before it turns off on 6.8mF cap...
the hex inverter is probably a lower idle; but I couldn't get down to 0.01A....
the simulator was able to charge the primary power cap; so I really think that in theory this should work as a long runner
I had it for a while doing the double-spike thing, but it turned out the secondary became disconnected from ground.
I had a couple flyback cores but they have since disappeared.... so waiting on a few replacement parts from mouser.
The wire should be large-ish; I used 20ga primary and 28ga secondary which works OK; but there's not enough room to use like 14ga wire primary in the pot core....
Have you all seen LS latest? Amazing how little power to light the LED! 8) O0
http://laserhacker.com/?p=420
@All
News from Akula0083
Download It before disappear....!!!
https://www.youtube.com/watch?v=9sYLdekiLD0
https://www.youtube.com/watch?v=FHIF49ZcavE
https://www.youtube.com/watch?v=uG705Ayant8
https://www.youtube.com/watch?v=OSvePx2Akhg
https://www.youtube.com/watch?v=XB5_njwaK8o
Quote from: TutorialFE on 2014.06.28, 11:47:59
@All
News from Akula0083
Download It before disappear....!!!
https://www.youtube.com/watch?v=9sYLdekiLD0
https://www.youtube.com/watch?v=FHIF49ZcavE
https://www.youtube.com/watch?v=uG705Ayant8
https://www.youtube.com/watch?v=OSvePx2Akhg
https://www.youtube.com/watch?v=XB5_njwaK8o
Hello All:
Did anyone save the listed videos?
I didn't get a chance to watch them.
Thanks
Take care, peace
lost_bro
Quote from: lost_bro on 2014.07.01, 15:02:01
Hello All:
Did anyone save the listed videos?
I didn't get a chance to watch them.
Thanks
Take care, peace
lost_bro
I've saved all the videos .
I'll charge it here betwen one week I will delete them.
https://www.youtube.com/watch?v=dl9KsDB_RYs&feature=youtu.be
Quote from: TutorialFE on 2014.07.01, 16:49:04
I've saved all the videos .
I'll charge it here betwen one week I will delete them.
https://www.youtube.com/watch?v=dl9KsDB_RYs&feature=youtu.be
Thanks TutorialFE,
I will download today,
Take care, peace
lost_bro
Hi guys,
After long exploration why there was current sensor in akula's 30W circuit I finally came to conclusion - it is automatic frequency control implementation to keep transformer in resonance. Obviously it is badly implemented as it should be current sensor right on the wire directly from transformer but still good try.
Here is good read (in Russian, use google translate) about this stuff:
http://www.icct.ru/node/85
Cheers!
Quote from: lost_bro on 2014.07.01, 15:02:01
Hello All:
Did anyone save the listed videos?
I didn't get a chance to watch them.
Thanks
Take care, peace
lost_bro
I only save and wach last video, but in it nothing important. Only showing Tesla coil. I think if he made private, let be private and not need it show for others, I in my computer also remove this video.
:)
@All
Does anyone know Mike_M?
He asked me to upload videos of Akula ...
@Dear All
last news from Akula
https://www.youtube.com/watch?v=7sJ6lg3mKS4&feature=youtu.be
https://www.youtube.com/watch?v=w_WCcUXvZK8&feature=youtu.be
Quote from: TutorialFE on 2014.07.30, 17:20:44
http://www.youtube.com/watch?v=w_WCcUXvZK8
Am I looking at a modern SM TPU here?
The disassembly portion of this video sure looks familiar.
Very interesting the Asian crowd that appears to be embracing this technology. Looks like we are going to get left in the dust if I don't get onboard soon.
Yeah, looks TPU-like.
Don't recall any schematics coming through for that..
https://www.youtube.com/watch?v=z37vpqjGKuE Oh; guess that's the one with horizontal 3 coils
I'd like to know what can cause the collector to go low and stay low.... lower than ground. Shouldn't the positive supply voltage at the other side bring it up quickly (even in a ringing like?)
First device in latest videos was 'lantern no4' not 30watt...
re 30watt: Pretty sure the right side diode between from the secondary and the load is backwards... otherwise that cap will never be higher than the supply cap, because it will conduct through the secondary... unless that's a way to cause the primary to reverse conduct back from power to ground, but in simulations there's not enough force to get the primary to back-conduct.
@d3x0r
Keep up the great research! I'm going to play with Lasersaber's simplified version, but have to wait till September when I'll have the time. Just too much I have to get done this summer.
4Tesla
laser saber's simplified version; recomplicated...
added more of the components from lantern 4 to use negative pulse to light more LEDs and recover some of the positive...
Should omit the load on the positive side... just added it because it can be done...
but too much and it stops having enough to trigger its own oscillation.
Another one I just bumped into:
http://www.youtube.com/watch?v=GmlpV1MWm40
Single transistor, interesting transformer and quite an interesting way to start it too.
Now he says this is not over-unity in any way, just highly efficient.
Schematic would be interesting.....
@ALL
What do you say to cooperate on this device?
https://www.youtube.com/watch?v=rBAEzeIRHqg&list=UU_zn0cdzv_qfsxQUm4e7R9w
https://www.youtube.com/watch?v=kAz0wqQD4Zw&list=UU_zn0cdzv_qfsxQUm4e7R9w
https://www.youtube.com/watch?v=xOc-7EXMvNc&list=UU_zn0cdzv_qfsxQUm4e7R9w
Hi,
Ruslan is basically experimenting on same device as Roman shown with 1 kW+ one.
Also that should go on separate topic ;)
....it would be nice if someone translate the Russian Ruslan in English ..
Ruslan last video showed everything that was to be shown and convinced me that it is not a fake.
I think it's not hard to replicate it the problem for me are the electronic components to be assembled in a technical way and not in an improvised way It would need like someone who did produce the pcb as Akula 30watt to weld all the components etc. etc.
A new topic would be right... if other users think that it is important
Quote from: TutorialFE on 2014.09.27, 19:18:37
A new topic would be right... if other users think that it is important
Hi TutorialFE,
yes i think its important to start a new topic.
It would help to publish the latest diagram from Ruslan so it can be checked by the experts here.
I f.i. need to know what the circuit with the 7400 and 7414 is doing (is it peak driving the kacher?)
Thanks, regards Itsu
I've watched this Rusian video a few times and cannot detect any faking, so yes, maybe worth looking into further under a separate thread. However, there is a fairly long period when the demo board is moved across to the table whilst the camera is put down, possibly long enough for an 'X' wire to be hooked-up. :-\
@Hoppy
...that's what I thought too, but this movement he has also made it in another video faster then I do not think that there are hidden cables
The principal point for me to remove all doubt is that the main components are the same for all the different inventions .. . Dally Akula etc etc there is always a large ferrite type youke tv or similar, large capacitors, 220v to 12v transformer and circuits for nanopulser and winding.
