2 Different Output Tubes in 1 Amp

In a VTA ST-120:
Can 2 KT120 tubes in the back and 2 KT88 tubes in the front be installed and function OK? Or is that just asking for trouble?

Dogstar wrote:In a VTA ST-120:
Can 2 KT120 tubes in the back and 2 KT88 tubes in the front be installed and function OK? Or is that just asking for trouble?

This is asking for trouble.

Dogstar wrote:In a VTA ST-120:
Can 2 KT120 tubes in the back and 2 KT88 tubes in the front be installed and function OK? Or is that just asking for trouble?

Theory: As long as the bias is set and correct for both pairs, and the net-current-draw of all filaments is within the capacity of the transformer, there are no electrical issues to worry about.

Reality: As long as you stay within the comfortable power handling envelope for the 'lesser' pair, of tubes, you should not even notice a difference.

I tried it once and nothing bad happened. I could not really detected any major sound differences and did not try it long. But I was listening in lower volume Class A only. In higher volume Class B operation you could have asymmetrical operation where one tube type operates much differently that a matched set of the same type. This could cause an increase odd harmonics.

I tried it once and nothing bad happened. I could not really detected any major sound differences and did not try it long. But I was listening in lower volume Class A only. In higher volume Class B operation you could have asymmetrical operation where one tube type operates much differently that a matched set of the same type. This could cause an increase odd harmonics.

Theory: As long as the bias is set and correct for both pairs, and the net-current-draw of all filaments is within the capacity of the transformer, there are no electrical issues to worry about.

Reality: As long as you stay within the comfortable power handling envelope for the 'lesser' pair, of tubes, you should not even notice a difference.

Thank you for the feedback. Rarely do I have the volume levels cranked up high. But I should make note that once in a while I do and I should be conscientious of have have different tubes in the amp.

My reason for asking is that I now currently have 3 KT120 tubes as one had red plated awhile ago. I was thinking that I could at least use 2 of the 3 to get more tube time our if the remainder. What I should probably do is buy 1 more KT120.

If I have the amp set to triode mode is it running strictly in Class A? If not how do you ensure that the amp is running in only Class A.

Give it a try and see what you think...

Triode mode also operates in Class AB.  But you will get less Class A power in triode versus ultralinear.  You would really need an oscilloscope connected to the speakers to determine class B operation.  I would say that my amp in ultralinear mode, at a modest volume, probably operates in Class A 99% of the time.  A lot will depend upon your system, speaker efficiency, and how loud you like to play.

An audio electronics expert told me a long time ago how to know when an amp is running in AB mode....but not being an electronics guy I forgot.

When the amp is operating in Class AB you will sometimes notice non-linear performance near the zero crossing point as one tube cuts off and the other carries the peak of the waveform.  This is particularly true in SS amps.  A distortion analyzer would be best and and display distortion as power is increased.  If the amp had low distortion until higher power output (but below rated power rating) could be a sign that operation is not linear and perhaps a mismatch of tube characteristics.  Also as tubes age tube matching may not be the same as when new.

http://www.aikenamps.com/index.php/what-is-crossover-distortion

It just doesn't seem like a good idea.  The design of the amp is for ideally a matched pair of the same tube.  While KT88s and 6550s are pretty close the KT120 is a pretty new invention and has different operating points for best results.  Ultimately it's your amp but I would advise against it.

There is no reason you can't use 6550As in the amp simultaneously with KT-120s. The 6550s draw a bit less heater current but it's unlikely the heater voltage will rise enough to be a problem.

Sonically it may be great - or it may not sound so hot. But if the tubes are biased properly the amp is in no peril. As long as both tube types can operate properly in the amp (I doubt a KT-120 and a 6AQ5 would work in the same amp!) then using them together is fine. Whether you like it or not? Who knows??

arledgsc wrote:When the amp is operating in Class AB you will sometimes notice non-linear performance near the zero crossing point as one tube cuts off and the other carries the peak of the waveform.  This is particularly true in SS amps.  A distortion analyzer would be best and and display distortion as power is increased.  If the amp had low distortion until higher power output (but below rated power rating) could be a sign that operation is not linear and perhaps a mismatch of tube characteristics.  Also as tubes age tube matching may not be the same as when new.

http://www.aikenamps.com/index.php/what-is-crossover-distortion

Just a couple of comments I'd like to make...

