Mu Follower with a fixed bias upper triode

[SoulFetish]

Whenever I Mu-Follower circuits, they are either configured using cathode biasing or grid leak biasing for the upper triode (In my experience). However, is there any reason that it wouldn't work with a fixed bias as shown in the first (A) drawing below? Both are driven by capacitive coupling via the bottom triode plate voltage? Maybe I'm missing something

(The examples below use 6111 tubes in a realistic example with voltages listed at the respective nodes)

[Chuck H]

Both are driven by capacitive coupling via the bottom triode plate voltage?

Yes? But when you say it like that it sounds wrong to me. When someone says "plate voltage" I think of the supply voltage. But of course it's voltage modulation on the lower triodes plate that is allowed through the capacitor to the upper triodes grid that is driving the upper triode.

And I don't see any reason the circuit couldn't be fixed bias as you've shown (with the exception that you may need to adjust the bias voltage divider circuit values). But I do wonder if there's any advantage to it. I'm far from expert on this but there's so little voltage bias that any amount of shift in the cathode bias arrangement is likely insignificant. And it may be possible that the cathode bias arrangement allows for better operational condition tracking with HT sag.?. Not sure about this either but I'm just thinking out loud. I do know you CAN bias it as shown in your schematic without a problem.

What is your goal in changing the circuit to fixed bias?

[Helmholtz]

To me plate voltage means (momentary) voltage at the plate, so DC + signal AC (except for CF).
Shouldn't be confused with plate supply voltage or "battery voltage" B+.
Here we're speaking of plate signal (voltage).

Fixed bias should work, but will worsen PSRR and make the circuit more dependent on B+ fluctuations.

Also you will lose the bootstrapping of the upper grid resistance. So in order to not attenuate bass even more, coupling capacitance will need to be larger.

[SoulFetish]

Great points. Just a design exercise, really.
But I don’t see any advantage to a fixed arrangement.

[Helmholtz]

The µ-follower is discussed in Merlin's HiFi book (highly recommended!).

[Enzo]

Agree with Helms, "plate voltage" is voltage at the plate. The power supply is as stated "plate supply" or I prefer B+ or even HV rail.

[Chuck H]

Well that's why I said it "sounds wrong to me". I do agree that 'plate voltage' is a mutually exclusive designator but easily nine out of ten times when someone refers to 'plate voltage' they ARE talking about the B+ DC voltage. I didn't have any trouble working out what was meant but I did need a moment for interpretation.

P.S. I spice modeled the circuit above (with a higher value coupling cap) and compared it to the cathode bias arrangement. It seems to work just fine with no significant difference in behavior or performance. But I can't be sure I looked for all the right things.

[Helmholtz]

P.S. I spice modeled the circuit above (with a higher value coupling cap) and compared it to the cathode bias arrangement. It seems to work just fine with no significant difference in behavior or performance. But I can't be sure I looked for all the right things.

Did you add a "ripple source" to the supply and did you measure frequency response?

[SoulFetish]

Merlin seems to include a portion of ripple voltage at the grid of the bottom triodes in examples of totem pole circuits in his Hi-Fi book - usually dialed in to provide optimum phase cancellation at the upper triode grid (via the bottom triode plate voltage )

[Helmholtz]

Merlin seems to include a portion of ripple voltage at the grid of the bottom triodes in examples of totem pole circuits in his Hi-Fi book - usually dialed in to provide optimum phase cancellation at the upper triode grid (via the bottom triode plate voltage )

Yes, at the grid of the bottom triode.

Cancellation requires out-of-phase signals.
The lower tubes inverts signal phase - the upper not as it's a CF.

[Chuck H]

Did you add a "ripple source" to the supply and did you measure frequency response?

I did not. I only viewed sine wave activity and amplification at similar bias conditions. Unfortunately I just plugged the circuit into an existing model and then didn't save it at the end. But I'll get around to redoing it tonight and repost tonight or tomorrow morning.

[SoulFetish]

Yes, at the grid of the bottom triode.

Cancellation requires out-of-phase signals.
The lower tubes inverts signal phase - the upper not as it's a CF.

Correct

[Chuck H]

Did you add a "ripple source" to the supply and did you measure frequency response?

Did you want ripple injected at the cathode (I think) or the B+ supply. I tried injecting ripple at the B+ but even adding 2ohms of ESR to the filters I wasn't seeing any significant ripple past three filter nodes so there was no significant ripple at the mu follower to make a comparison.

The schematic in question has a partially bypassed cathode. but it seems to me that a fully bypassed cathode would help reject ripple in the filament supply. I think Merlin used a diode in the cathode circuit of the lower triode to do this in his example?

[SoulFetish]

Ripple would exist in a DC filament supply.

[Helmholtz]

Did you want ripple injected at the cathode (I think) or the B+ supply. I tried injecting ripple at the B+ but even adding 2ohms of ESR to the filters I wasn't seeing any significant ripple past three filter nodes so there was no significant ripple at the mu follower to make a comparison.

My point was about power supply ripple rejection, defined as plate supply ripple divided by corresponding ripple at circuit output.
It seems obvious that the bias divider adds 2/3 of the supply voltage ripple to the output.
The idea was to add a 120Hz ripple source (say 10mV) in series with B+ in simulation.
Dunno if that's possible with LTSpice or if all sources need to be grounded.

The PSRR result may not be a problem with a well filtered supply.