If you look at the devices Akula and Dally will always find these 5 main components O0
Quote from: TutorialFE on 2014.09.28, 20:45:36
@Hoppy
...that's what I thought too, but this movement he has also made it in another video faster then I do not think that there are hidden cables
The principal point for me to remove all doubt is that the main components are the same for all the different inventions .. . Dally Akula etc etc there is always a large ferrite type youke tv or similar, large capacitors, 220v to 12v transformer and circuits for nanopulser and winding.
If you look at the devices Akula and Dally will always find these 5 main components O0
no TV yoke in this last one http://youtu.be/kAz0wqQD4Zw?t=7m18s
still no thead? or do I have to find it?
Quote from: d3x0r on 2014.09.29, 04:44:47
no TV yoke in this last one http://youtu.be/kAz0wqQD4Zw?t=7m18s
still no thead? or do I have to find it?
http://www.overunity.com/12736/kapanadze-cousin-dally-free-energy/msg418645/#new
Quote from: TutorialFE on 2014.09.28, 20:45:36
@Hoppy
...that's what I thought too, but this movement he has also made it in another video faster then I do not think that there are hidden cables
The principal point for me to remove all doubt is that the main components are the same for all the different inventions .. . Dally Akula etc etc there is always a large ferrite type youke tv or similar, large capacitors, 220v to 12v transformer and circuits for nanopulser and winding.
If you look at the devices Akula and Dally will always find these 5 main components O0
It is clear from Rusian's various videos that he is basing most of his devices on Akula's various circuits. He shows a pile of Akula's circuit schematics on his work bench. However, Rusian has IMO carried out more convincing demos than Akula. This appears to have motivated a few experimenters on the OU.com forum to take him seriously and attempt replication.
Quote from: d3x0r on 2014.09.29, 04:44:47
no TV yoke in this last one http://youtu.be/kAz0wqQD4Zw?t=7m18s
still no thead? or do I have to find it?
Dear d3x0r and All.
The concensus of opinion seems to be toward launching a new thread ?
I can open one on my bench if you wish? Would someone like to suggest a title ?
I am at this time involved in more mechanical pursuits, whilst the weather here is still good enough to work in the back shop !!
I will probably put my electronic hat on when the snow comes !! :)
Cheers Grum.
Quote from: Grumage on 2014.09.29, 10:10:48
Dear d3x0r and All.
The concensus of opinion seems to be toward launching a new thread ?
I can open one on my bench if you wish? Would someone like to suggest a title ?
I am at this time involved in more mechanical pursuits, whilst the weather here is still good enough to work in the back shop !!
I will probably put my electronic hat on when the snow comes !! :)
Cheers Grum.
could be something generic like 'flavor of the week kapagens'
@All
I have not figured out why just published the video es Akula Ruslan immediately burst out of the schematics as if it was clear the operation of the device and then everything stops and does not go forward anymore apart from many of the comments ... someone has a logical explanation ?
@ALL
A new video... from Akula..
https://www.youtube.com/watch?v=awy9OgRd-D8
Quote from: TutorialFE on 2014.10.16, 21:29:26
@ALL
A new video... from Akula..
http://www.youtube.com/watch?v=awy9OgRd-D8
Thank you TutorialFE for posting this. I downloaded the video (as these seem to disappear after a short period of time) and have been studying it. Something hit me when looking at this and other similar devices that have a ground wire:
You all know about Tesla's Wardenclyffe project right? Do you suppose something like this was built in Russia for military operations? And instead of a tower it was housed underground. The idea being that a military unit could be dispatched and for their power needs, all they would have to do is drive a stake in the ground.
I don't know the background of Akula or Kapanadze, but if they knew about such underground signals, tapping into them wouldn't be that difficult.
Just something to toss out there. It would be interesting if someone knew for sure whether this type of energy production was a natural phenomena or if it has a man-made source.
Wardenclyffe will be open to the public on oct 25 for a few hours
@Matt
QuoteI don't know the background of Akula or Kapanadze, but if they knew about such underground signals, tapping into them wouldn't be that difficult.
the kapanadze device has a large Earth ground which when disconnected stopped the device not unlike T.H.Morays device. It was my though that the devices may operate utilizing the interplay between the Earth's electric and magnetic fields. The Earth has a vertical electric gradient rising by 100v/m and it also has a magnetic field. As Moray himself stated... we are swimming in a sea of energy, here is Moray's book "The sea of energy in which the Earth floats" which I consider required reading --http://www.free-energy-info.com/P26.pdf.
As well for those familiar with Moray, I researched Moray extensively a decade or so ago and geographically tracked his mysterious "swedish stone". It was in fact a rather rare form of quartz which led me to believe his device was a crystal radio of sorts, a 20,000 watt crystal radio.
AC
@Matt Watts
I've seen this tecnology personally..
http://www.iene.mediaset.it/puntate/2014/10/08/viviani-e-se-l'acqua-fosse-un-carburante-_8797.shtml
Do anybody here had experience with large long HV power lines ? I wonder if there is a limit of one section length between each ground connection (power line pole) ? What I search is if there is very long wire and I mean many miles,placed from South to North for example then would it generate substantial voltage between its ends ? Kind of Ferranti effect might be used to cover this effect ?
Quote from: TutorialFE on 2014.10.17, 15:33:12
@Matt Watts
I've seen this tecnology personally..
http://www.iene.mediaset.it/puntate/2014/10/08/viviani-e-se-l'acqua-fosse-un-carburante-_8797.shtml
Video, eight minutes, 41 seconds, correct?
I'm pretty poor at Italian, can you summarize TutorialFE? I'm seeing HydrOxy production here, but is the "aqua" fully running the car or just acting as a supplement?
Quote from: Matt Watts on 2014.10.17, 23:28:54
Video, eight minutes, 41 seconds, correct?
I'm pretty poor at Italian, can you summarize TutorialFE? I'm seeing HydrOxy production here, but is the "aqua" fully running the car or just acting as a supplement?
Matt it is just a supplement, I believe Nissan fit this in that car as standard. I saw a long video done in the USA on that, Nissan corrected the ECU to accommodate it with enhanced combustion and less fuel used.
I think only on this supper sport model which normally burnt a lot of fuel!! but I'm not 100% on that :-\
regards
Mike 8)
Quote from: Centraflow on 2014.10.18, 08:04:26
Matt it is just a supplement,
Mike 8)
I feel the same improve combustion and reduce consumptions
Marangoni start the study
http://tyre.marangoni.com/it/Tuning/Progetti/zHydro2010/Z-Hydronote.aspx#.VEIyVIusX_w
http://de.autoblog.com/2010/01/15/wasser-marsch-marangoni-showcar-nissan-370-z-mit-hydroantrieb/
https://www.facebook.com/pages/Nissan-370Z-Hydro-Marangoni-Tyre/193619767427?viewas=0&sk=info
here the man
http://www.mareasistemi.com/Downloads/LORENZO%20ERRICO/LORENZO_ERRICO.html
https://www.facebook.com/Hydromoving
@MattWatts
The invenction is based on the cold fusion discovery in the 1980's. The car work on 30% HHO and 70% fuel. The secret is the electronic that made a good signal for the cell to made HHo with low current.
