Testing the TK Tar Baby

[hartiberlin]

Hi TK,
please set the 10 K pot to the oscillation setting at around -4 Volts,
then remove it and measure the resistances from
the center tap to the 2 ends of the pot.

I guess it is still at least 1 kOhm in this setting,
so at 9 Volts then a maximum of about 10 Milliamps
can only flow as a DC current from the battery...
( 1 Kohm parallel to 9 KOhm as the lowest resistance !)

Maybe the oscillations just flow as an AC current then due to the stray capacitance
via your loose wires setup so the battery is drawn flat from the AC currents ?

With the DC currents it can not go flat, if your pot is set to about -4 Volts what you said.
as this is about the center of the pot and then you have at least about 3 to 5 Kohm on each side of the battery,
so  the DC current can not get bigger than 5 to 10 milliamps... It is just Ohms law...so 200 mA is wrong.

So if it gets flat it is either a bad old battery or the AC current sucks all the energy out of it....


Regards, Stefan.

[TinselKoala]

TK,

With as much as you've worked on this today, I guess you did indeed not get any "real work" done!

Your efforts are appreciated.

PW

Thanks. I managed to walk the wild canine, but that's about all I've really gotten done today.

[picowatt]

TK,

With only 9volts to work with, you will have to reduce the 50R as you did to get a similar bias current.  My estimates were based on the FG at full offset approaching -14 volts and an Rgen of 50R.  As I have been saying, 100-200ma of Ibias may not be required if other aspects are played with, i.e., changing wire lengths/layout for changes in inductance and strays.

I would think that the FG would act similarly to your 9V and be an additional "battery" in series with the main battery string and included in the whole loop.  If 100ma DC is flowing, then I would think the battery, or FG, is acting like a "battery" and discharging a similar amount of current.  As the overall available voltage is higher (i.e., the negative source in series with the main battery string) overall dissipation will increase a bit at the load due to the higher "loop" voltage.

Again, the required bias current may be much lower 150ma.  I use FETs all the time biased in the microampere range, but I have never attempted such low bias currents with a power MOSFET.  We do not know RA's FG settings so we cannot estimate the bias current in her tests.  In reference to one test she indicated full negative offset with the FG, and that is what I based my 150ma plus or minus 30-40ma and a maximum of around 200ma estimates on.  And that would be dependent on the Q2 device temperature as well.

With a fresh 9V battery, and its positive terminal tied to the MOSFET side of the CSR, a resistor of approx 30R between the 9V battery negative and the source of Q2 should give somewhere around 150ma of bias current (plus or minus 30ma or so).  Again, the actual current will depend on the FET turn on variances and the temp of Q2.

The reason I always "push" for 50R is that it allows strays to act similarly to the NERD circuit.  But as those are also not well defined, I would try to bias the circuit so that the oscillations look similar to the scope shots in the first paper (with regard to their symmetry and clipping).

PW

[TinselKoala]

Hi TK,
please set the 1ß K pot to the oscillation setting,
then remove it and measure the resistances from
the center tap to the 2 ends.

I guess it is still at least 1 kOhm in this setting,
so at 9 Volts then maximum of about 10 Milliamps
can only flow as a DC current from the battery...
( 1 Kohm parallel to 9 KOhm as the lowest resistance !)

Maybe the oscillations just flow as an AC current then due to the stray capacitance
via your loose wires setup so the battery is drawn flat from the AC currents ?

With the DC currents it can not go flat, if your pot is set to about -4 Volts what you said.
as this is about the center of the pot and then you have at least about 3 to 5 Kohm on each side of the battery,
so  the DC current can not get bigger than 5 to 10 milliamps... It is just Ohms law...

So if it gets flat it is either a bad old battery or the AC current sucks all the energy out of it....


Regards, Stefan.

No, I think you are still not understanding me, or I am not understanding you. The minus 4 volts is some kind of "floor". The oscillations just begin at low amplitude when you reach this value, and the current from the 9v is about 15 mA. Now if you keep turning the pot, the voltage _Indicated_ stays at 4 volts... because you are now feeding a low impedance load somewhere, and increasing the pot setting just sends more current but the apparent voltage stays at -4 volts _indicated_.  This is the "voltage floor" phenomenon that .99 mentioned from his sim and that apparently also occurs in the NERD device, since none of their negative going gate drive traces go more negative than this.
The oscillations grow in amplitude and the current from the 9v increases as you rotate the pot. With the combination of the partially depleted battery, the 10R series resistor and the pot set ALL THE WAY to one end, the current is 100 mA on the moving-coil meter, and could be much higher with a fresh battery and no series resistor.

The only way that the AC current---really RF---could suck the energy from the battery is by turning on the mosfets partially and allowing them to dissipate the 9v battery's power as heat.

There is no one "oscillation setting" of the pot. The oscillations start small and grow, until with ridiculous voltages like 15 volts _applied_ but not _indicated_  (the voltage floor) you are chancing exceeding the voltage limits of the mosfet gates-- but you have huge oscillations and the mosfets are passing a lot of current. 4 or 5 amps through the mosfets is easy to obtain if you have the main voltage up above 48 volts and can source 350 mA from the bias source-- and by Ohm's law this is telling us that the mosfets are completely or nearly completely on with that amplitude of oscillation, and of course the load heats very rapidly under that current.

My "10K" pot actually measures 11.2 K on the Simpson DMM, and the wiper is 380 ohms from the end that's connected to the negative pole of the 9v battery when the oscillations just begin and the Hickok meter is indicating 14 mA. Turning the pot so that the resistance is less, makes the current go up. And when the wiper is at the most negative end and the inline 10R is used, the current is 100 mA or slightly more. But my battery now measures 7,95 volts no-load, so all these numbers will change with a fresh battery or another current source.

[TinselKoala]

@PW: Let's go for 100 mA bias current. This keeps it on the Hickok meter, and it makes nicely formed oscillations, and more importantly it gives me a stable and manageable 190-200 mA on the inline DMM which seems to reflect what the load is getting fairly accurately as far as I can tell so far. This produces easily measurable heat and fairly rapid battery depletion with my 5 A-H batteries, and should give manageable run times with NERD's batteries too... ten times longer but still manageable.

So with a fresh 9v, or with a regulated 12 volts input to the Voltage Inverter/Clock, or some other source... it seems that the 10R series resistor is minimum and 50R is a max... maybe I should put a nice wirewound 100R precision pot in there instead of the fixed series resistor.