a push/pull tube o.d. for the summer...

[samhay]

I could never get LTP or self split (basicly the same thing) to work at low voltages. I'm sure it can be done but some kind of trickery is probably needed. That's why i chose a transistor Cathodyne pi. An inverting opamp solution would work too. But again, the simplicity of a single transistor and two resistors lured me in. .
If Samhay get it to work it opens up a world of new possible 9 volt tube designs. ..Very exciting. .

This seems to work on the breadboard and gives vaguely balanced output from the two halves of the LTP.

However, I don't think I am anywhere near the best bias point yet - most of the resistor values are not too criticial, which is a worry - and the gain is low. This needs some big improvements, but is a start.

[samhay]

Had a bit more of a play with the LTP version on the breadboad and have updated the schematic in the previous post.
Not sure if it is as verstatile as the cathodyne version, but when the two halves of the LTP are balanced, it sounds very nice and has good overdrive character.
Still need to play with it some more and tack on a tone control.

[Johan]

I see you liked my heater c.t bias trick. :-)
Do you get any gain from the ltp? It would seem like that common cathode resistor is unusually large compared to that plate resistors....But then again, i never got ltp to work at this low voltages..am happy with how my last version turned out and plan on boxing it up in a few weeks when i have a little more time..

[samhay]

Yes - the heater tap is a nice idea. Only problem is the resistance is low, so we may need a very large resistor to keep power supply noise down.

I think the gain is very roughly unity from each side of the LTP (based on listening), so you get about a gain of 2 when you mix them. It might be a bit more, but I haven't had a good look with a scope yet.
I needed to increase the cathode resistor and signifcantly decrease R7 in order to get the grid negative and the two halves reasonably balanced. I don't really know what I am doing, so it may be a long way from optimum. However, I would imagine that the BJT PI makes more sense here as it gives you more freedom to play with the gain.

[samhay]

Had a look at the gain of the LTP circuit. With the drive at minimum (input op-amp gain = 1) the gain is about 0.6!
It does overdrive very nicely, with pretty good cancellation of second harmonic. Some interesting crossover distortion when the drive pot is turned up.

I think we can put that design to bed now.