• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Common mode feedback

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
dirkwright

dirkwright

Member
Joined 2011
smoking-amp said:
"Gp P K VALUE={1.147E-6*(PWR(V(2),1.5)+PWRS(V(2),1.5))/2}"

Yep, 3/2 power modeling. (the 1.5 powers, not sure why its got PWR and PWRS though, seems to be averaging them) With constant Mu (the 95.43 above).

Might be interesting to vary the Z of the CCS to see if the distortion is maximized at some specific Z or just at the inf. Z.

I wonder if this common mode feedback scheme would lower distortion for a diffl. driver and output stage combo.

The usual scheme for minimising 3rd harmonic is to use a specific Z in the tail of a diffl. pair. But this common mode fdbk scheme seems to be doing the opposite. Maybe the 2nd stage is doing the distortion comp. by being dragged by the 1st stage? Would be interesting to put a pure sine wave in and observe the 3rd harmonic at the output of the 1st diffl. stage and at output of the 2nd diffl stage to see if the 3rd H distortion % first goes up then back down to cancel. This could be very significant if it can be used to cancel odd H distortions.
Click to expand...

The only choices for the CCS impedance is infinity or 1M ohm. I suppose I could construct something more real to see if there's a difference.

The common mode feedback is only for the common mode signal, as I understand it. The differential signal is not affected.

One thing about differential amps is that the distortions become compounded after each stage. If a first stage has both differential and common mode signal outputs because the common mode rejection ratio for that stage is not really high, then the following stage amplifiers both signals and produces distortions off of each kind. The common mode signals become differential mode distortions, in other words. Common mode feedback is helpful, as I understand it, by reducing the effect of the common mode signal on distorting the differential mode signal. I assume it consequently increases the CMRR for the whole amplifier as well.

Also, let me know if you find a better spice model for these tubes.
 
smoking-amp

smoking-amp

Member
Joined 2001
I know I've seen some cathode to cathode (across subsequent SE stages) feedback before. But it seems to be positive feedback for 2 stages. How is that done to get neg. fdbk? Was it over 3 stages?

------------------

"Common mode feedback is helpful, as I understand it, by reducing the effect of the common mode signal on distorting the differential mode signal."

You saw an improvement of 10X in the differential distortion, correct? (was that with a superimposed common mode level or signal at the input?)

One could use a center tapped inductor load for the 1st stage to remove common mode signal. Would that be just as, or nearly as, effective in lowering the final diffl. mode distortion? Maybe this is the reason some prefer interstage xfmr coupling.

------------

You can just put a series RC across the CCS to lower the impedance (big C).
 
Last edited:
dirkwright

dirkwright

Member
Joined 2011
I have been fooling around with using common mode feedback (CMF) in a power amp without much luck. I seem to be stuck at pretty much the same point as I was 20 years ago when I tried to build a tube power amp using CMF but all I made was an oscillator. I gave up and sold them.
 
dirkwright

dirkwright

Member
Joined 2011
smoking-amp said:
I know I've seen some cathode to cathode (across subsequent SE stages) feedback before. But it seems to be positive feedback for 2 stages. How is that done to get neg. fdbk? Was it over 3 stages?

------------------

"Common mode feedback is helpful, as I understand it, by reducing the effect of the common mode signal on distorting the differential mode signal."

You saw an improvement of 10X in the differential distortion, correct? (was that with a superimposed common mode level or signal at the input?)

One could use a center tapped inductor load for the 1st stage to remove common mode signal. Would that be just as, or nearly as, effective in lowering the final diffl. mode distortion? Maybe this is the reason some prefer interstage xfmr coupling.

------------

You can just put a series RC across the CCS to lower the impedance (big C).
Click to expand...

The thing is this: if you apply ONLY a common mode signal to a differential amplifier, it will produce BOTH common mode and differential mode output signals. If you apply ONLY a differential mode signal to a differential amplifier, it will produce BOTH common mode and differential mode output signals. The common mode signals appear at the joined cathodes, as I understand it. I'm no expert. I don't understand this well enough to make too many more comments.
 
smoking-amp

smoking-amp

Member
Joined 2001
I don't see any cross coupling in posts #1 or #5, but that (cross coupling) would fix the differential pos. fdbk (to neg. fdbk) only.

Hmm, I don't think it (the cross coupling) even fixes the diffl. mode if the tails are all connected together. Doesn't make any difference what side a phase is on then.

Maybe you are referring to a diagram without the common tails?
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom