Rosemary Ainslie Quantum Magazine Circuit COP > 17 Claims

[MarkE]

So the circus is threatening to close and quietly disappear. Never fear. The lack of attention will quickly bring it back to life.

It looks like .99 has missed the deadline and will henceforth be banished to purgatory. All he had to do was acknowledge that partially defined loony tune tests are valid. Hang on,  he must also agree to the results before they are measured - Sounds like Zimbabwean elections. Also please stop raiding the Cape Town computers. Obviously this is where TK gets all his data.

That Mosfet is tough. It's avalanche rated so providing the duty cycle is short enough it will take that punch almost indefinitely

That's the hilarious part.  Ms. Ainslie claims that she wants the world to see her alleged discovery.  In the meantime she's saying that she won't show any evidence because Poynt99 won't go along with her demands.  She says she's drawing a line that she should have long ago.  Yet, just a couple of weeks ago she plead with Greg East to be polite to Poynt99 because she felt he was critical to legitimizing her experiments.  Now, she's put herself between a rock and a hard place.  If she shows nothing, she's back at square one.  If she shows experiments that conform with reality, then she's sunk.  If she shows experiments performed incompetently, then she's sunk.  I think that TinselKoala is most likely correct:  Whoever is performing her experiments are not getting the unreal results she's previously reported.  Now, she's: confused and embarrassed and looking for a way out.

[TinselKoala]

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

I put a MUR1560 in and I can see little or no discernible effect on the VBatt trace. With the capacitor alone (I used 0.1 uF 400v poly film) or in conjunction with the diode, the whole effect is killed, the performance of the circuit is entirely altered. Perhaps the frequency of 187 kHz is too fast? The resultant waveforms on both Vcvr and Vbatt are nearly sinusoidal and look nothing like what I expected. I'll try to get a scopeshot later on. I have to change some trigger parameters on the scope to capture that waveform properly.

However, with the MUR1560 alone, with anode connected but cathode free, I can get the rectified Drain spike at the cathode and use it to run, eg, the NE2 ring oscillator. The non-rectified spike (taken from the anode end) will barely run the Ring Oscillator, but the rectified spike from the cathode end runs it really well. There is a visible effect on the Vbatt trace with the ring oscillator: the first positive spike decreases in amplitude by a few volts. I see nothing happening on the bottoms of the ringdown spikes. I've made a Supplement video that kind of shows these effects (but not the capacitor effect yet) but it's not too coherent, I haven't had enough coffee yet.

[MarkE]

I put a MUR1560 in and I can see little or no discernible effect on the VBatt trace. With the capacitor alone (I used 0.1 uF 400v poly film) or in conjunction with the diode, the whole effect is killed, the performance of the circuit is entirely altered. Perhaps the frequency of 187 kHz is too fast? The resultant waveforms on both Vcvr and Vbatt are nearly sinusoidal and look nothing like what I expected. I'll try to get a scopeshot later on. I have to change some trigger parameters on the scope to capture that waveform properly.

However, with the MUR1560 alone, with anode connected but cathode free, I can get the rectified Drain spike at the cathode and use it to run, eg, the NE2 ring oscillator. The non-rectified spike (taken from the anode end) will barely run the Ring Oscillator, but the rectified spike from the cathode end runs it really well. There is a visible effect on the Vbatt trace with the ring oscillator: the first positive spike decreases in amplitude by a few volts. I see nothing happening on the bottoms of the ringdown spikes. I've made a Supplement video that kind of shows these effects (but not the capacitor effect yet) but it's not too coherent, I haven't had enough coffee yet.

The drain waveform should look something like this, provided that you have the scope probe ground right at the MOSFET source pin, and you've kept the loop area small.  A picture of the wired set-up would be helpful.

[TinselKoala]

The drain waveform should look something like this, provided that you have the scope probe ground right at the MOSFET source pin, and you've kept the loop area small.  A picture of the wired set-up would be helpful.

I know it should look like that, but it doesn't.

No, the wiring is more like what Ainslie is probably using. These trials aren't particularly designed to do "proper" measurements, they are designed to replicate Ainslie's scopetrace, and move on from there. I'm trying to get heat results measured properly, correlated with scopeshots and circuit parameters as used by Ainslie.

Here's the current physical layout, with the locations where I put the diode and capacitor sketched in.

[TinselKoala]

Actually, the Drain trace does look very much like that, except there's more ripple and a LF sinusoidal wiggle in the "off" portion with the cap in place. The diode eliminates the drain positive spike.

But here is what the Vbatt and Vcvr traces look like with the capacitor installed. Adding the diode makes only a tiny difference on these traces here. This is made by setting up the same screen as before, with the 200V Vbatt spikes and the nice Vcsr pulse shape and negative spikes, then simply connecting the cap as shown. The traces collapse to this:
(Also including the non-cap screen below for comparison)