OverUnity Research
  
Welcome, Guest. Please login or register.
+  OverUnity Research
|-+  Electrical / Electronic Devices
| |-+  Tariel Kapanadze's Devices
| | |-+  Akula0083 Lantern v3
Pages: 1 2 3 [4]
Topic: Akula0083 Lantern v3  (Read 7690 times) Print
Grumage

Position: Sr. Member
****
Posts: 315
« Reply #75 on: 2014-05-11, 14:44:17 »
Dear All.

Have a look at this !!  Afro

https://www.youtube.com/watch?v=6B79UJGoNJE

Simplicity !!

Cheers Grum.
 IP: [ Logged ] WWW
EMdevices
Position: Hero Member
*****
Posts: 1056
« Reply #76 on: 2014-05-11, 15:30:01 »
Nice!  

with a 100 % efficient circuit a 3000 uF cap charged to 9 volts should last quite a few hours.

Let's see,   Energy in cap = 1/2 C V^2 = 0.12 J

Who can estimate how much average power those pulsing LEDs consume?  1 uW?  maybe more?

EM

 IP: [ Logged ]
ION
Group: Elite
******
Posts: 2009

It's turtles all the way down

« Reply #77 on: 2014-05-11, 15:52:49 »
Nice work Grum.

John Blade has an interesting build of the dual TL494 circuit that is exhibiting the long sought after "reverse ringing" or growth of a oscillating waveform, one possibility: energy is somehow being added.

http://www.overunity.com/14610/akula0083-light-no3-dual-tl494/msg402033/#msg402033

Watch the video on that post.

http://youtu.be/8_Bal_7F06s
 IP: [ Logged ]
d3x0r
Position: Newbie
*
Posts: 21
« Reply #78 on: 2014-05-11, 18:22:28 »
Nice work Grum.

John Blade has an interesting build of the dual TL494 circuit that is exhibiting the long sought after "reverse ringing" or growth of a oscillating waveform, one possibility: energy is somehow being added.

http://www.overunity.com/14610/akula0083-light-no3-dual-tl494/msg402033/#msg402033

Watch the video on that post.

http://youtu.be/8_Bal_7F06s
the higher frequency ringing is at 582Khz, so every time the faster TL494 fiires, it adds to the signal; not a somehow
 IP: [ Logged ]
MenofFather
Position: Newbie
*
Posts: 21
« Reply #79 on: 2014-05-12, 08:47:32 »
Nice!  

with a 100 % efficient circuit a 3000 uF cap charged to 9 volts should last quite a few hours.

Let's see,   Energy in cap = 1/2 C V^2 = 0.12 J

Who can estimate how much average power those pulsing LEDs consume?  1 uW?  maybe more?

EM


One led can consump 2 volts and 0,1-5 miliamperes, so 0,2-10 miliwatt.
 IP: [ Logged ]
EMdevices
Position: Hero Member
*****
Posts: 1056
« Reply #80 on: 2014-05-12, 17:53:46 »
thanks MenofFather,  that's what I thought.    So  if the LED is pulsed, we can expect an LED to consume power in the uW range depending on the pulse width and duty.    I'm pretty sure my joulethief circuit long ago was drawing around a few uWatts.

If this guy can get the cirucit to operate for a few hours if not a day, it may be overunity, and that would be nice.   I'm interested to see the voltage drop in the 3000 uF capacitor as a function of time.
 IP: [ Logged ]
lost_bro
Position: Jr. Member
**
Posts: 56
« Reply #81 on: 2014-05-12, 19:19:40 »
Good day all:

He mentioned he switched to a 2n1304 germanium trans.   '(S.J.L) lasersaber.'
ICmax=300ma
Vbe    =.22v

So apparently needed a more efficient switching device to get it up and running.

attached 2n1304 datasheet.

take care, peace
lost_bro
-------------------------------
* datasheet 2n1304 germanium.pdf (271.59 KB - downloaded 48 times.)
 IP: [ Logged ]
MenofFather
Position: Newbie
*
Posts: 21
« Reply #82 on: 2014-05-16, 14:49:23 »
thanks MenofFather,  that's what I thought.    So  if the LED is pulsed, we can expect an LED to consume power in the uW range depending on the pulse width and duty.    I'm pretty sure my joulethief circuit long ago was drawing around a few uWatts.

