PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2024-03-29, 04:42:28
News: Registration with the OUR forum is by admin approval.

Pages: 1 ... 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 [22] 23
Author Topic: Graham Gunderson Energy conference High COP demonstration  (Read 211988 times)

Group: Elite Experimentalist
Hero Member
*****

Posts: 4056

When loading the secondary (FWBR with capacitor and 100 Ohm resistor), the primary signal is still lost.
No more resonance at whatever frequency and with whatever parallel capacitor.

Only way to show some signal on the primary in resonance is with an unloaded secondary.

I moved the 7H choke to be in the 12V supply line to the H Bridge and now (unloaded secondary still) the voltage across the primary looks somewhat better,
see screenshot 1

But when measuring the 12V AFTER the choke, i see a very wavy / spikey signal see screenshot 2, so i think the primary signal mostly comes from this
wavy input voltage, and not from resonance.
As ION mentioned, my 220VAC wall adapter choke sucks
 
And like verpies said, "A full H-bridge is a tricky beast on the MOSFET/iso driver level"


Itsu
   
Group: Experimentalist
Hero Member
*****

Posts: 1665
When loading the secondary (FWBR with capacitor and 100 Ohm resistor), the primary signal is still lost.
No more resonance at whatever frequency and with whatever parallel capacitor.

Only way to show some signal on the primary in resonance is with an unloaded secondary.

I moved the 7H choke to be in the 12V supply line to the H Bridge and now (unloaded secondary still) the voltage across the primary looks somewhat better,
see screenshot 1

But when measuring the 12V AFTER the choke, i see a very wavy / spikey signal see screenshot 2, so i think the primary signal mostly comes from this
wavy input voltage, and not from resonance.
As ION mentioned, my 220VAC wall adapter choke sucks
 
And like verpies said, "A full H-bridge is a tricky beast on the MOSFET/iso driver level"


Itsu

Itsu,

I was going to simulate your setup but something does not compute in regards to your transformer measurements. Using your values of primary inductance = 2.3H and primary inductance with secondary shorted = 2.3mh and using the formula for coupling K = (1-Lleak/Lpri)^1/2 = .9995. This value of coupling is way too high for the physical transformer you show in your video IMO. The K factor for your transformer should be in the .8 to .9 range give or take.

Even though I have the equipment to measure the inductance, I prefer to set up a test fixture and apply a switched voltage to a winding and measure the di/dt and calculate the inductance from the results. With this, one can set the peak current to the actual operating level and check for any saturation, etc.

pm

   
Group: Experimentalist
Hero Member
*****

Posts: 1665
ION,

OK, here is your sim that I've modified by using the IR2184 sub circuit downloaded from the Infineon website rather than the model by Analog Spiceman. I then used the auto generating feature to create the symbol from the subcircuit. I also changed to the alternate "solver" in Tools>Control Panel>Spice with all else unchanged. This was run in LtSpice XVII.

pm
   
Group: Elite
Hero Member
******

Posts: 3537
It's turtles all the way down
ION,

OK, here is your sim that I've modified by using the IR2184 sub circuit downloaded from the Infineon website rather than the model by Analog Spiceman. I then used the auto generating feature to create the symbol from the subcircuit. I also changed to the alternate "solver" in Tools>Control Panel>Spice with all else unchanged. This was run in LtSpice XVII.

pm

Dear pm

Thanks for taking a look at it and the helpful hints. I tried the alternate solver, but that didn't help. I have not yet put the new parts in the library, but I am running the LTSpiceIV as I'm stuck with WinXP, so maybe that's also part of the problem. If you post the .asc, I'll see if it runs ok here, with the appropriate library changes of course.

Regards
ION



---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   
Group: Experimentalist
Hero Member
*****

Posts: 1665
Dear pm

Thanks for taking a look at it and the helpful hints. I tried the alternate solver, but that didn't help. I have not yet put the new parts in the library, but I am running the LTSpiceIV as I'm stuck with WinXP, so maybe that's also part of the problem. If you post the .asc, I'll see if it runs ok here, with the appropriate library changes of course.

