TPU-ECD FAQs answered from the builders

[hartiberlin]

Hi All,
this thread should only be here for FAQs which have been answered by the builders
and replicators of the TPU-ECD, so this thread is locked and sticky, so only
I can post the answers from the builders.
( There is unfortunately no other way to control the posting this way, as there is
not yet any posting moderation in the forum software and I don?t want any NON-builders
to post silly questions or comments over here...!)

Read first these PDF files inside this thread:
http://www.overunity.com/index.php/topic,2383.0.html
and then continue here:


Here are the latest tips from Roberto:



Hi Stefan,

1 - I do have a 6" and a 15" ECD both at the moment OK but to be eventually reconnected to breadboard in order to do additional tests. Of course it take time. And at the moment I'm just doing it slowly (as I can work at it only from 9PM to 10PM.......lot of Office work to do) and reassembling the 'big eye' to make the film I promised you. So......it is necessary to wait for measures. I do remember you that this design is unique in the sense that power output seems only related to the percent 'opening' of the 'door'. Again....when conversion is operating the PS delivered current -> to ZERO. Of course ZERO is impossible to reach as in that case you shoul have INFINITE POWER!!
You see by yourself that probably in this case the COP is meaningless for this class of devices that should be put outside the classic theories.

2 - In the past I've used analog testers as well ...but I found the EM fields tampered them in many situation. So a solution could be to put ECD into a metal grid box in order to lessen the effects. But it take time... and I've so few. Please let the other that are only lurking or just only speaking DO SOMETHING.

3 - Me and Otto gave to all of you FOR FREE an incredible DISCLOSURE. ARE YOU SURE THAT YOU ALL ARE REALLY UNDERSTANDING HOW BIG GIFT IS IT??? - THAT'S SCIENCE FANTASY!!!!!!!!  I CALL THIS DEVICE: STARGATE. IT JUST OPEN a door to ZPE I'm now confident.
ronotte

Hi Stefan,

I do assure you THAT ALL IS REAL just follow exactly the schematic I delivered in TPU=ECD and you will be 100% OK. Anyway no problem to answer you:

1 - Mobius loop is A MUST. Only 1 turn: it is a MUST.
2 - Loudspeaker cable (like Twinflex...every wire sections) is OK.
3 - The CC are transformers. Primary is winded in parallel way with Secundary. Of course as primary (optimized) wire lenght is 4.2 mt. (0.5 -1 mm diam. wire) and secundary (optimized) lenght is 9 mt (about 0.16 mm. diam. wire), it does mean that you start winding together the 2 wires, when you reach the end of the first wire (4,2 mt) you simply continue winding till the secondary wire end. Transformer width (from right side to left side) is 25.4 mm.

Please feel free to ask me for any problem. I inform you I'll put soon a FAQ page.

Good luck

Roberto


Hello Stefan,

I'm giving lot of similar infos to Jason and all is team in order to ease their replications.

1 . Transformer. This is an Otto's idea. He just optimized all to obtain best efficiency. So for the moment (!) just do so....no standard answer (bad for me too...).

2 - You can put the 'LOAD' phisically where you want. Don't think to change the schematic. Vary anything ...and it will not operate..that's a MUST. Just do it as in my doc: this work for sure as I did it with my hand.

3 - many guys think to 'adapt' the MOSFET driver circuitry. DON'T TRY. IT WON'T WORK ANY MORE. I know you don't believe, sorry but you have . I've been a digital designer during my yougtness for almost 15 years doing really research for the italian nuclar reactor control room. I DO Know what I'm saying!

Summarizing: as MOSFET's driver you can use with similar results: IRF7307BC or UCC3724 or MIC4427. If you don'make use of one of them NO-WAY to put in 'Avalanche mode' the MOSFET. I spent a week of my precious time to do all the possible way around as: CMOS driver, TT driver, high speed switching trannys, operational aplifirers, etc.  NO-WAY.

This is the last time I will answer to this kind of questions. It seem to me that the guys won't believe/understand...so let them do what they want...but don't complain not obtaining my results!.

Stefan I'm in my Office so I've to leave you.

