What to do with one of a pair of matched tubes

[d. spree]

I just had to replace the output tubes on my tweed deluxe - one was bad. They are JJ 6V6s from Tubedepot and were a matched pair.

I have a couple of questions: how can I test them - other than putting them in the amp, and what do I do if one of them is good? I presume the amp needs a matched pair - although I've also read that this may not be super important on this model amp.

Do I end up junking both of them or is there some was I can use one or find a match for it? Its not the cost so much (a pair is probably less than $30 with postage, I just hate to throw away a perfectly good tube.

[pdf64]

Best thing would be to hook up with someone that has a Champ or other single ended 6V6 amp.
Or just keep it safe as a spare for a rainy day.
For tube life, I think your amp would benefit from having 6V6 tubes that were a good match with each other. Reason being that both tubes, if evenly balanced, are working pretty much at max dissipation. If the tubes aren't evenly balanced then the hot one may be operating in excess of its max dissipation and so will wear out quickly.
Tube are a regular frequent consumable, with that amp especially, so you could soon build up a stock of odd ones that a tech may be able to match up.

[d. spree]

Thanks for the feedback - appreciate it.

[km6xz]

Put it in an amp and measure the characteristics. Even when "matched" it means nothing technically other than at some static dc state, with some plate load, at some unknown bias, a relative number is choosen for it. Tube rebranders do not publish specs for their tubes or describe what their matching number or letter grade represents. It might represent Mu but without details of where in a tube's plots this occured. It is perfectly possible for a low numbered tube in an actual working amp, to have more gain than a higher numbered tube.
So basically you know very little about the tube and the rebranders will not tell you, to which I suggest testing it a different levels of bias in your circuit measure it at different cathode currents to draw your own curves. That way, if another tube comes your way you have a basis for comparison. Grid curves would tell something meaningful.
In a 2 tube push-pull class AB amp, modest imbalance will not hurt anything but you might notice a difference in harmonic spectra, and a stronger 2nd harmonic. For many this would sound pleasing, particularly bass rigs.

[Chuck H]

+1
I don't get too compulsive about it but when I have several tubes of a type and brand I do test each one for current at a given bias at idle and with signal and just match them up in closest pairs. I don't even test for output but you could if you want to be more accurate. A little mismatch is NBD or, as noted, even desireable.

km6xz, the photo avatar is a nice addition. I may do that.

[Enzo]

The amp may be push pull, but it will function with one tube, at least well enough to test the tubes. So if you don;t already know which was the bad tube, try one at a time to see which one doesn;t work. Then IMMEDIATELY throw the bad one in the trash. Absolutely no reason to have bad tubes around.

Holding the good tube for potential CHamp use is a fine idea. But having it as a spare for this amp is also a good idea. It doesn't matter if the tubes match. I buy matched sets for their hum cancelling effect, but no amp is made with matched tubes. Fender makes amps, and when it comes time to stuff tubes in them, the tech reaches into the box and pulls out two tubes and sticks them in sockets. If you have one tube fail in the future, having the odd spare tube is far better than having an empty socket.

[km6xz]

