Channel imbalance from 6c45 tubes

[MikeSattler]

Hey all, I've been breaking in some new 6c45 tubes in my mainline for the past few weeks and lately I've noticed a slight channel imbalance. Nothing major, but the stereo image is very slightly skewed to the right.

Out of curiosity, I played a test tone through the amp with the volume at max and measured the AC voltages on the tip/ring connection. Sure enough, the ring measured about .1V higher than the tip. Swapped the 6c45 tubes and the .1v difference is on the opposite channel now. So, it looks like one tube is outputting a slightly weaker signal. I made sure both channels are biased properly but that didn't seem to change anything. Do I need to be looking for a matched set of tubes or is there an easier way to get both output voltages more in line?

[Paul Birkeland]

When you're measuring the 0.1V difference, what are the actual voltages?  If you have 5V and 5.1V, that's 0.17dB and not that big of a deal (pretty much inaudible).

If you have 0.3V on one channel and 0.4V on the other, that's 2.5dB of difference and that would be worth addressing. 

[MikeSattler]

Hey Paul, thanks for the quick reply!

For extra context; I'm running a 60hz tone from a Chord Mojo at it's line out setting which is 3V I believe. With the amp set a tad higher than my normal listening volume I'm getting the .1v difference: .61 for the left channel and .72 for the right.

At max volume the voltages are 7.98v and 8.67v.

[Paul Birkeland]

I would let the amp run with that test tone going for a few days to give the tubes a chance to cook in.  For whatever reason, Russian tubes tend to not get a lot of break in hours at the factory, and we find that they will somewhat dramatically change their operating parameters after everything has gotten some time to cook in. 

[MikeSattler]

Okay cool I'll just let them run for awhile. Just out of curiosity, I read in some other parts of the forum here that matched tubes aren't required and I was wondering how the circuit design accounts for the variations between tubes? I had figured that was what biasing was for but it looks like that doesn't affect the tube gain at all?

[Paul Birkeland]

Tube matching is really helpful when you have something like a long tail pair phase inverter with a resistor under the tail that will only perform its best when each triode is doing the exact same thing.  With an amp like the Mainline, small differences in gain are stepped down by the output transformer rather than being further amplified by additional stages, so they end up not being particularly audible. 

[MikeSattler]

Okay great, thanks for the info!