Regards
ION

ION,

I sent the info to you by pm so let me know if it doesn't work for you.

pm
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4056
I changed to another transformer, its a MOT (Microwave Oven Transformer), 220V in, 2000V out.
Primary measures 220mH @ 1KHz, and when secondary shorted it drops to 53mH @ 1Khz.

With a 600nF capacitor parallel to the primary and the secondary loaded, i get resonance around 520Hz.

I still have my 220VAC transformer choke (7H) on the DC  input side (12V).

When loading the secondary with a FWBR (4x UF4007) into a 910mF capcitor with a 100 Ohm resistor across, i still have some resonance signals
which now can be manipulated into the desired 30% positive, 30% negative, 30% flat voltage trace, see screenshot 1

The 2 input signals used can be seen in screenshot 2

For ION about his IR2184 "forbidden state" question, i think it does not matter as the load is between 2 H-bridges which do not communicate. 

Regards Itsu
   
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 1567
Frequency equals matter...


Buy me a drink
JDO300's controller permitted <50ns on off ramp speeds. I guess I am confused about the thread target here.


---------------------------
   
Group: Elite
Hero Member
******

Posts: 3537
It's turtles all the way down
Itsu said:

Quote
For ION about his IR2184 "forbidden state" question, i think it does not matter as the load is between 2 H-bridges which do not communicate.

I am basing what I said with regards to partzman's earlier comment that all four  FET's are to be on in the "clamp phase"

I don't think it is possible to have all four on with the IR2184

You can arrange the timing so that an upper pair or lower pair are on at the same time, but not all four due to the nature of the (non-negotiable) inversion of the LO with respect to HO. You can also have all four off at the same time by pulling both SD's low.

I assume you still have a transistor inverter for the right side chip for the marked up drawing.
Regards, ION


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   
Group: Experimentalist
Hero Member
*****

Posts: 1665
Itsu said:

I am basing what I said with regards to partzman's earlier comment that all four  FET's are to be on in the "clamp phase"

I don't think it is possible to have all four on with the IR2184

You can arrange the timing so that an upper pair or lower pair are on at the same time, but not all four due to the nature of the (non-negotiable) inversion of the LO with respect to HO. You can also have all four off at the same time by pulling both SD's low.

I assume you still have a transistor inverter for the right side chip for the marked up drawing.
Regards, ION

Yes, if only the upper pair or lower pair of mosfets in the bridge are on during the clamp phase, the current in the primary will be frozen but the input choke current will be dumped during this phase. The choke current can also be clamped however using diodes as shown in the partial schematic below.

pm
   
Group: Elite
Hero Member
******

Posts: 3537
It's turtles all the way down
Yes, if only the upper pair or lower pair of mosfets in the bridge are on during the clamp phase, the current in the primary will be frozen but the input choke current will be dumped during this phase. The choke current can also be clamped however using diodes as shown in the partial schematic below.

pm

Very clever pm. Your experience is showing.

Regards
ION


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   
Group: Guest
Gentleman,

You can not turn on all 4 transistors, clamp, etc.  for 1/3 cycle.  That's a short on the power supply!!!!
The clamping/load connected is most likely in the secondary that you see reflected in the primary.

Ben
   
Sr. Member
****

Posts: 460
How about crazy idea about disconnecting power supply for that split second moment you want to dump energy.Just thinking, sorry I'm not following this thread carefully
   
Group: Experimentalist
Hero Member
*****

Posts: 1665
Gentleman,

You can not turn on all 4 transistors, clamp, etc.  for 1/3 cycle.  That's a short on the power supply!!!!
The clamping/load connected is most likely in the secondary that you see reflected in the primary.

Ben

Ben,

You are correct if one does not use a choke between the DC supply and the H bridge however, with the choke inserted and positioned as shown in the schematic below, one may turn on all the mosfets in the bridge to freeze the flux in the primary winding as well as the flux in the inductor during the clamp phase which satisfies Graham's original disclosure IMO.

