Output tube plate resistors

[Tdale]

Do most of you guys use a 10K trim pot for the plate resistors on the output tubes, as shown in many schematics? Is the pot there for acurate adjustment?

I also see some using different values on the plate resistors on the output tubes, why is that?

Tommy

[dobbhill]

The trimpot is there for AC Balance of the output section. Someone please correct me if there are errors:

How to set AC balance using the trimmer between the PI plates:
Hook up a suitable dummy load to the amp (or get out your earplugs!).
Apply a sine wave signal to the amplifier input that gives a reading of about 20 volts AC measured at the point where the bias voltage is applied to the output tube grids. I like using a 400Hz tone, but others may do. Some people use pink noise. I stay away from higher tones, as they may have more harmonic content, and 400Hz is within the guitar's range.
Adjust the trimmer until the voltage is equal on both output tube grids. I like 20volts or a little less because my Fluke gives 2 decimal points up to 20volts.
You have just set the AC balance.
I have heard from people who hook up the output of the transformer to a scope and adjust the balance until the positive and negative halves of the waveform are equal in voltage. That method looks at the PI, output tubes, and transformer as a complete system. Overkill, IMHO, but some think AC balance controls are overkill.
This insures that the output tubes are both being driven by the same voltage (symmetrical drive).

Plate resistors on the output tubes???
D

[Bob-I]

Do most of you guys use a 10K trim pot for the plate resistors on the output tubes, as shown in many schematics? Is the pot there for acurate adjustment?

I also see some using different values on the plate resistors on the output tubes, why is that?

Tommy

You sound a little confused. There are no plate resistors on the output tubes. You must be thinking of the plate resistors on the PI.

The goal of the trim pot on the PI plates is to balance the output of the 2 plates. I personally don't hear the difference, but some folks do. Asl dogears, he can hear the difference.

[Tdale]

Damn! I didn't mean the output tubes, I of course meant the PI tube, my mistake

The schematics show a trim pot for the plates on the PI tube

Anyway, I guess if I use two accurate resistors, it shouldn't be necassery with that pot.. I se several boards are made without the trim pot.

Tomy

[dobbhill]

It's not the resistors that are "out of balance," it's the two halves of the PI tube. They do not amplify exactly the same.
D

[glasman]

An interesting side effect I noticed a while back while adjusting the trimmer on one of my amps. I use the scope on the output method personally. As you increase the signal into the amps you will start to hear the the output transformer start to squeal at the signal generator frequency. The squeal was minimum when I brought the clipping to a symmetrical setting. Adjusting one side or the other from nominal and the squeal would increase.

Gary

[VacuumVoodoo]

There are several opinions as to how the phase inverter symmetry should be adjusted.

Adjusting for equal signal amplitude at the PI plates leaves two other stages of the power amp out of the chain i.e. power tubes and output transformer. Even if you have a "matched pair" of tubes they're usually matched for plate current at one specific operating plate voltage, not many suppliers match also transconductance.
This means that the transconductance most probably will not match at the bias point of your choice so equal amplitudes at the grids will result in different amplitudes on the plates. Add to this the fact that output transformers aren't perfectly symmetrical either. You might get lucky in that the OT's non-symmetry will cancel tube mismatch but don't rely on luck.

So, if your goal is perfect symmetry you should adjust for symmetrical onset of clipping of the waveform at the speaker output.

My opinion is that "symmetry trim" should be done by ear (yes, i have an oscilloscope and a spectrum analyzer and what-not ) - I 'm pretty sure "Dogears" will agree with me that what looks pretty on oscilloscope screen doesn't always sound good in our ears.

[VacuumVoodoo]

An interesting side effect I noticed a while back while adjusting the trimmer on one of my amps. I use the scope on the output method personally. As you increase the signal into the amps you will start to hear the the output transformer start to squeal at the signal generator frequency. The squeal was minimum when I brought the clipping to a symmetrical setting. Adjusting one side or the other from nominal and the squeal would increase.

Gary

Symmetrical clipping introduces 3rd harmonic while asymmetrical gives you 2nd harmonic. The squeal you hear is mechanical vibration of windings and core, it's easier to move the winding at twice the fundamental frequency than thrice. Therefore less squeal when clipping symmetrically. I have yet to find an OT that will not squeal when output waveform clips.

[glasman]

An interesting side effect I noticed a while back while adjusting the trimmer on one of my amps. I use the scope on the output method personally. As you increase the signal into the amps you will start to hear the the output transformer start to squeal at the signal generator frequency. The squeal was minimum when I brought the clipping to a symmetrical setting. Adjusting one side or the other from nominal and the squeal would increase.