Lorenzo is an old invetor that has Patented many inventions in his life as electronic components.
He made a simulation of approval and the results are wonderful in terms of abatement of pollution - 80% and the cost of gasoline consumed - 30%.
Initially, the Nissan had commissioned to make this change to the invention of the car, and after 5 years of Lorenzo work then he succed. Than ì Nissan executives have changed their mind and abandoned the project and left Lorenzo HHO alone with his invention .
They've given the car to Lorenzo but do not support it in order to develop the technology. All the devices are made in a traditional way and then have high costs but if they could enginered the Kit costs would be lowered to € 3000.
@All
I found this new video demonstration that there is no other trick in the free energy device to Akula0083
https://www.youtube.com/watch?v=hIqEHc9_Bwo&list=UUbcaIfmxRZhnnYVqG81kCZg
Quote from: TutorialFE on 2014.11.02, 16:54:59
@All
I found this new video demonstration that there is no other trick in the free energy device to Akula0083
https://www.youtube.com/watch?v=hIqEHc9_Bwo&list=UUbcaIfmxRZhnnYVqG81kCZg
Dear TutorialFE.
Thank you for the video link, most interesting!
One really big thing that has me sceptical, how did his camera remain functional near that Tesla coil?
When I recorded my Akula 0083 video last year, the first run was a complete bust! The camera completely froze up for most of the time. My second attempt, being aware of the problem I managed to film to the last 30 seconds before the video froze but luckily the audio carried on working.
See here......... https://www.youtube.com/watch?v=LOqjTHmmumU
For me, Personally I think something is not quite right here! What do you think?
Cheers Grum.
Quote from: Grumage on 2014.11.02, 17:39:46
For me, Personally I think something is not quite right here! What do you think?
The one thing I think isn't quite right is our understanding of what
energy actually is. Also, where does it come from and where does it go.
I do think energy is akin to life. Energy <--> Life
If energy can neither be created nor destroyed, it's likely that too holds true with life.
Most of us know how complicated it is to accurately define what life is, I think the same holds true with energy.
I can't speak for everyone, but tell me, when you see a self runner like this one of Akula's, doesn't something inside you recognize the device as being alive?
Hopefully I'm not the only one that sees the connection here.
Quote from: Grumage on 2014.11.02, 17:39:46
Dear TutorialFE.
Thank you for the video link, most interesting!
One really big thing that has me sceptical, how did his camera remain functional near that Tesla coil?
When I recorded my Akula 0083 video last year, the first run was a complete bust! The camera completely froze up for most of the time. My second attempt, being aware of the problem I managed to film to the last 30 seconds before the video froze but luckily the audio carried on working.
See here......... https://www.youtube.com/watch?v=LOqjTHmmumU
For me, Personally I think something is not quite right here! What do you think?
Cheers Grum.
Initially, the shooting of the video was taken from a distance and was approached by zooming,
then was very close.
I don't have an explanation why the camera does not broke.
Your video was enought ok... This device is not very powerfull only 20w Mybe is the reason of the camera didn't broke.
Hello everyone and Happy New Year !!!
We are still of the next steps of Akula0083 inventions that have been replicated by Ruslan Kulabuhov.
It has posted on Russians forums also electronics diagrams.
Someone of you is able to produce these electronic boards?
https://www.youtube.com/watch?v=EyP5IcVzLHk
http://realstrannik.com/forum/ustanovka-kulabukhova/170-analiz-ustanovki-bez-kachera.html
For producing (low-cost) PCB's. Yes it is made in China, good quality fast and low/cost.
I would like to suggest publish my contact details as follows:
Name: Grace Liu
Website: http://pcb.hqew.net/?s=545651
Email: hqewpcb1@gmail.com
Mobile: 86-15220275629
Skype: hqew-three
So, if and when you wish to order, just email Grace (a delightfully helpful and pleasant young lady!), mention the name Les Banki or Peter and attach the pcb file(s) you want made.
You can also use their web site directly to get instant quotes, without even having to "register"!
(So far, I have not found any other web site where you don't have to "register" just for getting a quote!)
@ Kampen72
Thanks for the info.
Let's see if other people are interested in this project.
Quote from: TutorialFE on 2015.01.20, 22:13:21
Hello everyone and Happy New Year !!!
We are still of the next steps of Akula0083 inventions that have been replicated by Ruslan Kulabuhov.
It has posted on Russians forums also electronics diagrams.
Someone of you is able to produce these electronic boards?
https://www.youtube.com/watch?v=EyP5IcVzLHk
http://realstrannik.com/forum/ustanovka-kulabukhova/170-analiz-ustanovki-bez-kachera.html
Hello
I looked at the zip files, and I don't see the NC drill files or the Gerber files needed to process these pcbs.
Is there more information that I am missing?
Thanks
take care, peace
lost_bro
@Lost_Bro
It is all I've found browsing the russian forum.
I hope that someone who knows the Russian language can be helpful to find the missing informations.
Cheers
It returned the shark ....
https://www.youtube.com/watch?v=yKoksIxSg1A
Don't be eaten!
Quote from: TinselKoala on 2015.07.21, 01:37:30
Don't be eaten!
The public finances of the states of the world are empty ... there are bigger sharks then Akula .... :P
Quote from: TutorialFE on 2015.07.21, 12:46:42
The public finances of the states of the world are empty ... there are bigger sharks then Akula .... :P
Right -- and how is it possible that all the countries of the world (essentially) are in enormous DEBT? debt to whom? what is going on here?
Quote from: PhysicsProf on 2015.07.21, 13:59:13
Right -- and how is it possible that all the countries of the world (essentially) are in enormous DEBT? debt to whom? what is going on here?
You hit the nail right on the head, where is all the money? apart from the few (thousand or so) super rich in the world sitting on their fortunes which will do them no good in the end, they will burn in hell or rot in their cofins with or without their money, like anyone else.
Well the two biggest banking families is probably the answer, and they don't keep it in a bank either, and it is all to keep it in the same way for future family generations for world power, they are the world power, always have been and always will be, unless one could put them all in one place and.......................... :D leave the thought up to you >:-)
regards
Mike 8)
For the "Forum Purists" this discussion would be deemed
"Off Topic" but, on the other hand, you're on a roll!
All Hail our Banking Overloads. (http://investmentwatchblog.com/all-hail-our-banking-overlords-we-work-for-them-plain-simple/)
Quote from: PhysicsProf on 2015.07.21, 13:59:13
Right -- and how is it possible that all the countries of the world (essentially) are in enormous DEBT? debt to whom? what is going on here?