I mean in not nominal brightness that consumps (because in Lasersaber seems in not nomianal LEDs shine). If in nominal brightness, then one LED consumps 20-50 miliamperes, you can look datasheet. Smiley

If speek abou this lanter with two TL494, then transistors must be fast, open must around not more than 50 nanosecond and close around not more than 50 ns. If then not be that fast, then this divice not works how selfrunner. In Akula ferite feroresonance is about 3.5 megaherc. So feroresonance is qiute hight frenquency, so transistors must be like I say, fast closing and opening, not all can close with 50 ns time.
 IP: [ Logged ]
stprue

Position: Full Member
***
Posts: 214
« Reply #83 on: 2014-05-16, 17:24:14 »
LS now has an updated unit with a different transistor.  Fantastic run time!

http://laserhacker.com/?p=410

 IP: [ Logged ]
TutorialFE

Position: Jr. Member
**
Posts: 57
« Reply #84 on: 2014-05-19, 21:40:21 »
A new record.....


http://www.youtube.com/watch?v=T5EXisVd9Xs

new replicators

https://www.youtube.com/watch?v=QZhebKXdJNM
https://www.youtube.com/watch?v=MQXGJgeGbh8
 IP: [ Logged ]
TutorialFE

Position: Jr. Member
**
Posts: 57
« Reply #85 on: 2014-05-21, 16:28:36 »
Other replicators....


http://www.youtube.com/watch?v=Cz-i8JoY8nU

http://www.youtube.com/watch?v=EjJ1iVNHd4s

http://www.youtube.com/watch?v=t3UJ6tVWrmg

https://www.youtube.com/watch?v=rdWoat2Zjoo

https://www.youtube.com/watch?v=8QbRiR5xsSc
Last Edit: 2014-05-26, 14:32:32 by TutorialFE
 IP: [ Logged ]
Itsu

Position: Sr. Member
****
Posts: 404
« Reply #86 on: 2014-06-07, 21:45:46 »

Finally got around to put together this dual tl494 akula device on a breadboard, at least the first part without the coils and load.

I used the updated handdrawn diagram which you can partly see in the picture.
I had to modify some things to get to the mentioned 16.6KHz on the lower B4 tl494 by adding a 10K resistor between pin 6 and the 20K pot.
Also the 180 Ohm resistor at pins 9/10 of this B4 chip needed to be upped to 1K as it got hot (i also used a 1K on the upper chip H4 pin 10).
Transistors used (PNP) are 2n5401 and the MOSFETs are IRFU320.
Nothing attached to the drains of the MOSFETs yet other then a 10K resistor to + (17V here).

H4 chip is set to 14KHz   / 44% (max.) duty cycle as per the diagram.
B4 chip is set to 16.6KHz / 89% (max.) duty cycle as per the diagram.

See scopeshot 1

The drains output is shown in screenshot 2 and shows the reverse of screenshot 1, (same frequency, but reversed duty cycles).

Next is to wind the coils and attach the leds

Regards Itsu
-------------------------------
dualtl494.png
Akula0083 Lantern v3
* dualtl494.png (595.79 KB, 640x480 - viewed 374 times.)
outputs.png
Akula0083 Lantern v3
* outputs.png (23.66 KB, 640x600 - viewed 341 times.)
drains.png
Akula0083 Lantern v3
* drains.png (23.41 KB, 640x600 - viewed 336 times.)
 IP: [ Logged ]
Itsu

Position: Sr. Member
****
Posts: 404
« Reply #87 on: 2014-06-08, 10:39:30 »

Ok, as a side note concerning free energy;

today (8 June) it is exactly 1 year ago that our solar panel installation was started up.
This 6Kw (24 panels of 250W) has delivered 5912Kwh = almost 6Mwh of energy during that year.

As we used 3768Kwh ourselves we delivered/sold the rest to the net.

Record day was on 8 June last year (first day in operation) with a yield of 44Kwh.

My wife already uses this to charge the battery of her electrical bike, so she goes green, but there is enough
room to charge a little electrical/hybrid car which will be our next goal.

