Rosemary Ainslie Quantum Magazine Circuit COP > 17 Claims

[TinselKoala]

Gmeast, I think we are acknowledging that you have skills and knowledge, and I don't think that is seriously in doubt. But the question is about your own analysis of the things Ainslie has claimed all along, and most particularly in the past few weeks, and even more pointedly in the past several days. I know you read her posts. So... what do you think of her claims now, from a practical scientific viewpoint, and what do you think of her attitude of willfull ignorance and arrogance, from an interpersonal, psychological viewpoint?

We respect you enough to be asking you these questions, that should be plain.

For example, do you believe her plain accusation that I cheated by showing a stored waveform, or in any other way?

[MarkE]

I'll post which make and model non-inductive resistors I used in all of my tests as soon as I can find the Mouser invoices. I used two values depending on what I was measuring. I used 0.1Ohm resistors for measuring lower voltage drops like for PWM power determinations and 0.05Ohm for the higher voltage drops like for the Inductive Resistor Heater Element. However, I don't use the Q-Array. My setup is much more basic, but measurement is just as rigorous and requires more than just basic measurement skills (and knowledge)... which (even though you don't believe it) I do have.


Regards,


gme

If you have pictures of your arrangement that would help assess where you likely are with respect to parasitic inductance.  You should always use the largest sense resistor value that you can tolerate from a voltage drop and power dissipation standpoint.  The R/L angular cut-off frequency obviously falls with R, and rises with L.  A lot of resistors that are sold as "non-inductive" exhibit 100nH or more inductance.  In the case of Ms. Ainslie's circuits the oscillations occur around 3MHz ~= 20E6r/s.  A 100nH inductance looks like 2 Ohms at that frequency.  The sense resistance would have to be 6 Ohms in order to limit the amplitude distortion to 10%. 

There is another way:  You can make up a sense circuit that places a small series R/C network across the sense resistor that compensates for the R/L response with an RC response.  The capacitor connects to the common side of the sense resistor and you read across the capacitor.  The resistor is adjusted to get a flat frequency response.  This is basically the same idea as the frequency compensation adjustment on an oscilloscope probe.  In order to make the adjustment you can drive the circuit with a fast rising or falling current at a low repetition rate, like 1kHz.  Then just like the scope probe adjustment, dial in a flat response.

[TinselKoala]

Gmeast, has MarkE given you sufficient enlightenment or would you like to discuss my videos? Even though I think that the explanations given in them are perfectly clear, if you start at the beginning--- and especially if you have watched the Negative Bias playlist from the beginning... I would be more than happy to explain any thing in them and I would indeed like your honest criticism of the points that are being made that relate to Ainslie's claims. For example in the post you quoted, she predicted, very nastily, that I would blow out the FG if I used six batteries in series and put those in series with my FG. As you can see from the videos, I have done just that, and not just with the simplified half-Ainslie circuit but also, in the Tar Baby demonstrations, I did it many times.

So... your opinion please. And why are you critical of my attitude towards Ainslie? There is nobody, nowhere that she has insulted more than me, and with all her false accusations, proven lies and ridiculous claims, coupled with her deliberate ignorance and refusal to follow simple explanations... can you blame me that I am angry with her? I have worked long and hard in an attempt to explain the action of the circuit, nobody has made any refutations of my work at all, and yet Ainslie still persists in her remarkable set of errors. Do you deny, for example, that I have shown how the Q2s are turned on in the circuit by Grounding the Gate and Lowering the voltage at the Source pin of the transistor, something that Ainslie clearly believes, still, to be impossible, in spite of the references I have given her to the Common Gate Amplifier configuration? Something that Ainslie has actually accused me of FAKING by showing a stored waveform... on an analog oscilloscope from last century? Would you not be frustrated and angry if someone refused to believe your results that are perfectly ordinary, but rather accused you of FAKING them?

[MarkE]

Ms. Ainslie must be a big fan of the old Monty Python routine: "The Argument Clinic".  Even though TinselKoala's videos systematically disprove Ms. Ainslie's claims she's adhering to those disproven claims in a game of:  "No it isn't."  Likely there are very few that Ms. Ainslie fools other than herself.

[gmeast]

If you have pictures of your arrangement that would help assess where you likely are with respect to parasitic inductance.  You should always use the largest sense resistor value that you can tolerate from a voltage drop and power dissipation standpoint.  The R/L angular cut-off frequency obviously falls with R, and rises with L.  A lot of resistors that are sold as "non-inductive" exhibit 100nH or more inductance.  In the case of Ms. Ainslie's circuits the oscillations occur around 3MHz ~= 20E6r/s.  A 100nH inductance looks like 2 Ohms at that frequency.  The sense resistance would have to be 6 Ohms in order to limit the amplitude distortion to 10%. 

There is another way:  You can make up a sense circuit that places a small series R/C network across the sense resistor that compensates for the R/L response with an RC response.  The capacitor connects to the common side of the sense resistor and you read across the capacitor.  The resistor is adjusted to get a flat frequency response.  This is basically the same idea as the frequency compensation adjustment on an oscilloscope probe.  In order to make the adjustment you can drive the circuit with a fast rising or falling current at a low repetition rate, like 1kHz.  Then just like the scope probe adjustment, dial in a flat response.

You're talking 'apples & oranges'. Looking at the images in my slide show, what's shown are not the types of oscillations seen in the Q-Array or in any other part of Rosie's papers or in other replications. In mine, all you see is a (pretty much textbook - maybe plus one extra cycle) damped oscillation seen when ANY switch that has been charging an inductor is opened. There are only about 12 complete oscillations in the wave, with each one of them being successively of lesser amplitude. The waveform you see in my figures is exactly what you would see from a high-voltage scope with no probe attenuation and set on a 1:1 scale. There's no parasitic 'anything' in such a wimpy waveform as mine and any math scope will return an accurate mean value for the voltage drop across a non-inductive current measuring resistor applied to that waveform.


Something NOT in my data, but that kept me optimistic, yet sober, was the simple fact that my batteries recharged faster following the pulsed loading tests (powering the Heater Element) vs. the purely resistive load tests ... and not by just a little bit ... roughly 6-Hrs vs. 8-Hrs respectively. This was a common factor throughout the dozens of tests I ran. If you look at my 'crazy' startup procedure, you'll see that the batteries were charged (on a charger made for charging AGM batteries) that automatically tapered off and shut down to 20mA. Then the batteries were unhooked from the charger and left to sit for 8 more hours before being loaded on any test.  And also NOT in my published data (and I consider my slide show as "published") is the same battery-charging counterpart as applied to the second resistive load test wherein the lesser load and pulsed load had the same starting and ending battery voltages but wherein the lesser resistive load caused a lower temperature (rise over ambient) than the pulsed load on the same Heater Element test fixture ... yet they took nearly identical times to fully recharge the batteries after those tests. So there's yet another, albeit indirect, result in support of my conclusions as presented in my slide show.


There are too many resultant factors in blatant support of one another. My conclusions are simply NOT the result of measurement error or measurement anomaly.


... but thanks anyway. Regards,


gme