It's called the "Central Banking System". I think if you dig deep enough you'll find that none of the Central Banks are owned by the countries they are located in, ie. From what I can tell the Federal Reserve Bank is not owned by the U.S. citizens, the Reserve Bank of Australia is not owned by the Australian citizens ect. ect. In other words they are privately owned and these private banks loan (numbers or digital money) to our governments and oversee the printing of the Cash amongst other things, The only thing the Government and by extension the citizens own is the paper or metal of the Hard Currency.
They also determine the interest rate for the "retail" banks to pay on non existent money (numbers) which the retail banks loan to us, then we do real work and repay the principal (which never really needed to exist) as well as the interest, with our labor compensation (pay), the principal can then be made to be completely non existent again, and the banks divvy up the interest. Something like that, much better explanations of the entire setup can be found online.
Here's the thing, it seems that the average person that is always busy working or having fun (fair enough at that too) but they do not seem to see through the veil as often as those that are less hectic.
It does take some "getting".
..
The beginning of the end was when they forced most peoples pay to be deposited directly into the bank. Imagine living in Greece and earning 200 dollars a day at your job but you can only withdraw about $70 a day. OMG. No way I would do that, my employer would need to change to cash money for wages or I would walk.
I think if they can get most of us to do that we would deserve all we get when we got it whatever that may be.
..
Quote from: Farmhand on 2015.07.22, 01:43:52
Imagine living in Greece and earning 200 dollars a day at your job but you can only withdraw about $70 a day.
It's a test and once they find the proper formula, you can count on it going global.
Quote from: Matt Watts on 2015.07.22, 02:19:51
It's a test and once they find the proper formula, you can count on it going global.
Those who had money in the bank is retreating them ...
The elites have mounted the brain like Hitler in the third rich ...
There are poisoning even with Chemtrails ...
Who would believe more in this economy ....?
to pay more and more taxes ....
People are tired and retreating all the money from the banks ...
Their link to the microchip under the skin we see if they can.... :-X
http://www.tzetze.it/redazione/2015/07/il_debito_mondiale_oltre_200_trilioni_di_dollari_lo_studio_se_crolla_la_cina_salta_tutto_il_pianeta/
Hi,
It would be good if you could move off-topic discussion into separate thread so akula devices repeated stuff would be not under not related things.
Cheers!
Quote from: T-1000 on 2015.07.22, 16:58:18
Hi,
It would be good if you could move off-topic discussion into separate thread so akula devices repeated stuff would be not under not related things.
Cheers!
Dear T-1000.
You beat me to it !! :)
Whilst I fully agree with your sentiments gentlemen, please start a new thread.
The Akula 30 W project might actually benefit from the use of TinMan's potted transformer. It's a thought that has been nagging away at me for some time now.
Your thoughts ?
Cheers Grum.
New akula videos (Save for offline before it get deleted):
https://www.youtube.com/watch?v=E0qVW3zNPkQ
https://www.youtube.com/watch?v=UViTEKQUCZQ
Quote from: T-1000 on 2015.07.30, 08:21:26
New akula videos (Save for offline before it get deleted):
https://www.youtube.com/watch?v=E0qVW3zNPkQ
https://www.youtube.com/watch?v=UViTEKQUCZQ
Rapid heating and cooling...?
I would have never thought of that.
Of course it only works until the core breaks, but it's a start.
Quote from: Matt Watts on 2015.07.30, 09:05:15
Rapid heating and cooling...?
I would have never thought of that.
Of course it only works until the core breaks, but it's a start.
In general it is changing material properties so it goes from magnetic to non-magnetic state and the heating is one of examples. And doing it with very low consumption of energy makes induction far greater than energy used to manage process... :)
Also by doing that with ferrites it breaks core so Roman had to glue it multiple times for recycling until it got to the point where there was no more magnetic coupling. This is what he said in video and also he still kept secret how he manipulates magnetic properties of material.
Quote from: T-1000 on 2015.07.30, 16:10:24
This is what he said in video and also he still kept secret how he manipulates magnetic properties of material.
Partially. He showed the copper ribbon used in there. If those strips do what I think they do, I imagine a fair amount of thermal activity going on within the pot core.
So my question is...
How might this apply to the Ruslan device having only an air core? Can the magnetic properties of the copper windings be manipulated in a similar fashion?
This device does have a powerful induction heater winding, but can it heat (and cool via thermal conductivity) the grenade winding rapidly enough to gain the same effect shown with this device. I suspect it might and if so, this is the reason no one has been successful at replicating it--we are all tuning to an incorrect operating principal.
@to All
....perhaps we have solved the problem of communication .... this speaks English ... O0
https://www.youtube.com/watch?v=nDbEsk89yMQ
T
that vid comes up as a Rodin coil tutorial
Rodin Coil Tutorial - How to Wrap a 12 Pointed Coil -
Chet
PS
I see you corrected link below O0
That's wrong video, sorry.
The https://www.youtube.com/watch?v=Pw8wyw2ZaAQ is the correct link.
The purpose of the video was to show 2 devices sharing same effect and drawing explanation. Roman talked a lot about magnetron effects adopted to that specific case but in my opinion it is all same with creating resonant directed electron radiation from Anode to Kathode (the https://en.wikipedia.org/wiki/Secondary_emission described that in vacuum tubes) where absorbed electrons are added to the current flow and this is where additional power came from...
Can I play too?
http://www.youtube.com/watch?v=80D92QaOcM4
http://www.youtube.com/watch?v=X9wxuRZV-Ro
"In my opinion" Akula is not showing you everything you _really_ need to know to understand how his devices work.
To light up LED it is enough but not enought to light up 1kW+ loads where same fundamentals are applied. Just there you see high voltage source such as Tesla coil for electron emission while for LEDs it is not required.
@ T-1000
What is Ruslan saying exactly in this comment back to Roma:
Quote from: Ruslan Kulabuhov
Я 5 копеек вставлю свои , Рома...
Ты намекаешь на меня ? Ок , не вопрос ! Но тут что я увидел сейчас , могу сказать одно - народ может и поверит. Вон космонавт попу подлизывать пытается не соображая ничего... А вот остальных жаль ,очень жаль ! Рома , последние минуты или даже секунды видео убили нафиг :))))))))))))))) И после этого ты из меня дурака делаешь ? Многие уже понимают о ком это ты :) К тому же .... куда то делся водород и атамарные процессы. Согласен , теорий много и у меня было. Но этот видос меня реально зацепил... Ну хоть про магнитные поля мы заговорили и особенно про отрицательные. Так уже хоть лучше выглядит :) Я не собираюсь тебя учить , но вот есть правда в Твоих словах Рома " Все системы работают по одному принципу" Согласен ! Но не эти барабаны и катушки :)) Киловатт , Германия 25 ватт и те видео , работа совершенно отличается. Собирать заряды и вообще как это делать , знаю очень хорошо. Это так... Что бы Ты понимал что я знаю всё от чего. Просто не буду тут писать много.... Обидно придуркам давать , которые обзываются. Я и тебя понимаю тоже конечно. Но в следующий раз показывай пальцем , а не намекай :))))))))) Я не рассыплюсь !