One pitfall discovered up till now is that during operation on a hot day >25°C the panels get so hot (up till 70°C)
that the efficiency drops with almost a third, so it won't deliver more then about 4.3Kw on a hot sunny day compared
to peaking 6.2Kw on a cooler cloudy day (when the clouds move away that is).

Regards Itsu
-------------------------------
solar1.png
Akula0083 Lantern v3
* solar1.png (441.21 KB, 1012x654 - viewed 321 times.)
 IP: [ Logged ]
Peterae

Group: Administrator
*****
Posts: 2888
« Reply #88 on: 2014-06-08, 11:43:58 »
That's quiet interesting Itsu regarding the performance, thanks for the insight  Afro
Have you an idea how long to pay back the cost of installation

Thanks
Peter
 IP: [ Logged ] WWW
lost_bro
Position: Jr. Member
**
Posts: 56
« Reply #89 on: 2014-06-08, 15:37:53 »
Ok, as a side note concerning free energy;

today (8 June) it is exactly 1 year ago that our solar panel installation was started up.
This 6Kw (24 panels of 250W) has delivered 5912Kwh = almost 6Mwh of energy during that year.

As we used 3768Kwh ourselves we delivered/sold the rest to the net.

Record day was on 8 June last year (first day in operation) with a yield of 44Kwh.

My wife already uses this to charge the battery of her electrical bike, so she goes green, but there is enough
room to charge a little electrical/hybrid car which will be our next goal.

One pitfall discovered up till now is that during operation on a hot day >25°C the panels get so hot (up till 70°C)
that the efficiency drops with almost a third, so it won't deliver more then about 4.3Kw on a hot sunny day compared
to peaking 6.2Kw on a cooler cloudy day (when the clouds move away that is).

Regards Itsu

Good day Itsu

Great to hear of your success with the solar installation.  I run a small company here in Mexico, we install 'Grid tie' Solar systems......  Is your system's topology 'String' inverter or 'Micro' inverter?
 I would assume you are located in the Northern latitudes;  It is good to see that some countries are actively promoting grid 'interconnection' type installations (paying for excess generation).
 Mexico does NOT pay for excess production, only gives an energy credit good for up to one year.  After that amount of time, if you don't use your produced excess, it becomes government property...... What a scam....
But you are correct, solar is the closest thing to 'Free' energy that I have ever used!!!!

take care, peace
lost_bro
 IP: [ Logged ]
Itsu

Position: Sr. Member
****
Posts: 404
« Reply #90 on: 2014-06-08, 16:22:12 »

Peter,

up till now we saved about 1020 Euro in this year (went from 90 euro a month to 50 Euro in July 2013 till febr. this year, then got refunded 490 Euro,
then a further reduction to 40 euro / 2 months now which is all transportation costs).

So with an investment of about 9000 Euro it will be around 8.5 till 9 years, but the feeling of being partly green is nice.





Lost_bro,

its a 'String' inverter topology still, it has 3 strings.
I am located in the Netherlands and together with Germany there is a strong push to promote this kind of green energy.

From my colleagues in Italy, Spain and Croatia i learn that they get taxed or even fined for putting up such a solar installation and/or that
an installation half like mine costs the double of what i have paid  Cheesy
Real crazy considering the double amount of sunhours they got overthere.


Is there something to do against this efficiency drop when the panels get hot?
I mean i see some d.i.y. solutions like cooling ribs at the backside of the panels or air ducting or both.
Perhaps you can push some manufacturers to think about that as i see it as a loss.

 
Thanks,  regards Itsu
 
 IP: [ Logged ]
Peterae

Group: Administrator
*****
Posts: 2888
« Reply #91 on: 2014-06-08, 20:44:42 »
Thanks Itsu, that's pretty good then, we also get subsidized here in the UK, it was a really good deal when it started about 5-7 years ago, but they have recently dropped the subsidy, but just the fact you have no bill is great, i would love to build an electric car, if i had the spare cash i would definitely be tinkering, i have had an electric bike for must be 5 years now, great fun.
 IP: [ Logged ] WWW
Itsu

Position: Sr. Member
****
Posts: 404
« Reply #92 on: 2014-06-14, 22:10:46 »
Concerning the dual tl494 device, as promised, i have made a few U-core transformers and was tuning the last few days.
But its like ION said before, we have 5 pot meters and no idea where to aim for, so its like shooting in the dark.

