RF Osc for Audio Output Bias.

[pwdrive]

I don't think I've ever seen this technique before.

Lawrence.

[Eddystonepete]

Seems like an awkward way to avoid dissipation in a cathode resistor. They could have gone the whole hog and used a cap dropper for the heaters too. That said the live chassis aspect means that it limits the amps usefulness. Further thought, why not tap off the heater chain with a diode to give the neg. bias ? That could be used with a cap. dropper.

Peter

[Kalee20]

Well, it's an AC/DC circuit, and neither a capacitive heater divider, nor tapping off grid bias by a rectifying diode from the heater chain, are techniques available when you operate the thing from DC mains.

As the bias voltage is 32V, this would be a sizeable amount to lose from the HT supply if cathode bias is used. Power output would be compromised.

Earlier this year, 'Radiophile' magazine published a design someone put together for a radio using 12V as power supply (ie 12V HT). I can't be certain of the output valves - EF98's I think, in push-pull, but again, grid bias for these was produced by a self-rectifying oscillator circuit.

[pwdrive]

Yes, it's unusual, the only reason I can think of for that arrangement is so that the output valves can produce the maximum power that is possible with a 240ac derived HT supply, also the -ve bias derived from the oscillator would remain constant possibly.
Yes, a diode etc off the heater chain could provide bias but I'm not sure if silicon were generally available when that circuit was designed.

Lawrence. [EDIT: Whoops....missed Kalee's post...ships in the night and all that, so the above is a reply to Peters last post.