I sense some animosity in his comment. Has Ruslan made significant progress he sees Akula's work to be a mere toy?
Please explain if you will.
Also, I attached his sketch where he describes the coils themselves, one being an anode and the other a cathode. Interesting concept when applied to coils of wire.
(caution: editorial comment, skip if you find it offensive)
If you watch this video (or many youtube videos) with the sound off, you realize how little information is conveyed for so much bandwidth used.
There is much hand waving, picking things up and putting them down, only to pick them up again, then more hand waving.
Then there is the really bad schematic sketch.
A simple pdf document, would take far less storage space, and we could at least attempt to translate it into English, to get some information.
Youtube is really great for some things, and they do have their place, but often time fall far short at trying to replace a well written technical document.
There I go getting all critical again >:-).....but then why I don't make youtube videos?..I certainly have the skill to do so....I'd just rather write good technical descriptions. :o
(the technical part)
Interesting that for all his devices, he is now trying to duplicate something that looks like SM's first tpu, which never showed any power output, only voltage. Anyone else notice that he is very careful not to move the device out of the horizontal plane, lest perhaps the leds would dim. This is suspect of a hidden source of RF powering the device at a distance, so I agree with TK.
Regarding the sketch: Is there any proof that electron cascade effect can occur outside of a vacuum environment? Such was not mentioned in the wiki link, but if possible I find it interesting.
Edit: A search for "electron cascade effect in free air" generated some interesting hits.
Quote from: ION on 2015.09.08, 20:04:25
Edit: A search for "electron cascade effect in free air" generated some interesting hits.
One of which mentions plastics. Hmmm, I've heard the term Delrin come up before. Any chance this material may be suitable for some sort of test bed?
I'm pretty certain most of us have focused on the wire and the electronics with these devices; wouldn't it be a kick in the pants to learn the insulation around the wire is the important part. Talk about hidden in plain sight.
This link certainly got my attention:
https://gmwgroup.harvard.edu/pubs/pdf/1169.pdf
And a few notes from Wesley:
https://www.youtube.com/watch?v=ghRDcgrMwp0
It certainly does look a lot like the first TPU.
Where is the source of electrons if this thing works like a magnetron?
Quote from: Matt Watts on 2015.09.09, 06:28:31
And a few notes from Wesley:
https://www.youtube.com/watch?v=ghRDcgrMwp0
Thanks Matt, the Wesley commented video clears up some aspects of the video.
Notice the resemblance of the large device to this original SM video from 1997 or thereabout.
https://www.youtube.com/watch?v=sHswoNpc0Tk
Use of magnets for startup in both devices, similar construction. Beware of the huge amount of disinformation or bad information regarding the TPU by individuals on youtube. Rely only on the original videos and your common sense.
The paper you found on contact electrification is also very interesting.
Delrin is the brand or trade name of an easily-machinable plastic, generically called "Acetal resin" or chemically, polyoxymethylene. I have used it a lot in various mechanical builds. It's strong, dimensionally stable and easy to work with machine tooling. Its position on the triboelectric scale is noted in the pdf linked above.
Imagine how trivially easy it would be for Akula (and Ruslan) to demonstrate _unequivocally_ that their self-running claims are true. Yet they do not choose to do these simple things. Meanwhile, everything they have presented can be easily "faked" by using small concealed batteries, and/or hidden wires, and/or wirelessly transmitted power.
I "conditionally" believe in Wesley's honesty and integrity. So why, if he understands so much about what Ruslan and Akula are doing, has he himself not replicated any of the allegedly self-running circuits they have talked about and that he has "explained"? If anyone can do it surely Wesley, with his many thousands of dollars worth of sophisticated lab equipment that he talks about often, could do it.
If only it were actually possible, that is. But Akula, Ruslan and Kapanadze are fishing for big fish and don't care, obviously, about providing any actual proof of their claims to little minnows like us.
Quote from: poynt99 on 2015.09.09, 12:52:12
It certainly does look a lot like the first TPU.
Where is the source of electrons if this thing works like a magnetron?
The resemblance is no doubt deliberate.
There is no evidence that this thing "works like a magnetron" and no experimental data are presented that even suggest that it does so.
The source of the power that allows it to function is conveniently not shown in the videos. All that is shown in the videos are the Red Herrings that are used as bait for the big fish.
Quote from: T-1000 on 2015.09.08, 15:48:44
To light up LED it is enough but not enought to light up 1kW+ loads where same fundamentals are applied. Just there you see high voltage source such as Tesla coil for electron emission while for LEDs it is not required.
To light up heavier loads, a different method is obviously used. Concealed wires, heavy hidden batteries, etc. Some of the concealed wires have been noticed before. And if any of those "tesla coils" actually started working as real TCs, you'd certainly notice it... the "electron emission" would cause arcing and sparking all over the place, insulation breakdown, etc. It's clear to me that the "tesla coils" shown in Akula and Ruslan videos are just there for show and aren't actually operating as TCs... and they know it too. Look at how they are handling those items!
Quote from: TinselKoala on 2015.09.09, 15:28:59
Delrin is the brand or trade name of an easily-machinable plastic, generically called "Acetal resin" or chemically, polyoxymethylene. I have used it a lot in various mechanical builds. It's strong, dimensionally stable and easy to work with machine tooling. Its position on the triboelectric scale is noted in the pdf linked above.
Imagine how trivially easy it would be for Akula (and Ruslan) to demonstrate _unequivocally_ that their self-running claims are true. Yet they do not choose to do these simple things. Meanwhile, everything they have presented can be easily "faked" by using small concealed batteries, and/or hidden wires, and/or wirelessly transmitted power.
I "conditionally" believe in Wesley's honesty and integrity. So why, if he understands so much about what Ruslan and Akula are doing, has he himself not replicated any of the allegedly self-running circuits they have talked about and that he has "explained"? If anyone can do it surely Wesley, with his many thousands of dollars worth of sophisticated lab equipment that he talks about often, could do it.
If only it were actually possible, that is. But Akula, Ruslan and Kapanadze are fishing for big fish and don't care, obviously, about providing any actual proof of their claims to little minnows like us.
Agreed
It certainly is curious that these guys chew up thousands of megabytes with almost useless youtube videos when a simple "how to do it" technical document would get the job done......that is if you really wanted to convey the operating principle, it could be done in a small .doc file or .pdf including embedded pictures and schematics.
Instead we are subject to countless videos showing tiny leds being lit to some arbitrary brightness or larger incandescent lamps that always seem to need a very fishy looking ground connection.