From the schematic we can see that the choke inductor limits the current drawn from the supply during the clamp phase of each cycle.

pm
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4056
Quote
I assume you still have a transistor inverter for the right side chip for the marked up drawing.

Correct ION.


I added the current signal from the primary, see green trace (current controller was set to 2mA/Div.).
This signal is  very close in shape to the original GG current signal (in blue there).

EDIT, 2th screenshot shows the math function yellow x green in red.


Itsu
   
Group: Experimentalist
Hero Member
*****

Posts: 1665
Correct ION.


I added the current signal from the primary, see green trace (current controller was set to 2mA/Div.).
This signal is  very close in shape to the original GG current signal (in blue there).

EDIT, 2th screenshot shows the math function yellow x green in red.


Itsu

Itsu,

Thanks for posting the scope shots. Do you now have the choke inside or outside the H bridge?

pm
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4056

PM,

since post #525  its outside the H-bridge.

Itsu
   

Group: Professor
Hero Member
*****

Posts: 3354
since post #525  its outside the H-bridge.
So the choke constitutes a constant current power supply for the H-bridge ...at least for a while.
   
Group: Guest
Dear All,

Graham Gunderson does have two technical videos placed on YouTube. One is 10 minutes long while the other is 20 minutes long. These were filmed just prior to the convention and apparently deal with additional technical details. Now that his password to this forum site is working again he says that he will release these films to the public. Originally he wanted to make them private because of the information they contained, however he would have to provide YouTube with an email address for each person that wanted to look at them. I don't know specifically what they are about.

But, don't expect anything soon. Graham says that he will release these films on this forum after he writes up a suitable introduction with background information.

The real reason his response is going to be slow is that he has contracted for some very lucrative side work that pays many times more than his day job wage. So his evenings are now spent making the big bucks to help pay the tax man off. He is rebuilding a piece of aircraft test instrumentation from the 50's that is irreplaceable.

On the technical front Graham talked about another mode of operation that he discovered while getting ready for the convention. This mode was just screaming with at least 2K VAR of reactive energy oscillating in his tank circuit. He talked about how the current through the choke changed. It was in this mode that a slight rotation at an arms' distance of a couple of PM's would greatly impact the operation of the circuit. He didn't explore this mode for very long since the convention was on the horizon, but he wants to get back to it and is certain that the present H-Bridge is not up to the challenge.

Graham is not happy with the performance of his present H-Bridge. I suspect that it has stability issues and was responsible for toasting the last set of MOSFETS.  It is like he wants to disown that sub-circuit, despite all the hours spent during its construction. That was his old love. His new love is some new fly-back controller chip that has provisions for the regulation of up to three output voltages and somehow "looks through the transformer" to get feedback information. I don't know how this is done, nor do I know what the chip is at the moment.  Unfortunately there is no commercial transformer that will fill the design bill. He has to design and hand wind at least four (4) of these to make things work. Thus the need for the transformer winder machine. Hopefully the machine that is available can make transformers large enough the do the job - To be determined. The Jovil machine can only wind torrids up to 6mm in diameter with three conductors of #30 AWG wire.

Graham has been following the posts on the forums, but didn't have access to OUR until recently.  He says he has already gone round and round in the past with fruitless discussions of "gate leakage" and "skew parameters", which are important issues but subjects that he doesn't wish to defend and comment on again.

He really thinks that his presentation was a global flop. I agree, that being as it may, he thought that if this technology had been available 5 years ago many people would be all over it, that hasn't happened. I suspect that most all the real builders have passed on, become burnt out, lost the dream, or have moved into assisted living facilities. Anyway he feels a real sense of disappointment in the reception of these new ideas, perhaps he got his first royalty check for the DVD and didn't make enough to buy both his sons a value meal at MacDonald's. So he feels no urgent need to move ahead with developments since in his mind hardly anybody cares. He appreciates the efforts and comments of those on this forum, but the six or so interested people are far less that what he expected.