I'm not aware of any reason that a SS amp would have more tendency to have xover distortion that a tube amp. It all depends on where the tubes or SS devices are biased. There is nothing inherent in SS circuits that would be the cause.

Also the crossover distortion explanation Randall describes is theoretically spot on  - he's a SHARP guy! But in the world of audio amps (and not so much musical instrument amps) I would say an amp with audible or "scopeable" crossover distortion is either in need of service or is a poor design. Simply biasing the tubes a bit deeper into class A (class A means both tubes in a push-pull amp conduct on 100% of the waveform) will minimize or eliminate the problem. Yes, the designer does have to be sure he isn't overstressing the tube by exceeding the tube dissipation ratings. and there are some other design issues involved.

Keep in mind that class AB operation can encompass a wide range of bias points since class A means both tubes conduct all the time and class B means the tubes conduct exactly 50% of the time. So class AB means the tubes conduct more than 50% but less than 100% of the time. And as you can imagine the difference between an amp operating at 50.01% and 99.9% is significant! At 50.01% even a small audio signal applied to the tube grid can drive the output tubes into cutoff; whereas at 99.9% it takes a very large signal to drive the stage into cutoff.

So the odds of experiencing crossover distortion vary depending on how deeply into class A the amp is biased! Remember deeper into class A means FARTHER from class B and the distortion that results.

Jim McShane wrote:

arledgsc wrote:When the amp is operating in Class AB you will sometimes notice non-linear performance near the zero crossing point as one tube cuts off and the other carries the peak of the waveform.  This is particularly true in SS amps.  A distortion analyzer would be best and and display distortion as power is increased.  If the amp had low distortion until higher power output (but below rated power rating) could be a sign that operation is not linear and perhaps a mismatch of tube characteristics.  Also as tubes age tube matching may not be the same as when new.

http://www.aikenamps.com/index.php/what-is-crossover-distortion

Just a couple of comments I'd like to make...

I'm not aware of any reason that a SS amp would have more tendency to have xover distortion that a tube amp. It all depends on where the tubes or SS devices are biased. There is nothing inherent in SS circuits that would be the cause.

Also the crossover distortion explanation Randall describes is theoretically spot on  - he's a SHARP guy! But in the world of audio amps (and not so much musical instrument amps) I would say an amp with audible or "scopeable" crossover distortion is either in need of service or is a poor design. Simply biasing the tubes a bit deeper into class A (class A means both tubes in a push-pull amp conduct on 100% of the waveform) will minimize or eliminate the problem. Yes, the designer does have to be sure he isn't overstressing the tube by exceeding the tube dissipation ratings. and there are some other design issues involved.

Keep in mind that class AB operation can encompass a wide range of bias points since class A means both tubes conduct all the time and class B means the tubes conduct exactly 50% of the time. So class AB means the tubes conduct more than 50% but less than 100% of the time. And as you can imagine the difference between an amp operating at 50.01% and 99.9% is significant! At 50.01% even a small audio signal applied to the tube grid can drive the output tubes into cutoff; whereas at 99.9% it takes a very large signal to drive the stage into cutoff.

So the odds of experiencing crossover distortion vary depending on how deeply into class A the amp is biased! Remember deeper into class A means FARTHER from class B and the distortion that results.

Jim, correct me if I'm wrong here, but matching bias of tubes is also dependent on plate voltage presented to each tube. So if the plate voltages don't match up for whatever reason, then the bias settings would need to be adjusted accordingly.

deepee99 wrote:Jim, correct me if I'm wrong here, but matching bias of tubes is also dependent on plate voltage presented to each tube. So if the plate voltages don't match up for whatever reason, then the bias settings would need to be adjusted accordingly.

You are correct sir- and even more variance would be caused by different screen voltages.

Jim McShane wrote:

deepee99 wrote:Jim, correct me if I'm wrong here, but matching bias of tubes is also dependent on plate voltage presented to each tube. So if the plate voltages don't match up for whatever reason, then the bias settings would need to be adjusted accordingly.

You are correct sir- and even more variance would be caused by different screen voltages.

Yeah, forgot about the screen volts. Thanks, Jim.
I think output tube-swapping in an ST-120 is just asking for trouble. In the M-125 Monoblocks, a quad of matched outputs of different flavours for each amp might be a fun experiment, just to see which suits you best.
Put on a Phil Specter album, all of which were recorded in mono, and if there's a difference you'll hear it for sure.