Whether going for the big fish or not, the small fish keep the hit counter turning some small change for the people in these poor countries, so magic shows are good for them.
@TK
QuoteTo light up heavier loads, a different method is obviously used. Concealed wires, heavy hidden batteries, etc. Some of the concealed wires have been noticed before. And if any of those "tesla coils" actually started working as real TCs, you'd certainly notice it... the "electron emission" would cause arcing and sparking all over the place, insulation breakdown, etc. It's clear to me that the "tesla coils" shown in Akula and Ruslan videos are just there for show and aren't actually operating as TCs... and they know it too. Look at how they are handling those items!
I believe that's debatable because a true Tesla Coil does not cause arcing and sparking all over the place, insulation breakdown etc.. and if it is then obviously your doing something wrong. A true Tesla Coil is supposed to produce an extreme EMF with no arcing which is essentially wasted energy. In any case my TC can produce very high voltages and a massive glow discharge seen best in a dark room however it also produces no visible discharges. Arcs and sparks are for kids...silly rabbit.
AC
Quote from: AllCanadian
Arcs and sparks are for kids...silly rabbit.
Oh, I dunno. The brilliant flashes, the loud bangs
and crackles, the faint smell of the Ozone often
spiced with Nitrous Oxide...
Almost as much fun as a pyrotechnics display.
Of course it is possible to put all of that high voltage
to work under less spectacular circumstances, but the
thrill of the "high voltage fireworks" is still captivating
even at advanced age.
Maybe some of us are just "kids" at heart who never really
ever grow up...
Quote from: Allcanadian on 2015.09.09, 17:46:19
@TK
I believe that's debatable because a true Tesla Coil does not cause arcing and sparking all over the place, insulation breakdown etc.. and if it is then obviously your doing something wrong. A true Tesla Coil is supposed to produce an extreme EMF with no arcing which is essentially wasted energy. In any case my TC can produce very high voltages and a massive glow discharge seen best in a dark room however it also produces no visible discharges. Arcs and sparks are for kids...silly rabbit.
AC
Either you are deliberately misinterpreting what I've said, or you simply don't understand what a Tesla Coil actually is or how it works. Or, considering other things you've posted... both.
You will, I presume, have no trouble overwinding your "TC" with extra windings, operating it in close proximity to solid-state logic gates, having sharp projections from the top capacity, and driving a kilowatt load with it, then. Nor will it change its behaviour or performance when it's picked up by hand. Please feel free to demonstrate these things in a video, especially the quarter-wave resonance and operation by the principle of VRSWR.
@TK
QuoteEither you are deliberately misinterpreting what I've said, or you simply don't understand what a Tesla Coil actually is or how it works. Or, considering other things you've posted... both.
I have posted pictures of my 24" TC here before.
QuoteYou will, I presume, have no trouble overwinding your "TC" with extra windings, operating it in close proximity to solid-state logic gates, having sharp projections from the top capacity, and driving a kilowatt load with it, then.
No trouble, I started using HV mosfet gate switching around a decade ago and the rest you have mentioned is common TC tech.
QuoteNor will it change its behaviour or performance when it's picked up by hand. Please feel free to demonstrate these things in a video, especially the quarter-wave resonance and operation by the principle of VRSWR.
This is why I use mosfet HV gate switching and the TC generally finds it's own point of auto-resonance, it is quite easy. In any case you can spend hours tinkering with your TC but at the end of the day it's nothing more than a toy if it has no real purpose other than arcs and sparks. Why not go and buy yourself one of those plasma globes from radio shack, lol.
Me, I use my TC as a HV AC/DC power supply for experiments and I could care less if it doesn't fit the purists view of a true TC.
AC
Here is a 500w TC running at 4Mhz
VSWR tuned of course at this frequency and power, could easily run at 1Kw :D
regards
Mike 8)
Quote from: EMdevices on 2015.09.10, 15:24:09
The internal fins of the small coil shown, resemble the structure inside a magnetron, and their size corresponds to a frequency in the GHz range, so I can believe the 1.4 GHz frequency suggested. However I think all these devices are fakes with hidden, and not so hidden, super capacitors!
The larger device that obviously tries to imitate the small TPU built by SM, has integrated circuits on the PWB, so these need at least 5 volts. There is no way you can just start such a device from a weak phenomena that builds up in voltage, it indicates he already has a voltage source hidden of at least 5 volts.
Nice try but these bring confusion to the TPU discussions.
EM
EM: resembling a cavity magnetron is not quite the same as being one, add to that there are no magnets used on the smaller device, so talk of magnetron operation in this smaller device doesn't hold water.
Mike: very nice schematic and coil.........your "Ham" knowledge is showing. Thanks for posting it.
Matt Watts said:
QuoteI'm pretty certain most of us have focused on the wire and the electronics with these devices; wouldn't it be a kick in the pants to learn the insulation around the wire is the important part. Talk about hidden in plain sight.
That got me thinking in a different direction so I took a closer look.
Now that some of us agree on the many ways the Akula devices can be faked, let's speculate and suppose just for a moment that they do work as demonstrated.....how do we guess it might be done? Is there enough ambient 1.4 gHz energy to get the "process" or at least the kickstart oscillator started by some sort of mixing or other down conversion? Is there a precharged cap for this?
Then the next step is to take a few wild guesses regarding what the
actual "process" is. Does the magnetron effect hold any water ? Would it be closer to a cyclotron effect or some variant?
Note that some of the white wire bundle of the larger device have ends twisted together, resulting in 4 connection points. With a high voltage AC or DC (shown) applied between the bundles we have a number of dielectric charges with air spaces where ions or freed electrons (from Townsend avalanche) can linger or be rotated with the application of the permanent ring magnets. Attached are some simple sketches of what may be the rough start of an approach worth examining.
Thanks Matt
centraflow, allcanadian.... you still aren't quite getting the picture.
Of course Tesla Coils can work with solid state drivers, of course SSTCs can _operate_ in the 1kW range. But most definitely NOT as Akula/Ruslan have presented them.
Centraflow, let's see your SSTC drive a 1kW bank of incandescent bulbs. AllCanadian, let's see your TC work with several extra coils overwound upon its secondary.
You don't have to try to educate me about SSTCs or regular spark-gap TCs. What you do need to explain, IF you believe in Akula/Ruslan, is how _their_ "Tesla Coils" operate within their devices as demonstrated.
http://www.youtube.com/watch?v=XeQ5WnziKBA (all solid state, CMOS, PLL, single mosfet, less than 100W)
http://www.youtube.com/watch?v=aIZClhoU2Xk (Hybrid SS-SG, operating on 24 vdc input)
Quote from: ION on 2015.09.10, 16:11:13
Now that some of us agree on the many ways the Akula devices can be faked, let's speculate and suppose just for a moment that they do work as demonstrated.....how do we guess it might be done? Is there enough ambient 1.4 gHz energy to get the "process" started by some sort of mixing or other down conversion?
