Testing the TK Tar Baby

[MileHigh]

Hey TK:

More thoughts with standard disclaimers....

Not sure what the bandwidth of your scope is but I think the 4 MHz "spikes"  (spears?) are probably pretty accurate.  Usually if a signal gets mushed up it becomes a sine wave.  It's highly unlikely that a mushed up signal will change into something with a lot of high frequency components.  Also, a lot of Poynt's simulations had very similar looking spear waveforms.

Note that because you basically have a MOSFET array screaming in the RF band that everything becomes very touchy and sensitive and Heisenberg's uncertainty principle is in play.  Perhaps you put the inductor in parallel with the two LEDs to allow some DC current to pass through the inductor and thus reduce the chance of popping an LED?  Something happened and your LEDs across the CSR changed in behaviour.  It's possible (but I am starting to feel like I am shooting blanks) that my proposed 4 + 4 back-to-back LEDs in series with the battery positive is still viable.  It's possible that the "outbound" LEDs pointing towards the circuit will glow more brightly than the "inbound" LEDs because of the fact that the current flow is DC with AC superimposed on top of it.  Perhaps more importantly, they may not disturb the circuit too much such that both LEDs across the CSR will still glow evenly.  Your addition of an inductor in series with the battery positive may have been too much of a disturbance.

The "hard core" analysis that the RAT team never even discussed is the oscillation itself and it's mechanism.  My gut feel is that it's just a variation on the theme where a microphone + amplified speakers cause high-pitched feedback in a PA system.  It's arguable that understanding that oscillation is the key but I am not asking you to go there.

In the world of modelling electronic circuits with amplification you can do some fancy footwork with simplified differential equations that can characterize the amplifier as a "filter" that is described by it's "S" parameters.  If I recall correctly, "If there is a "pole" or poles in the right half of the complex "S plane" then the filter can spontaneously oscillate at one of the poles.  Different poles correspond to different frequencies.  I am sure that you have played with a circuit that might be oscillating at say 1 KHz and then you touch it and it "jumps" and starts oscillating at 3 KHz.  There you go....  The S plane in action.

MileHigh

[MileHigh]

TK:

About your FM radio squealing at 88 MHz.  That's very interesting because FM radio was specifically designed to be immune to static and noise sources like kitchen appliances, etc.  Frequency modulation vs. amplitude modulation, a very important difference.  It's very rare that you hear interference from an electrical device on the FM radio band when you are tuned into a station.

I suppose that it's possible that the Tar Baby is pelting the radio with tar balls and it's one of those very rare cases that you are overcoming the inherent FM noise immunity.  However, there is another distinct possibility.  The other possibility is that your Tar Baby is so damn noisy that it's disturbing the radio _after_ the FM demodulation is done in the radio.  So the Tar Baby is injecting noise into the circuit board of the FM radio itself.  That's one noisy baby.

Just imagine the nightmare of living in a world full of pulsing inductors everywhere with collapsing magnetic fields.  We would all have to wear radiological exposure badges and sleep in Faraday cages lest we all fry our DNA.

MileHigh

[picowatt]

TK,

If you will:

With the circuit oscillating, and while observing the drain signal, what happens if you place a cap across the points where your FG is connected to the circuit?  Try using a cap with decent HF characteristics and at least a .1uF of capacitance.  A larger value is better if it has good HF properties.  A voltage rating of 15V or better should be used.  A ceramic or tantalum would be preferrable.

Just briefly touch the cap across those points (at the end of the FG cable) and let us know what happens regarding the oscillation.

PW

P.S., FM IF is 10.7MHz if I recall...

[fuzzytomcat]

Yes, I thought so too.

I'm wondering... has there been any progress towards testing the NERD device? I haven't been looking at other threads here and I'm wondering if the NERD RATs have dropped their application or decided not to pursue the matter further.

Because I think we've identified some very easy little tests and trials that might reveal a lot about that device's performance, if somebody would only do them. If the NERD RAT device still exists somewhere, and is the revolutionary COP > Infinity device it has been claimed to be..... wouldn't you think at least _somebody_ would be willing to stay up late, in a room full of electronic test equipment and a computer and a camera or two, testing and reporting until everybody understood what a miracle it is and just how well it worked.

I know I would be, if I believed, really deep down, that I actually had such a device. Heck, I am anyway, and I doubt very severely that I will be able to get anywhere near COP > Infinity with the Tar Baby.

Heck, I'd be happy if it just did half that.

Hey TK,

Your right about the NERD RAT testing, you would think that any of the authors that appear on the papers that were submitted for publication in a accredited journal or magazine would be all over the proof or verification of the COP>INFINITY claim. Especially Donovan Martin whom owns Donix Embedded Systems http://www.donixes.co.za/  specializing in "electronic circuits" and his name appears in all four (4) papers over the years Rosemary was a part of. You would think Donovan would be around right now backing up the claims in the COP>INFINITY papers and posts at OU Rosemary made .... but so far a no show ever. 

FTC

[TinselKoala]

TK,

If you will:

With the circuit oscillating, and while observing the drain signal, what happens if you place a cap across the points where your FG is connected to the circuit?  Try using a cap with decent HF characteristics and at least a .1uF of capacitance.  A larger value is better if it has good HF properties.  A voltage rating of 15V or better should be used.  A ceramic or tantalum would be preferrable.

Just briefly touch the cap across those points (at the end of the FG cable) and let us know what happens regarding the oscillation.

PW

P.S., FM IF is 10.7MHz if I recall...

Well, all the tantalums I have in stock are in the multi uF range. I have some good ceramic caps, Y5U dielectric, marked 500M 15kV. Depending on the FG amplitude setting, this either increases the magnitude of the oscs or has no effect. Other caps I tried either diminish or kill the amplitude of the oscs. I didn't look at high resolution, though.

I'm running a temperature profile test right now, bipolar pulses so both Q1 ( one 830a) and the Q2s are active, with the one PG50 in a Q2 slot seemingly taking most of the load on that side of the phase. Load is heating like an overworked espresso machine, DC inline ammeter indicating 1.45 amps. I'm running with the back-to-back LEDs with parallel inductor in series with the battery; both LEDs glowing brightly.

The rest of my PG50s have still not arrived.