Rosemary Ainslie Circuit Demonstration, June 1 2013

[TinselKoala]

@PicoWatt, .99, MileHigh:

Here's a bit more information. I've just done a few more experiments with Tar Baby, and I've examined, again, _all_ of Ainslie's scopeshots that I can find.

The Figure 3 scopeshot and some others that do not show any current during the Q1 ON times definitely show an applied Gate signal of +12 volts.  And.... these are also the only ones that show Gate signal voltages over about +8 or +9 volts positive. All Ainslie's scopeshots that show current flow in the Q1 ON periods, also show lower gate drive voltages. NONE appear to reach even as high as 10 volts.
Please take a look:

http://seani.justemail.net/rosemary_ainslie/

In line with my experimental results, when the Q1 is "in tact", in place and operating properly, along with all the Q2s, even my Interstate doesn't produce a Gate voltage signal that goes over about 8-10 volts. Only after I pull the Q1 mosfet out of its socket, or place a known bad one in there, can I get the gate voltage signal to go over +10 volts. And in Ainslie's own data we see the same thing. The Gate drive signal does not go over 7-8 volts positive when the Q1 is operating properly. Only the suspicious shots that show no current in the Q1 ON times, also show Gate signals over 10 volts.

This fact also supports the conclusion that the Q1 mosfet is missing or blown or miswired in the Figure 3 scopeshot.

However.... that scopeshot made its first public appearance here:
http://newlightondarkenergy.blogspot.com/2011_03_02_archive.html
on the 2nd of March 2011.

And the Demo Video was shot, apparently, only ten days later, on the 12th of March, and clearly all mosfets were functional then... and clearly it was known, by someone, that the Q1 was stressed, because they removed one battery and only used four for the second part of the demonstration.

And yet they claimed all mosfets in parallel at the March 12 demonstration... but they weren't.

Then later, in April, Ainslie reports blowing several mosfets and damaging the Function Generator and getting a new one.

There is something extremely fishy going on here.

[MileHigh]

Can Rosemary read the datasheet for a MOSFET?   

I think all parties have staked out their respective territories.  Some amazing trash talk has been produced also!  lol  Perhaps a moratorium on all fronts could be declared?

We have two big and exciting new events happening.  Like many I won't sign up to see the video but something is telling me it will end up online.  World saved or nothing exciting, stay tuned.  I wonder how the companies that make MOSFETs are trading these days?  ha ha

There is the fantasy of the Zip-o-resistor reactor.  Giant coiled up tubes, the insides lined with thousands of pulsing resistors.  The super-heated steam output drives a turbine generator.  So you self-loop and also export power.  They ended up calling them Zippos as we entered the Age of Aquarius.  Minor side effect, they predict that by 2019 all terrestrial radio traffic will be drowned out by the Zippos.

[picowatt]

TK,

When the voltage at the gate of Q1 exceeds Vgs(on), Q1 turns on and the voltage at the drains of the MOSFET's is pulled low.

The gate of Q1 is also connected to the source of Q2 and the body diode of Q2 conducts when the voltage applied to the gate of Q1 exceeds the drain voltage plus the Vfwd of the Q2 body diode. As the Q2 body diode turns on, the voltage at the Q1 gate is clamped at (Vdrain+Q2 Body diode Vfwd).

At a Vbatt of 72 volts, Rload of 11R, an Rds(on) of 2R and a CSR of 0R25, the voltage at the drains will be about 12.2 volts when Q1 is on [(72/13.25)X(2.25)].  The Q2 body diode would then begin to conduct and clamp if the Q1 gate voltage were to attempt to exceed (12.2volts+Vfwd), or about 12.8  to 13 volts.

So at Vbatt=72 volts, even if Q1 is operating properly, one would expect to be able to apply +12volts to the gate of Q1.

However, if Vbatt is 48volts, the drains will pull down to about 8.1 volts and the Q1 gate will be clamped by the Q2 body diode at about 8.7 to 8.9 volts.

If Vbatt is 36volts, the Q1 gate will be clamped at about 6.7 to 6.9 volts.

The niumbers are not exact, as they are dependent on the Vfwd of the body diode which is current and temperature dependent, but in these instances is likely close to the .6 to .8 volts used.  As well, the Rds(on) of Q1 is dependent on both its gate voltage and temperature and may not always be the 2R used in the above examples.

PW

[markdansie]

I am with you Milehigh lets wait a few days in peace lol
Kind Regards
PS your welcome on my new site Mile you will never get censored

[TinselKoala]

Thanks for that clear explanation, PW. I knew that was the case but I never before actually examined Ainslie's traces for the gate voltage clamping before. My rough measurements agree with your calculations.

So we appear to have further evidence that the Figure 3 scopeshot is _not_ simply the result of some special FG settings, but rather does reflect something not quite kosher in the circuit.

What is your opinion of the time course of events? Whatever the problem was on the 2nd of March when Fig. 3 was taken, seems to have been fixed by March 12.... and they do _not_ demonstrate a Figure-3-like scopeshot during that demonstration. Then by March 26 she is saying that there is something badly wrong with the apparatus, then on march 31 she reports 2 blown mosfets. All this is happening while everybody still believes that there are all 5 mosfets in parallel: the true schematic wasn't revealed until the 19th of April. April 7-- still down, replaced FG. April 8, finally up again. Still ten days before the "reveal" of the true schematic used.

So there is all this work done on the circuit, replacing blown mosfets and the function generator, _after_ the March 12 demo and _before_ .99 discovers and reveals the true schematic: the Q1-Q2 difference.

How can this be? This is looking more and more like outright fraud to me, the more I look at it. Somebody knew the true state of the circuit and the true nature of the stress to the Q1 mosfet well before the demo video of March 12. Yet the narration, the schematics and all the forum posts until April 19th..... all claimed all five mosfets in parallel.

Unlike .99, I find it extremely improbable that people could work on the apparatus, troubleshoot and repair it after a disastrous failure, and _not_ notice that the Q2s were in "backwards" or not conforming to the claimed schematic.