This is going over two forums at once so I will just link my post from http://overunity.com/12736/kapanadze-cousin-dally-free-energy/msg460560/#msg460560
Quote from: TinselKoala on Today at 05:37:23 PM
Perhaps Wesley would like to explain how a device that produces _identical_ scope traces to Akula's "self running" unit, still doesn't manage to run itself.
The identical scope traces do not say you are under same conditions for phenomena to be met. The electronics can be doing perfectly same but the thing is not there. It is all in how power amplification happen on the wire.
Again, for power to be added to the wire we are talking about charged particles joined electron flow from the outside. If you think outside of the box the ionization/radiation/electrostatic transfer all start making sense as origin of the additional power directed on same wire. Same as two river streams joined together make big flow... My point is there, look for what make electron flow run faster/more electrons going on the wire which initially did not have it from the main power source.
P.S> There is good old read about harvesting energy from the environment - http://www.extremetech.com/extreme/148247-german-student-creates-electromagnetic-harvester-that-gathers-free-electricity-from-thin-air
In Kapanadze style device even the Tesla coil is good source for this... ;)
Quote from: T-1000 on 2015.09.10, 16:15:58
This is going over two forums at once so I will just link my post from http://overunity.com/12736/kapanadze-cousin-dally-free-energy/msg460560/#msg460560
Quote from: TinselKoala on Today at 05:37:23 PM
Perhaps Wesley would like to explain how a device that produces _identical_ scope traces to Akula's "self running" unit, still doesn't manage to run itself.
The identical scope traces do not say you are under same conditions for phenomena to be met. The electronics can be doing perfectly same but the thing is not there. It is all in how power amplification happen on the wire.
You got that right... the "proper conditions" for "self running" aren't there: My device is not supplied with a hidden source of power like Akula's is. But it can easily be so supplied. However, your explanation isn't really an explanation at all, and it is rather laughable to me, because you are essentially saying that scopetraces and other instrumental measurements are useless, they don't really tell you anything important. And I reject that.
http://www.youtube.com/watch?v=vn8kadM61jw
Quote
Again, for power to be added to the wire we are talking about charged particles joined electron flow from the outside. If you think outside of the box the ionization/radiation/electrostatic transfer all start making sense as origin of the additional power directed on same wire. Same as two river streams joined together make big flow... My point is there, look for what make electron flow run faster/more electrons going on the wire which initially did not have it from the main power source.
P.S> There is good old read about harvesting energy from the environment - http://www.extremetech.com/extreme/148247-german-student-creates-electromagnetic-harvester-that-gathers-free-electricity-from-thin-air
In Kapanadze style device even the Tesla coil is good source for this... ;)
The scope and electronics are only one part of circuit. The second and biggest part is the effects you are trying to achieve. Unfortunately most just failed there...
Quote from: ION on 2015.09.10, 16:11:13
Mike: very nice schematic and coil.........your "Ham" knowledge is showing. Thanks for posting it.
Thanks ION
I do have one of similar design, but that is not mine, mine is only around 150w with a 12v input. I was really just showing a SS TC of
exceptional power using RF and a class E output of 90+% efficiency.
I think I have made it clear in many places that I think multiple frequencies (noise) manipulated in the right way and added together will give you what you want, be it RF or hard wired, it is the heterodyning effect compounded into one output :) Re: my post on self running drum thread of SA of STEAP note the power of each frequency, and the main frequency.
Regards
Mike 8)
Quote from: Centraflow on 2015.09.10, 17:00:58
I think I have made it clear in many places that I think multiple frequencies (noise) manipulated in the right way and added together will give you what you want, be it RF or hard wired, it is the heterodyning effect compounded into one output :) Re: my post on self running drum thread of SA of STEAP note the power of each frequency.
That is the thinking into right direction. You have signal #1 directly from the power source. The signal #2 must come not from the same circuit and in most cases it can be in different energy form such as stray EM field from external transmitter or radiation source of charged particles(not limited to radioactive particles only!). Combining it into single output gives what you are looking for.
P.S> I am finding difficult to speak in same language when it comes to terms I just tried to explain. Most people are limited to closed loop only understanding and the open circuit working are just ignored..
Out there in the wild world of the Internet exist a recent presentation by Graham Gunderson. Graham is noted as being a respected power measurement engineer. The equation he uses as a basis I took a screen shot of and posted below, the middle term being the most overlooked and most important when it comes to achieving COP values near or above unity.
In his presentation, he uses high quality measuring devices, claimed to be calibrated to five significant digits. He goes on to show a COP of 1.015 from a device that should be 0.900 at best. He states the difference is the v x B term and a rotating magnetic field. As I recall, Tesla also mentioned a rotating magnetic field as being a prize discovery.
Watching Graham's presentation it isn't clear to me where this rotating magnetic field comes from, but I can speculate it might be encourage by the process of heterodyning--some other signal present that comes from the environment. This signal however started, becomes a sympathetic resonance and continues while the device is in operation.
In summary, if you take this man's presentation at face value, there is an avenue for exploration. Whether Akula is using this technique in his devices is in no way certain, but if we assume it is, it would be prudent to attempt understanding how and where it might be used.
Matt, could you kindly post a link to the presentation you speak of? I was only able to find one interview with the mystic Marikumi and it was not very impressive to my point of view. On another note a long while back I built one of Gunderson's "patented OU devices", the swiss cheese toroid and it did not do as claimed. Maybe I should have served it up with crackers and a little wine
Now a little research it seems reveals Gunderson has joined the FE circus conferences promoting books with "the secret sauce".
Quote from: ION on 2015.09.10, 21:56:38
Matt, could you kindly post a link to the presentation you speak of?
Sent you a PM.
I have seen an output from a large toroid type transformer lighting an LED with no apparent input right in front of me, on further investigation it was receiving power from a regular 50 Hz inverter working close by, that could be measured as an infinite C.O.P. because there was no input power to measure from the usual input leads. From memory I don't think I uploaded the video. Basically just wireless power and the input is no measured in totality.
I honestly believe that that bunch of scammers actually engineers those demonstrations to bamboozle folks into buying the secret sauce books. It would not be so difficult to do. I could almost do it on my own, and I do produce the effects of some other peoples claims to show it can be done if we want to show it. Madgick :P
Actually I did upload the video and it is embarrassing but the link is below, at about 5:00 mins I disconnect the power and surmize it is scope noise, haha lol. it was some time ago. Anyway I disconnect the positive input wire and the output to another battery I also disconnected but I think I left the negative of the input wires connected to the battery and an inverter was running from a different battery which transferred the noise from the inverter from its battery to the battery that was connected by the negative wire to my circuit, I measured the wave form of the noise from the inverter by attaching my scope to my circuit. Hilarious. But by the same token if it is done right a similar thing can be done and not shown, that would give the possibility of easily deceiving people with video demonstrations, or even the demonstrators could accidentally deceive themselves in a similar way to what happens in the video maybe. The power from the noise signal was ample to light a couple of LED's but I don't think I demo'd that in the video. But I lit LED's from that noise by connecting the LED's to my circuit. I'm no professor but eventually I worked out after the video. At the very end of the video I was baffled though, I didn't realize the inverter was on.
https://www.youtube.com/watch?v=i-rsdt28Ajk
.
