Rosemary Ainslie Quantum Magazine Circuit COP > 17 Claims

[MarkE]

Nice replication TK.

Could you try a flyback diode across the resistor to see if the Vbatt trace changes?

Thanks.
.99

In order for a clamp diode to be effective you will also need a decoupling capacitor as shown in the drawing below.

[TinselKoala]

Film at 11:

http://www.youtube.com/watch?v=YPzuPOK2lio

[TinselKoala]

Oh Daddy, havent you analised her enough yet ? You have. 

You proved her circuit works better than v!agr@ 

Now please give some rest to your tool(s)  or else your chips will fry with overload.

Google from www.viagranalising.com 

Ps: Even Mur@kamis (snake oil salesman), throat is choked now. Take some rest. 

But Google..... it seems that Ainslie still hasn't had enough yet! Look at her latest emissions, where she threatens all of us, accuses us of more impossibilities, and announces that she is NOT going to make anything public unless Poynt99 licks her boot. I declare, the woman's delusions of paranoia are absolutely stunning! How she manages to fall out of bed and dress herself in the early afternoon, I'll never know.

But ROFL! It is just as I predicted.... the lying troll queen has nothing. She has no thermal RATE data and never has had. She has nothing but silly scopeshots and low heating, while her batteries just continue to discharge and discharge and discharge...... Poor old Donovan Martin must be sweating bullets by now, since he must realize at this point that there is no hope, no hope at all, of ever seeing any COP over 1 when he's connected properly and does his sums right.

And now, like some huge pale South African cockroach, the Troll Queen is trying to get out of the light, trying to hide under her soggy rotten log again. She's trying to apply for OverUnity Prizes on Open Source websites.... while hiding everything she can hide. Her hypocrisy is only exceeded by her arrogance and her willfull ignorance.

[TinselKoala]

HEY AINSLIE.... DON'T FORGET THAT I HAVE IMAGES OF YOUR MANY POSTS WHERE YOU INSULT, MAKE FALSE ACCUSATIONS OF CRIMINAL ACTIVITY, AND OUTRIGHT LIE ABOUT YOURSELF, ABOUT OTHERS, AND ABOUT WHAT YOU HAVE DONE. Your imaginary lawyers will choke when they see all of that solid EVIDENCE of your fraud, your duplicity and your threats.

DON'T FORGET THE SOLID EVIDENCE WE HAVE UNCOVERED THAT YOU FABRICATED DATA FOR YOUR "PAPERS".

[TinselKoala]

That is a great demonstration of parasitic wiring inductance and capacitance.  The faster that the MOSFET switches, the more pronounced inductive distortion is in the current sense, visible as rising to a higher value that then declines exponentially during the on phase.  And Warp 10 is right for that turn-off.  With limited parasitic capacitance to dump charge into, the greatly sped-up turn-off of the MOSFET requires a much higher voltage.  With a fast enough driver one can easily exceed the Vds rating of the MOSFET, which is why a TVS from drain to source is a good idea.

These PG50s are tough little buggers. I have not yet popped a single one from overvoltage spikes. I have had a couple go from inadvertently exceeding Vgs max, though, and of course I deliberately drove one to thermal failure for the Stress Test. Still, the TVS is a good idea, just for the sake of good practice if nothing else. I may have one of appropriate value in one of the stash boxes.

Right now I can make bigger VBatt spikes but I'm deliberately keeping them small enough to fit on-screen on the Link. I don't have a 100x high-voltage scope probe, and with the 10x probe the least sensitive input setting on the Link DSO is 20 v/div. To go to 50 v/div I need a 100x probe, or go to the analog Tek 2213a, which will do 50 v/div with a 10x probe. The Drain trace is spiking to well over 250 v already, as displayed on the Tek. For the CVR trace I am using the Ohmite non-inductive resistors but I am deliberately maximizing the probe connection wire loop by clipping far from the resistor bodies. Still, I probably have less inductance associated with my current sense system on this setup than Ainslie does at best.

It should be clear to just about any observer that these large spikes, 180 degrees out of phase, positive on VBatt coinciding with negative on Vcvr, will most probably produce large "negative power product" when processed in the Ainslie fashion. It is frustrating that the Link can't do live trace multiplication, but I can always do it manually or from a spreadsheet dump. (Yes, dear Ainslie, I can do data dumps into spreadsheets too, and surprise surprise, the spreadsheet data when plotted looks just like the scopeshot it came from. Duh. How could it be otherwise, since it's the same data?)

I need to try the gate boost circuit with the SWeir board, which is properly filtered and has the really really noninductive precision cvr with proper probe contact points, just a tiny loop which will produce minimal distortion in the current signal. I think it would be great to see the mosfet switching cleanly and the ringing filtered out or properly damped. Waste less power in the mosfet, dissipate more in the load element, boost efficiency that way.