Rebiasing Power Tubes After SS/Tube Rect Switch

[Theashe]

Bit of a beginner question here. In a cathode-biased amp with a power supply having switchable SS/tube rectification, if the power tubes have been biased to 70% dissipation assuming SS mode voltage, would it be a good idea to design a circuit to re-bias the tubes back to 70% dissipation after switching to tube rectification and losing some plate supply? I read a post in a different forum a long time ago where a user spoke about switching from SS to tube rectification in cathode bias making the amp lose tone because it was significantly changing the bias. I imagine some volume will be lost regardless due to the lower plate supply, but it might be a good idea to try to keep as much else the same.

With that in mind, I thought of a circuit that uses a 4P2T switch to control both the rectifier selection and change Rk for the power tubes in an amp I'm designing. This way the Rk's could be selected to maintain 70% dissipation. Please see the attachment. A few more poles on the switch, and you could also change the screen dropping resistors simultaneously to try to bring the current and volume back up.

[pdf64]

That's probably a good idea.
However, why do you perceive the need to tweak the bias for 70% plate dissipation? That guideline doesn't apply to cathode bias.
Due to its self regulating nature, cathode bias can tolerate higher dissipation / hotter conditions, as the increase in positive grid current this may cause (which would cause runaway in fixed bias) gets largely compensated for by the bias voltage Vg-k increasing.
Confirmation of this can be seen in tube info, ie higher values of grid circuit resistance can generally be tolerated in cathode bias than fixed bias.
My finding is that the key thing with cathode bias is to operate at or close to class A, as per the classic designs.
This means that plate dissipation reduces with signal, hence static conditions of up to 100% of the limiting value of plate dissipation can be tolerated.
It helps to ensure good tone, by limiting the squish / bias shift at high signal levels, which, if excessive, results in crossover distortion (which to me sounds more unpleasant in cathode bias than fixed).
See Aiken's thoughts on cathode bias / plate dissipation http://www.aikenamps.com/index.php/the- ... on-biasing

[martin manning]

The 70% rule doesn't really apply to a cathode biased stage, so it might be better to set it up to run 95-100% with the SS rectifier and let it back down a bit when the plate voltage drops on the vacuum rectifier. Since you're dropping all voltages at the rectifier, the screens will track the plates and maintain more or less the same match. No need to worry about that.

There was a thread a while ago where I helped a guy set up a SS-vacuum rectifier switch with resistors to reset the bias on a 5F6-A circuit (fixed bias). He recorded some clips to compare the two, and it was very hard to hear any difference.

[Theashe]

That's probably a good idea.
However, why do you perceive the need to tweak the bias for 70% plate dissipation? That guideline doesn't apply to cathode bias.
Due to its self regulating nature, cathode bias can tolerate higher dissipation / hotter conditions, as the increase in positive grid current this may cause (which would cause runaway in fixed bias) gets largely compensated for by the bias voltage Vg-k increasing.
Confirmation of this can be seen in tube info, ie higher values of grid circuit resistance can generally be tolerated in cathode bias than fixed bias.
My finding is that the key thing with cathode bias is to operate at or close to class A, as per the classic designs.
This means that plate dissipation reduces with signal, hence static conditions of up to 100% of the limiting value of plate dissipation can be tolerated.
It helps to ensure good tone, by limiting the squish / bias shift at high signal levels, which, if excessive, results in crossover distortion (which to me sounds more unpleasant in cathode bias than fixed).
See Aiken's thoughts on cathode bias / plate dissipation http://www.aikenamps.com/index.php/the- ... on-biasing

That's an excellent point to raise, and to be quite honest I didn't know that about cathode biasing before this. I used 70% in my example as a commonly quoted figure, but it's not necessarily a goal. I just want to see what more experienced users think of my idea of rebalancing the cathode bias/possibly the screen voltage when switching from solid state to tube rectifier, to prevent possible loss of volume or tone.

[Theashe]

There was a thread a while ago where I helped a guy set up a SS-vacuum rectifier switch with resistors to reset the bias on a 5F6-A circuit (fixed bias). He recorded some clips to compare the two, and it was very hard to hear any difference.

Maybe it would be a good idea for me to experiment with different biasing resistors until I get two distinct tones and volumes that I'm happy with when switching from SS to tube recto. The point of having the switch is to get the best character from both worlds, not to have them sound the same, so I suppose it's all in the ear to a certain degree.

[rp]

On my 5F6A clone I'll plug in a SS thingy now and then, when compared to a real Mullard GZ34 it's pretty much the same plate voltage, which is pretty cool. Granted they are expensive now but if you have one this would be the place to use it.

[Theashe]

On my 5F6A clone I'll plug in a SS thingy now and then, when compared to a real Mullard GZ34 it's pretty much the same plate voltage, which is pretty cool. Granted they are expensive now but if you have one this would be the place to use it.

Unfortunately, I'm sitting on a bunch of 5U4GB tubes and nothing else. There'll definitely be some drop through one of those.