Gary

Symmetrical clipping introduces 3rd harmonic while asymmetrical gives you 2nd harmonic. The squeal you hear is mechanical vibration of windings and core, it's easier to move the winding at twice the fundamental frequency than thrice. Therefore less squeal when clipping symmetrically. I have yet to find an OT that will not squeal when output waveform clips.

But this assumes that you drive the power amp into distortion. With a d-type amp (at least my opinion) I don't want (or really need) power amp distortion. These amps are meant to sing not crunch. Of coure IMHO....

Gary

[VacuumVoodoo]

An interesting side effect I noticed a while back while adjusting the trimmer on one of my amps. I use the scope on the output method personally. As you increase the signal into the amps you will start to hear the the output transformer start to squeal at the signal generator frequency. The squeal was minimum when I brought the clipping to a symmetrical setting. Adjusting one side or the other from nominal and the squeal would increase.

Gary

Symmetrical clipping introduces 3rd harmonic while asymmetrical gives you 2nd harmonic. The squeal you hear is mechanical vibration of windings and core, it's easier to move the winding at twice the fundamental frequency than thrice. Therefore less squeal when clipping symmetrically. I have yet to find an OT that will not squeal when output waveform clips.

But this assumes that you drive the power amp into distortion. With a d-type amp (at least my opinion) I don't want (or really need) power amp distortion. These amps are meant to sing not crunch. Of coure IMHO....

Gary

Got your point Gary, need to clarify: adjusting for symmetrical onset of clipping is easy to do visually using the oscilloscope. Unclipped sinewave can also be asymmetrical and by definition distorted but more difficult to observe on the scope.

One way to achieve very nice PI signal symmetry is to use a current source in the PI tail and equal plate resistors. But that will no longer be a D-type amp...

Shouldn't this topic be moved to "Technical discussions" ?

[greiswig]

(SNIP)
I have heard from people who hook up the output of the transformer to a scope and adjust the balance until the positive and negative halves of the waveform are equal in voltage.

Thanks for this. As usual, I need clarification: if I'm looking at the output of the transformer into a dummy load, unequal positive and negative swings would just look like DC offset, wouldn't they?

[heisthl]

I've noticed some amps sound better when you use your ears to purposely unbalance the PI (more bloom and overtones)

[greiswig]

I've noticed some amps sound better when you use your ears to purposely unbalance the PI (more bloom and overtones)

I know, Henry. I'm just trying to find my starting point. I figure "out of balance" is easy to hit. I'm pretty sure that I can hit that without trying. The right AMOUNT of out of balance, in the right direction, is a different story. So I wanted to try to find the balance point first, then start adjusting in either direction from there.

So I have an old 2-channel scope, and I can set it up so I can have both probes on one side of the PI output, put a 1kHz sine into the clean input, and can match the waveforms with the display adjustments. I then put one probe on the other output, invert it, and see if the waveforms still match. Then I try tweaking the PI trimmer, and I'm damned if I can see anything changing no matter how far to one side the trimmer gets adjusted. What am I doing wrong?

[heisthl]

I'm damned if I can see anything changing no matter how far to one side the trimmer gets adjusted. What am I doing wrong?

You're using the word "see" instead of "hear" Just play one note and let it ring, then a half step higher, let it ring, then a half step higher yet, let it ring - adjust the trimmer, repeat the 3 chromatic notes, you will find a spot where all three sustain/bloom/octave bloom the best. I like notes on the B string(3,4 and 5th fret), I think Dogears likes to listen to low notes. Do it right and complex chords have a natural chorus.

[greiswig]

I'm damned if I can see anything changing no matter how far to one side the trimmer gets adjusted. What am I doing wrong?

You're using the word "see" instead of "hear" Just play one note and let it ring, then a half step higher, let it ring, then a half step higher yet, let it ring - adjust the trimmer, repeat the 3 chromatic notes, you will find a spot where all three sustain/bloom/octave bloom the best. I like notes on the B string(3,4 and 5th fret), I think Dogears likes to listen to low notes. Do it right and complex chords have a natural chorus.

Well, I've been trying to see something because I haven't been able to hear anything. I get the feeling this is something that I have to watch someone with experience do, then I'll understand it. Your explanation is good, though, and different from what I have seen elsewhere. Do you do this on the OD channel, the clean channel, or what? How loud does it have to be...edge of feedback already?