Rosemary Ainslie Quantum Magazine Circuit COP > 17 Claims

[TinselKoala]

Heh... something like that, I think. Ainslie is definitely acronym-challenged.

Have you seen the latest stuff that she has emitted concerning Jandrell and Garrett? Take a look, you will be very amused.

[TinselKoala]

Back to the Figure 8 scopeshot.

Notice, in the text, that the FG is set to produce a cycle with 20 millisecond period, so she claims. But the scope is set for 500 microseconds per division, meaning that the whole screen only shows 5 milliseconds (500 us x 10 divisions) of the Q2 oscillation portion of the signal. The scope's timebase setting is shown at the top left of the screen.

Once again, this constitutes data fabrication, because Ainslie claims that the high heat in the load is produced during the data displayed on the screen. But it is not-- the high heat in the load "cooking it" is produced by the Q1 ON times, where the overall resistance of the circuit is low and there are no oscillations. During the Q1 ON time, the total resistance of the circuit is about 14 ohms, the current path does not go through the FG, and with a 72 volt supply the current will be a bit over 5 amps, dissipating 250 Watts in the load! (And stressing the Q1 mosfet on its tiny heatsink as used in the original setup for the "papers".) The average power dissipated at the load is dominated by this high power during the ON portion--not shown in the scopeshot at all. Since we don't know the duty cycle used, we can't compute the average power at the load, but it is likely to be in the range of 100 Watts, mostly coming through the single Q1 mosfet.

This Q1 ON portion of the total signal is NOT SHOWN on the Figure 8 scopeshot, since it does not show a complete cycle. The scopeshot is a selective bit of data, falsely put up to misrepresent the true behaviour of the system.

Note that she admits that the transistors are stressed by this setting. How do you think she found that out... I don't wonder. But they are not stressed by the voltage... I have used 830s here with the same effect as with PG50s... but rather by the current, and the power dissipation in the mosfet itself that is caused by its carrying 5+ amps. The Rdss of the PG50 is 2 ohms, so at a current of 5 amps the mosfet itself is dissipating 50 Watts... which it won't do for long, on that miserable tiny heatsink. The solder will melt and the wires will come off the pins! (Electronic solder melts at 190C but the PG50 mosfet doesn't fail until well over 200 C.)

[MarkE]

For those who want more from their AC than their DC:

[TinselKoala]

For those who want more from their AC than their DC:

uh-oh... straight lines in a log-log graph... I can already see the tears starting to flow....


As an aside, psychological research has actually shown that there is a large subset of the population that can't actually interpret graphically presented data. This appears not to be a matter of learning or experience, according to the researchers, but is actually some difference in the way visual information is processed. I find this last bit a little hard to believe, but the first bit is certainly true. We've certainly seen how hard it is for some people to interpret scope traces, without the numbers in boxes, and the few graphs of data in Ainslie's daft manuscripts are incorrectly presented, like Figure 2 in the first manuscript, which has the Dependent Variable (temperature over ambient) along the abscissa and the Independent Variable (supplied power) along the ordinate, backwards from conventional data graphing. Ainslie's display gives the viewer a completely different impression of the data than would a correctly presented graph.

Reference:
Tufte, ER (1983). The Visual Display of Quantitative Information. Pretty much the whole book.

[MarkE]

uh-oh... straight lines in a log-log graph... I can already see the tears starting to flow....


As an aside, psychological research has actually shown that there is a large subset of the population that can't actually interpret graphically presented data. This appears not to be a matter of learning or experience, according to the researchers, but is actually some difference in the way visual information is processed. I find this last bit a little hard to believe, but the first bit is certainly true. We've certainly seen how hard it is for some people to interpret scope traces, without the numbers in boxes, and the few graphs of data in Ainslie's daft manuscripts are incorrectly presented, like Figure 2 in the first manuscript, which has the Dependent Variable (temperature over ambient) along the abscissa and the Independent Variable (supplied power) along the ordinate, backwards from conventional data graphing. Ainslie's display gives the viewer a completely different impression of the data than would a correctly presented graph.

Reference:
Tufte, ER (1983). The Visual Display of Quantitative Information. Pretty much the whole book.

One way to overcome intuitive interpretation is to use direct interpretation:  One can simply find dependent results for particular independent input values.  If for example one had a 0.2uH inductor, a known input voltage of 400Vac, and a measured voltage of 4Vac, one would see that ratio is 1/100.  One could then go to the Y axis and find 1/100 as an approximation and then see that low and behold that corresponds to w=R/100L.  Knowing that F is 2.4MHz, and that w=2piF ~15E6, one could estimate that R ~500 Ohms.  Could someone then further use this dark alchemy to estimate the true current??  Could it be that the current would be a close approximation to:  Ipp = Vpp/jwL?  How much is 4V/5Ohms??  Is it 14A as Ms. Ainslie insists?