Testing the TK Tar Baby

[TinselKoala]

Hi PW....

Great minds think alike! We are tracking perfectly, as far as the series gate resistor and the supply to the 555 goes. I have a 0R3 on the board in that position, but I will add another 10R in series after I've woken up and see what's up. The 555 I'm using is a Philips branded NE555N, which I have a handful of. I also have some TI branded 555P units. Adding the little heatsink helped stability, and I can operate at 320mA sustained with the 830as in all positions. With the PG50s in Q2 slots, the oscillations happen at low power input to the 555, like 5-7 volts indicated on the PS meter, and go away as power is increased and I can't sustain more than about 240 mA in oscillation mode. The circuit will oscillate with only a few tens of mA indicated but the best undistorted envelopes seem to happen at around 90-110 mA (the zoomed oscs are almost perfectly sinusoidal at low bias currents). Maybe this will change when I put the 10R in the gate drive line (source bias, gate drive... what a crazy circuit.) Even though the Q1 is "out to lunch" during the negative drive pulse Q2 osc mode, it needs to be in its socket for the oscs to appear, using the PG50s as Q2. But with the 830s the oscs are easier and more persistent.

I use the upside down air can trick all the time to cool components. Works well on hot glue to cool it off quickly too!
And... for the 830s and the 555, I have simply superglued the components to their heatsinks. No separations yet, even though they've been hot enough to burn my fingertips!  The PG50s of course are bolted down with heat transfer compound, and I have a fan handy.

I don't want to use a LiPo battery for the 555 supply because of the danger of short circuits and superposed spiky crap damaging the battery. Right now I'm running the 555 on a regulated bench supply at 10 V indicated. I think it might be possible to use a couple of 78 series regulators to cobble up a quasi-floating supply that could run the timer off of the TarBaby's running batteries, or perhaps use the 555 to switch a transistor to make the negative going pulse somehow. If you have any ideas I'll be more than happy to hear them and try them out. I think that the best solution would be running from the circuit's own supply, of course.

(Well, actually the BEST solution involves a deep lake, a bass boat, a bucket of readymix concrete, and a six-pack of Shiner Wild Hare, but that comes later...)



(deepbunker, debunker..... get it? I know.. it's weak, and I'm not actually in a basement any more..... oh well.)

[gyulasun]

To your attention:

http://www.youtube.com/watch?v=-c1py6CrgOY

Gyula

[TinselKoala]

Well done, Gyula. I'm glad you posted that...

What did you discover about the 555 circuit? Did you have to make any changes from the original published diagram in the Quantum magazine article? I'd also like to know the resistance actually used in the series gate potentiometer... I always wound up running mine at almost minimum resistance. (You probably know that I did a lot of work with this Ainslie circuit a couple of years ago).

You got very little heat in your load at Ainslie's reported 3.5 or so percent duty cycle. Try flipping the duty cycle exactly inverted to 96.5 percent ON, and see if your resistor heat profile matches what RA published. Mine did....... 

What's the scope app that you are using? Does it use the sound card for input? I've been looking for a good scope app that will run on Linux (and is free... of course.....)

Thanks again, that was a blast from the past !

[TinselKoala]

The "official" schematic for Tar Baby with the 555 timer driver.

(Thanks to GL for the 555 rendering; and thanks for the basic diagram, whoever drew it and released it into the wild...)

(And not showing the 10R that might be inserted into the pin 3 output as discussed above... watch for the next revision if it works out.)

[TinselKoala]

For some reason, this tune reminds me of you-know-who every time I hear it.
http://www.youtube.com/watch?v=tDl3bdE3YQA