The present big U-core i use has secondary 30 turns and primary 22 / 11

I tried severall different MOSFETs, presently using big IRFP264's but even those can get very hot.
Voltage across the leds plus resistor runs up till 40V, so needed to increase the current limiter resistor from 10 Ohm till 470 ohm now, while also putting 4 LEDs in series (3V leds).

Input is 9V with current limited to about 2A from my bench PS (a 9V battery won't last very long).

I have these 12V transil diodes, but they get very hot when putting one across the upper H4 = Q1 drain to ground.
Not sure if i can put 2 in series to up the protection to 24V,  anyone?

I have a 140V varistor across the lower B4 = Q2 drain to ground

Anyway,  no abnormalities seen yet.

Video here:    https://www.youtube.com/watch?v=Ojsj_JMH1J0&feature=youtu.be



Regards Itsu
Last Edit: 2014-06-15, 00:18:28 by Itsu
 IP: [ Logged ]
d3x0r
Position: Newbie
*
Posts: 21
« Reply #93 on: 2014-06-15, 08:52:18 »
One or more of the coils may be reversed since there is no start/end indicator; and even if there were, it's quite possible it was just indicating 'a connection to the coil'

Akula's first 30kW thing; or whatever, which was 2 tesla coils was a mixer of 1 and 3x frequencies... but; what is the MO of this circuit?  At least the version 4 with a choke I can vaguely see some sort of intent...

is the intent somehow to use one coil to drive power for the LEDs and the other to restore power for the first?  (which may require additional diode things)

I can't see them working in parallel; it would just be a loss then, unless 2 signals striking at the same time moves x^2 current or is some other exponential; but then I'd think they'd be 100% in-sync instead of variable...

if a short duty cycle is used on the slower one (which drives the load) then a higher duty cycle one can be used in the blank spots; hadn't quite expected to see quite so square of signals....  I just made a 1.7H coil (1700000uH... driver on it is 176uH; with low current signal generator can build a resonance of 100x the input voltage...) but this coil is very sinusoidal with 3V drive...

My other issue is; I can't really find anything that has 2 drivers on the same core except a conventional 'resonant transformer' which has a primary and 2 secondaries, 1 secondary bound to a cap to provide an LC tank that can cover gaps in primary power... because all other coils ...(I guess a push-pull uses 2 coils, but they are never driven at the same time) get affected by the primary's change in current so opening a gate on another coil at the same time is really only draining the power from the coil uselessly...

if 2 coils do conduct power to ground at the same time... then they should just be parallel inductances?  which is less than the lowest inductance; which leaves less power for the pickup coil...

and if one is reversed... then it would be 2 coils in series; which would be a bifilar with ends shorted; which is a square increase in inductance (if both were 1x, 2 of them in series is 4x inductance... )  but then again; why not just us a bifilar and 1 drive?
 IP: [ Logged ]
Itsu

Position: Sr. Member
****
Posts: 404
« Reply #94 on: 2014-06-15, 10:45:02 »
One or more of the coils may be reversed since there is no start/end indicator; and even if there were, it's quite possible it was just indicating 'a connection to the coil'............



Thanks d3x0r,

the handdrawn diagram that i use does have the start/end indicators, and i used the video in which Akule dissected this coil to copy the configuration.
But indeed, there could be a (deliberate) mix-up, but i did swap over the secondary leads without any difference noted.
I could swap over the primary coils too.

I will also measure the coils inductance's, both connected parallel (primary) and when disconnected at the common side.

Regards Itsu

 
 IP: [ Logged ]
d3x0r
Position: Newbie
*
Posts: 21
« Reply #95 on: 2014-06-15, 16:03:52 »

if a short duty cycle is used on the slower one (which drives the load) then a higher duty cycle one can be used in the blank spots; hadn't quite expected to see quite so square of signals....  I just made a 1.7H coil (1700000uH... driver on it is 176uH; with low current signal generator can build a resonance of 100x the input voltage...) but this coil is very sinusoidal with 3V drive...

higher duty cycle actually means means higher frequency
 IP: [ Logged ]
« previous next »
Print
Pages: 1 2 3 [4]  
« previous next »
Jump to:  


Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!
Forum and Banner by OUR Team