Best regards from

Roberto

P,S, yesterday night I have started tests at my reconnected 15" ECD.

[hartiberlin]

This is from Jason:

A lot of people seem to think that in the demo video, my bulb us pulling 150W+ or so to light the bulb! This is absolutely ridiculous and I can prove it! In actuality, the bulb is only being powered by a fraction of a fraction of this amount because the control circuit is GROSSLY inefficient.

Anyone who knows anything about transformers (especially pulse transformers) will tell you that the power is only transferred to the secondary coil when the square wave is switched on and off. In my case, ANY power the bulb would get from the power supply would HAVE to be transferred in the 46 ns that it takes for my MOSFET drivers to switch on the FETS, and an additional 46 or so ns to turn it off! the other tons of ms that the square wave is on (and drawing loads of current) is all a complete waste of the input power! I am driving my FETs with a 50% duty cycle square wave and this will be the first thing that I correct.

The second thing people need to realize is that the PS voltage went up to +44V not because the regulator circuits inside of it were tricked (not that I'm completely ruling this out), but the main reason is that there is LOADS of BEMF voltage spikes being shoved into it! If you look at Otto's diagram, you can clearly see that the coils are switched to ground, which means that when the MOSFETs are shut off, all of the energy loaded into the magnetic fields will reverse and head straight towards the power supply. This is what I believe is giving the appearance of increased voltage across the meters.

But, this appearance of the BEMF (according to Roberto) is a clear sign that I don't have the frequencies adjusted correctly. To be honest, I never really tweaked the second frequency to get the results that you all see in my video. I basically tuned to find a good sweet spot with the first frequency, then I used another sweet spot we found earlier for the second frequency. and as soon as I switched it on, BAMO! it worked! BUT! Even with that alone, I can use the frequencies to fix the current draw issue. Roberto said that when it is tuned right, It is suppose to draw little to no current from the PS. I personally witnessed this several times when I was adjusting only one frequency. Normally the bulb would draw about 1.5A constantly from the supply while the bulb was "out of tune." But when I got into the right frequency range, I saw the input drop as low as 690-700mA! And that is with my inefficient, wasteful 50 duty cycle square wave input!!

So I see NO reason why this TPU is not FAR overunity here! A few simple modifications to the driving circuit should show this immediately. As for the BEMF spike problem, it's really not a problem at all. If we make the pulse width small enough, then pretty much all of the energy will be stored in the magnetic field of the pulse coil, which means that we will get almost 100% of it returned (minus the small resistive losses in the coil). When it comes back, we simply store it in a capacitor (which would be in parallel with the power supply). Since I clearly saw the voltages far exceeding the 12V input of my DC PS, it would be no problem at all to charge the capacitor above the PS voltage. Once this happens, it would be possible to almost run the TPU entirely off of the capacitor! Of course, there would need to be a diode between the cap and the power supply to keep it from discharging into the supply...

This is my current line of reasoning. Soon, I don't see why we couldn't run this thing off of a watch or AA battery once it is in tune right!
God Bless,
Jason O

[hartiberlin]

This is from Darren.
Many thanks for redrawing it to get it more understable:


As an aid in helping me determine the theory of operation for Otto's and Roberto's device, I drew their circuit out in a manner that seems a little simpler....for me anyway. It's not too pretty, but it is hopefully clear and makes sense.

Just thought others might benefit by seeing it drawn a different way.

I won't go into the theory of operation, but one may notice that all 3 primaries, and all 3 secondaries are in parallel.

Any one of the 3 switches will energize all 3 primaries, somewhat in an "OR" fashion.

Darren

http://www.overunity.com/index.php?action=dlattach;topic=2535.0;attach=9662

[hartiberlin]

Here is a first movie of Roberto,
just showing his measurement TPU-ECD setup.
Not yet lighting the bulb:

http://www.megaupload.com/?d=SIL3B8KL

[hartiberlin]

Hi All,
Roberto posted a full schematic plan of his TPU-ECD over here:

http://www.overunity.com/index.php/topic,2535.msg37309.html#msg37309