I told him about the 577 views of some of the photos on overunity.com. We both wondered why so many lookers and so few do-ers these days. Then again this is not an easy circuit to replicate compared to a Bedini School Girl Motor.

Obviously nobody has approached him to discuss an outright purchase of the technology or to offer grant funding.

He also believes that there are probably 50 ways that this approach could be utilized for OU, such that a patent on this particular method is not all that valuable (thus no buyers). There may well be several people looking at these documents with the express intent of developing a patentable variation in other countries. Well more power to them. Perhaps in the past there was more collaboration than what we see now. I'm sure the 2009 changes to the US patent laws haven’t helped.

Graham seems ambivalent about the Open Source issue. He has the impression that people would expect complete and detailed plans with an up to date parts source list, and then complain about unresolved construction issues and forgotten notes. He says that nobody would follow the plans anyway. Why spend the hours compiling all this data for a setup that he doesn't like for a very small group of unimpressed viewers.  He says that the real talent, if they see any merit in what has been done to date, will run with this and make the necessary improvements as they go.

Otherwise he is going to continue to move at his own pace and take advantage of the extra income while it is available.

Make no mistake he is still very interested in this phenomenon and is continuing the exploration of the fundamental physics, which he has no idea about.

Meanwhile the rest of us, especially me, are still playing catchup.

Spokane1

PS: Does anybody want me to re-post the photos of the H-Bridge and coil winder that I posted on overunity.com while I was having access issues?

PS: Graham did say that the intent of the choke was to make the inductor act like a current source. He didn't know what a "Swinging Choke" was (Apparently little exposure to Ham Radio). He said that he custom built the one he has to reduce the inductance so that it would store more energy and not saturate as fast.
   

Group: Experimentalist
Hero Member
*****

Posts: 755
Believing in something false doesn't make it true.
He does not have to make his videos public if he doesn't want everyone to see them.  And he doesn't have to make them private and go through that hassle either.  All he has to do is make them unlisted.  Then if he wants to share them he can privately share the link to the video.  If it is unlisted then it will not show up on any searches.  He won't even be able to find it unless he uses his video manager to find it or saves the link for his use.  I have done this several times and it is very easy to do.  Just select unlisted instead of private or public.

Carroll


---------------------------
Just because it is on YouTube does not make it real.
   

Group: Experimentalist
Hero Member
*****

Posts: 2809


Buy me a beer
I follow this thread with great interest, I think OU is all about being able to change VAR's into real power without affecting the input power draw, we can produce hugh VAR's very easly, it is this converting circuit where the interest is.

I hope the few here will keep at it, I am short on time at the moment with the methane production unit etc

Regards

Mike 8)


---------------------------
"All truth passes through three stages. First, it is ridiculed, second it is violently opposed, and third, it is accepted as self-evident."
Arthur Schopenhauer, Philosopher, 1788-1860

As a general rule, the most successful person in life is the person that has the best information.
   

Group: Professor
Hero Member
*****

Posts: 1828
While you guys are looking at circuits I am concentrating on the magnetics, examining what could be going on in the magnetic domain.  Graham know this.  I keep reaching Eureka moments where I think I have discovered the secret, then on further study find the flaw in my arguments.  I am convinced that the very short interrupt time plays an important part, so I look at the magnetic time-constants.  One possibility applies to the mylar filled gaps which may not have constant dimension thus introducing a variable that is subject to inertial mass, which is a slow response that would be unable to follow that short interrupt time.  Another is the frequency response of the ferrites where the high permeability core is likely to have a lower cut-off then the low permeability one.  I tend to concentrate on the effect of a variable gap where a tiny change of dimension can create a significant change in reluctance.  If that gap has constant flux its changing reluctance is simply a changing mmf drop across it, and the energy associated with that is the so called co-energy on the B v. H  or Flux v. current chart.  That is energy change that does not involve voltage, it is invisible electrically.  And as Aspden pointed out, if you have a saturated core the co-energy to the right of the BH curve (invisible) is greater then the normal energy to the left (visible electrically).  So a combination of saturated core and variable gap has some potential for unusual performance.  You'll be the first to know if I hit the jackpot.

