How is bias tremolo "different"?

[Mark Hammer]

I ask this same question every now and then, and I never get a very good answer, so it's time to ask again.

If you've ever played through an amp that has tube bias tremolo, as opposed to LDR-based or other types of electronic tremolo, there is a very different feel.  Stompbox tremolo can sound like someone doing deep-knee bends or squats: up-and-down-and-up-and-down. It's not uninteresting, but if you've played tube bias tremolo, that method sounds like someone stretching and collapsing, instead of simply standing up and sitting down.  My Tone Core Tap Tremolo has a "bias" setting on it, which I like very much.  I can't speak to whether other commercial pedals capture this in a similar or even better fashion.  All I know is that I like it and prefer it to any of the other solid-state tremolo units I have.

So what the heck is it that is different about tube bias tremolo?  I don't mean in an electronic sense, but in terms of what happens to the signal?  My hunch is that, as something relegated to the output stage, there are implications for what happens to output tube coloration.  In some respects, this is analogous to the difference between what Doppler sounds are like when fed into an amp (via a flanger, chorus, or even modelled device), versus mapped onto the amp's output by a rotating speaker cab.

I'm a scientist by training, and principle #1 in science is that you have to be able to describe before you can explain, then predict, then reproduce and control.  So, could any of you whizzes out there voice an opinion about what you think goes on with a tremolo-processed signal in the tube-bias context that can make it sound and feel different?  What are the audible dimensions that, for you, change?  Or, as Eric Idle would say "You've, you know, BEEN with a tube-bias tremolo, eh?  Nudge-nudge-wink-wink, say no more, say NO - - MORE.  ...... What's it like?"

[BlueToad]

I'm hardly a whiz compared to some of the people on the board, but two thing strike me. First, the phase shift oscillator in an amp with photocoupler is followed by a gain stage to drive to drive the photocoupler's lamp. This is probably going to add a degree of distortion to the oscillation. Could the oscillator signal be closer to a pure sine wave in a bias trem?
            Second, with the bias tremolo the only signal change is going to be at the power tubes. With the photcoupler tremolo, the oscillator tends to be connected before the phase inverter. Therefore with a bias tremolo the phase inverter is less affected, and is going to distort or whatever as normal. The photocoupler reduces the signal input to the phase inverter, and consequently any distortion it might produce. Considering that the pentode/beam tetrode output tubes are going to make more 3rd harmonic distortion, while the triode phase inverter tubes are going to produce mostly 2nd harmonic distortion. So with the bias tremolo, you are disproportionately reducing the third harmonic distortion, while keeping more of the second harmonic distortion. This is going to be more pronounced in push-pull amps, b/c they don't produce even harmonics (according to Mr. Keen) like the AC-15 or Princeton. I can't think of a single ended amp which uses photocoupler tremolo, so comparing that could be hard.
            It's entirely possible that I am speaking garbage, and if that is the case I'd love to be corrected.
Carl

[MartyMart]

they do seem to sound more "sine wave LIKE" to me too .... could be wrong !
Perhaps "smoother" and less "pulse-like" is a better description, I agree BTW that the
tap trem's "bias" setting is superb ( as is the Space chorus's "Tri Chorus" setting )

[R.G.]

Well, in addition to modulating gain, they modulate the amount of crossover distortion in the output stage.

I think they also modulate the % distortion in a single stage which is bias-trem modulated as well.

You have to have a pretty small signal to not be affected by some of the tube-trem bias changes.