matched power tube question

[smolder]

I know they did not bother with this 'back in the day', but is matching up power tubes less of a concern with cathode bias amps than with fixed bias?

 

[Blue Strat]

No, same concerns as with fixed bias amps.

 

[Beam Tetrode]

I have a pair of 6L6 tubes that are slightly mismatched. The amp is a stock 65 Super Reverb. Considering that most phase inverter circuits do not produce perfectly balanced signals, which socket should I put the hotter tube in?

 

[Blue Strat]

I have a pair of 6L6 tubes that are slightly mismatched. The amp is a stock 65 Super Reverb. Considering that most phase inverter circuits do not produce perfectly balanced signals, which socket should I put the hotter tube in?

The PI balance isn't really relevant. How mismatched are the power tubes? Anything within 5mA is more than close enough. Try switching positions and listening. If one way sounds better than the other that's the right positioning.

 

[TweeDLX]

is matching up power tubes less of a concern with cathode bias amps than with fixed bias?

Think of it this way: You can use mis-matched tubes, IF...you set the bias for optimum performance from the tube drawing the highest current. Now you have one tube working well, and one ...not as well. The issue is when one tube is drawing too much current at whatever your cathode bias setting is and red-plates. Then, you have one happy tube, and one very feverish tube. That make sense?
An option would be to split the bias and give each tube its own (different) cathode resistor/cap combo.

 

[donnyjaguar]

I've found that paraphase invertors are normally pretty well balanced.

This is easy to test with a dual-trace oscilloscope with the amplifier dismantled. Plonk one probe on each half of phase invertor, remove negative feedback and inject a 1kHz tone at some level that won't drive the output into compression (or blow your ears off your head). Set the 'scope to A-B or simply overlay the traces.

Mind you, if you go to this extent its easy to tweak the resistor values so the outputs will be absolutely identical in amplitude anyway.

 

[pdf64]

The phase splitter balance and the output tube bias matching are separate issues.

An option would be to split the bias and give each tube its own (different) cathode resistor/cap combo.

That's by far the best way forward - even quite miss-matched power tend to balance out with similar cathode currents with this arrangement.
But the bypass caps need to be much larger (x10, eg =>220uF), as there's no 'shared cathode cancellation'. Pete.

 

[phsyconoodler]

Quote:"But the bypass caps need to be much larger (x10, eg =>220uF), as there's no 'shared cathode cancellation'. Pete."

Not really.I go the other direction with good results.For example if the amp had one cathode resistor and a 50uf bypass cap I am quite comfortable using two 25uf bypass caps.And they sound just fine with 25uf or all the way up to 1500uf. I recently cathode biased a paraphrase style amp with a 1000uf cap.Sounds delicious.

 

[Ronsonic]

No, same concerns as with fixed bias amps.

This is opinion and not techno-gospel, but IMO, matching tubes is even more important with a cathode biased amp than with adjustable fixed bias.

Let's use an extreme example, say one tube is dead and the other normal. In a fixed bias amp all that happens is you just run on one tube. It's distorted and low power but that's all. In a cathode biased amp the working tube overheats, melts down and takes stuff with it.

Sorta like the difference between having a big man and a weak one lifting a log. If they're both on solid ground they'll work it out. Cathode bias is more like putting them on a teeter totter. Differences are amplified over the fulcrum of that bias point on the cathodes. (wow, that was one tortured metaphor).

 

[donnyjaguar]

Splitting the cathodes on a cathode biased output stage isn't a good idea. Regardless of how much extra capacitance you put on the divorced cathodes it won't be as powerful.

 

[Tone_Terrific]

To check mismatch, I check voltage drop over the primary of the OT, for each tube.
Swap them around, and go with whichever way is more even.