Smudge   
   

Group: Administrator
Hero Member
*****

Posts: 3198
It's not as complicated as it may seem...
I told him about the 577 views of some of the photos on overunity.com. We both wondered why so many lookers and so few do-ers these days. Then again this is not an easy circuit to replicate compared to a ###### School Girl Motor.
It is fairly obvious to me why very few are kicking into action.

If I was truly interested in having folks take up the challenge of exploring what I presented, I would at least provide a starting point for them, i.e. a schematic of what I had demonstrated. Expecting someone to accurately reverse-engineer a complex circuit/system from a few photos in unrealistic.
   
Group: Guest
It is fairly obvious to me why very few are kicking into action.

If I was truly interested in having folks take up the challenge of exploring what I presented, I would at least provide a starting point for them, i.e. a schematic of what I had demonstrated. Expecting someone to accurately reverse-engineer a complex circuit/system from a few photos in unrealistic.

Dear poynt99

You have a very good point there.

Spokane1
   
Group: Guest
While you guys are looking at circuits I am concentrating on the magnetics, examining what could be going on in the magnetic domain.  Graham know this.  I keep reaching Eureka moments where I think I have discovered the secret, then on further study find the flaw in my arguments.  I am convinced that the very short interrupt time plays an important part, so I look at the magnetic time-constants.  One possibility applies to the mylar filled gaps which may not have constant dimension thus introducing a variable that is subject to inertial mass, which is a slow response that would be unable to follow that short interrupt time.  Another is the frequency response of the ferrites where the high permeability core is likely to have a lower cut-off then the low permeability one.  I tend to concentrate on the effect of a variable gap where a tiny change of dimension can create a significant change in reluctance.  If that gap has constant flux its changing reluctance is simply a changing mmf drop across it, and the energy associated with that is the so called co-energy on the B v. H  or Flux v. current chart.  That is energy change that does not involve voltage, it is invisible electrically.  And as Aspden pointed out, if you have a saturated core the co-energy to the right of the BH curve (invisible) is greater then the normal energy to the left (visible electrically).  So a combination of saturated core and variable gap has some potential for unusual performance.  You'll be the first to know if I hit the jackpot.

Smudge

Dear Smudge,

Keep up the outstanding theory work. I believe you are close if not have already figured out where the ball park is.

The Gray technology deals directly with a saturated core and a variable gap. This is acheived with two opposing open core transformers that are pulsed. What is novel in this design is that an isolated secondary is provided in each opposing transformer and connected such that an arc is developed directly between the iron cores that are now acting as gap electrodes. An arc is formed between the two moving away transformer sections. I propose that this arc creates a relaxiation oscillator process that changes its frequency as the gap widens. This casues a frequency scan that is broadcasted from the arc gases - like a spark speaker. The iron core and all of the windings are radiated with this sawtooth RF. Most of the resulting OU is harvested in the huge force that is created between the two opposing moving transformer sections. Some additional OU is collected from the primary. Gray's engineers were able to make several rotory engines based upon this novel principle, but they started out with simple popping coils and mass drivers.

Your discussion above reminded me of the Gray system.

Spokane1
   
Group: Guest
It is fairly obvious to me why very few are kicking into action.

If I was truly interested in having folks take up the challenge of exploring what I presented, I would at least provide a starting point for them, i.e. a schematic of what I had demonstrated. Expecting someone to accurately reverse-engineer a complex circuit/system from a few photos in unrealistic.

Not to mention the fact that true OU has not been demonstrated for the Gunderson system, by daisy-chaining or self-looping with excess output power running a load. Gunderson's systems, like all such systems, still require a power supply, and when this is disconnected there is very quickly no more power output.

Show me an electrical system with a true output power at least 1.3 times the input power and I'll show you how to self-loop it and/or daisy chain it so that it can run an external load and does not need a separate power supply after starting.
   
Pages: 1 ... 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 [22] 23
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2024-03-29, 04:42:28