Quote from: T-1000 on 2015.09.10, 17:13:55
That is the thinking into right direction. You have signal #1 directly from the power source. The signal #2 must come not from the same circuit and in most cases it can be in different energy form such as stray EM field from external transmitter or radiation source of charged particles(not limited to radioactive particles only!). Combining it into single output gives what you are looking for.
P.S> I am finding difficult to speak in same language when it comes to terms I just tried to explain. Most people are limited to closed loop only understanding and the open circuit working are just ignored..
That SA picture is of STEAP running with only passive components, a magnet is used to bias a reed switch inside a coil, the start voltage needed was 8v (a 12v run down LAB), the step up voltage output was over 1000v at that frequency. Apart from coils and caps, two very fast diodes are used and nothing else
As you might expect it sung like a kettle whistling at that frequency.
My idea was to let it find it's own natural frequency at which to switch, of course the coils and caps would influence this input frequency.
The start battery with feed back would show HV using a neon on one of the terminals, the reason why in time it would fail I think, also why supper caps could not be used, apart from caps have a linear voltage fall off from start, which is not good for self running unless there is a large COP to keep a flat line working voltage in the cap.
If anyone wants the circuit to play with, e-mail me
regards
Mike 8)
@All
This new design is really compact and functional, I hope that sooner or later the demonstrations can be made by anyone.
https://m.youtube.com/watch?v=v1zjkc5dYZU
https://m.youtube.com/watch?v=X-P0XPbtsnE
https://m.youtube.com/watch?v=OwKe6jkTY3o
Yes, Ruslan understands something fundamental we are still missing. I have some ideas, but far from proven.
Quote from: Matt Watts on 2015.11.20, 17:52:28
Yes, Ruslan understands something fundamental we are still missing. I have some ideas, but far from proven.
Me too. Not proven. :(
Quote from: Matt Watts on 2015.11.20, 17:52:28
Yes, Ruslan understands something fundamental we are still missing. I have some ideas, but far from proven.
In
this video (http://youtu.be/WRtyHAsFtFA?t=19m11s) he says that the width of the nanopulse determines the frequency of the push-pull driver for the yoke transformer.
It is a 40
th subharmonic of the nanopulser, ...or something like this.
As far as I know, none of the replicators tired to tune the push-pull driver to a subharmonic of the nanopulser's frequency.
P.S.
In this video (http://youtu.be/WRtyHAsFtFA?t=18m39s) Ruslan also says that 1kV and nanosecond pulses are impossible to achieve by the DSRD nanopulser, but Itsu's replication (https://www.youtube.com/watch?v=sCF8fwwqhKY) clearly shows otherwise.
Quote from: Matt Watts on 2015.09.10, 17:56:26
Out there in the wild world of the Internet exist a recent presentation by Graham Gunderson. Graham is noted as being a respected power measurement engineer. The equation he uses as a basis I took a screen shot of and posted below, the middle term being the most overlooked and most important when it comes to achieving COP values near or above unity.
In his presentation, he uses high quality measuring devices, claimed to be calibrated to five significant digits. He goes on to show a COP of 1.015 from a device that should be 0.900 at best. He states the difference is the v x B term and a rotating magnetic field. As I recall, Tesla also mentioned a rotating magnetic field as being a prize discovery.
Watching Graham's presentation it isn't clear to me where this rotating magnetic field comes from, but I can speculate it might be encourage by the process of heterodyning--some other signal present that comes from the environment. This signal however started, becomes a sympathetic resonance and continues while the device is in operation.
In summary, if you take this man's presentation at face value, there is an avenue for exploration. Whether Akula is using this technique in his devices is in no way certain, but if we assume it is, it would be prudent to attempt understanding how and where it might be used.
I have known Graham for many years, firstly on Dave Squires oubuilders forum, later working for MPI then later still for Chava. Never met him in person but spoken and corresponded many times and even video conferences. I can vouch for his integrity and I don't blame him from trying to make some money from all the work he has put in over the years.
Now about his transformer. Here is a photo taken from the Energetic Forum. My take on this is he uses two input coils, they are the top and bottom ones. The output coil is the middle one. He drives the two input coils separately at the same frequency but with a phase shift which is probably 90 degrees. The net result is that you don't get flux continuous around the core, where fields try to cancel flux gets thrown out radially (conservation of flux sees to that, you can't have flux lines that are not continuous). At that point the flux along the core is null. And because it is all cyclic, and with that 90 degree phase shift, the null point is not fixed, it moves. That is the moving field. If you imagine the null point moving around the closed core over one cycle you can consider that as a rotation, but note it is not the classical rotating field you would get from a rotating magnet. At that moving null the radial leakage field traverses any coil wound around the core so as to induce the classical E = v X B motional induction as a voltage in the coil. The output coil therefore sees two null points passing it each cycle, one with radial flux outward and one with radial flux inward. The motional induction is therefore AC supposedly superimposed on the classical dPhi/dt transformer induction.
I have always had trouble with this explanation because the two induction terms have a common source and you can easily verify that either term can give you the measured voltage. The radial field is maximum at this null point so both the v X B and the dPhi/dt yield the same result, a maximum induction as the null passes through the coil. Not only that but for the balanced system the two values are the same. So IMO they are just two different ways of calculating the same result. But that doesn't take into account the effect of the load current on the fields in the core. That load current has an effect on phasing and it could be this that unbalances the system in such a way that v X B effect takes slight preference over the dPhi/dt effect.
It may be of interest to note that Myron Evans ECE theory as portrayed by Horst Eckardt mentions rotating magnetic fields and his visualisation of SCR (spin connection resonance) is a rotation of the plane of the A field around a magnetic core, see image. That form of rotation is exactly what Graham achieves in his core.
Smudge
@ All
Happy New Year !!!
This is another replication working.... I hope someone help me to make the same result easily..... with all instruction...
Investors still doesn't come......
https://youtu.be/Rrl1FS9QwyM
https://youtu.be/v8_jNNvBAFE
It is very efficient, but what they don't tell you is the circuit only runs for a few minutes. I want to see a video of a circuit that really taps into zero point energy and doesn't require a jump start.
Happy New Year!!
Barefoot Free Energy.... from someone calling himself "Oxymoron" !!
Methinks someone is pulling your leg.
https://m.youtube.com/watch?feature=youtu.be&v=YTaK6CAOOms
News form Nelson Rocha