Tube matching of tested-to-be-good tubes is not very important except for sound but in ways that is not expected by those who think it is the key to proper tube operation.
Each half of the push pull Class AB amplifier can be treated as a fully functioning amp even if the other side was absent. In fact for circuit analysis, the entire amp's transfer function is thought of as a curve set for one tube, copied and reversed and pasted below the Zero line.
If tubes are perfectly balanced the nature of push-pull is cancellation of the second harmonic. What would a amp sound like with no second harmonic and another with a 10% imbalance? The second harmonic would be stronger in the second and for some styles of music, rounder fuller bottom and more musicality. When people claim "oh, you have to try XXX tubes, I put in a set and the bottom end was fuller and warmer" given that there is NOTHING in a vacuum tube itself that can create more bottom, I have to laugh. Not likely, name a tube that does not have a -3db power bandwidth of less than 25megahertz. What can happen is premature hf rolloff can make the listener think the bottom is bigger.
So what the reviewer who was claiming the tubes were miracle workers did not understand was that the new tubes might have just been unbalanced enough to notice less suppression of the second harmonic. Another tube swapped at random into one of the positions might have enhanced that even more. Yet the internet suddenly is abuzz about how great XXX brand tubes are. You can see the difference quite easily between perfectly balanced and less balanced on a spectrum analyzer.
Often claims of brighter highs and clear crystalline quality of one tube versus another when there is nothing in the tube that could cause that but a tube and circuit combination can. For example testing 12AX7s is interesting, few have plate load resistances that match the 12AX7 definition of the early 50s or 63k or Mu of 1800/1700 and 1.4pfd grid capacitance. Some are closer to 12AY7's, those with long plates for example with lower plate resistance. The value of the coupling cap and actual plate load can alter the HF response when a tube is swapped in with a lower or higher plate resistance than intended. That changes the roll off of the plate load and interstage coupling network. The tube is fine, the tube in the circuit gives unintended response however. What the golden eyes hear, if there is anything there at all, is usually traceable to very measurable deviations from intended operation, and not some magical property of a brand of tube. Fairy dust usually turns out to be just dust and a mismatch in circuit to substituted parts.
Try creating a simple test jig and plot the grid curves for some of the popular tubes and you will find they are all over the map and would not be called 12AX7 or 6L6 by an engineer in 1960. I play with a lot of tubes people bring to me and I check them on my curve tracer. Overall, the unit to unit consistency is not great at all even with tubes claimed to be "matched". But that is OK, it might even sound better this time. The problem comes in trying to reproduce that sound when those tubes wear out. It is not the brand but the individual tubes that vary. That is why I am not very agreeable with people who make wild sonic quality claims about a brand or series of tubes, they prove with every sentence that they are blowing smoke. Some do not realize it, others do it because a "golden" tube fetches 2-10 times as much.

So to the OP, save the extra tube you can use it in the future when one pops in the amp.

[d. spree]

Tube matching of tested-to-be-good tubes is not very important except for sound but in ways that is not expected by those who think it is the key to proper tube operation.
Each half of the push pull Class AB amplifier can be treated as a fully functioning amp even if the other side was absent. In fact for circuit analysis, the entire amp's transfer function is thought of as a curve set for one tube, copied and reversed and pasted below the Zero line
If tubes are perfectly balanced the nature of push-pull is cancellation of the second harmonic. What would a amp sound like with no second harmonic and another with a 10% imbalance? The second harmonic would be stronger in the second and for some styles of music, rounder fuller bottom and more musicality. When people claim "oh, you have to try XXX tubes, I put in a set and the bottom end was fuller and warmer" given that there is NOTHING in a vacuum tube itself that can create more bottom, I have to laugh. Not likely, name a tube that does not have a -3db power bandwidth of less than 25megahertz. What can happen is premature hf rolloff can make the listener think the bottom is bigger.
So what the reviewer who was claiming the tubes were miracle workers did not understand was that the new tubes might have just been unbalanced enough to notice less suppression of the second harmonic. Another tube swapped at random into one of the positions might have enhanced that even more. Yet the internet suddenly is abuzz about how great XXX brand tubes are. You can see the difference quite easily between perfectly balanced and less balanced on a spectrum analyzer.
Often claims of brighter highs and clear crystalline quality of one tube versus another when there is nothing in the tube that could cause that but a tube and circuit combination can. For example testing 12AX7s is interesting, few have plate load resistances that match the 12AX7 definition of the early 50s or 63k or Mu of 1800/1700 and 1.4pfd grid capacitance. Some are closer to 12AY7's, those with long plates for example with lower plate resistance. The value of the coupling cap and actual plate load can alter the HF response when a tube is swapped in with a lower or higher plate resistance than intended. That changes the roll off of the plate load and interstage coupling network. The tube is fine, the tube in the circuit gives unintended response however. What the golden eyes hear, if there is anything there at all, is usually traceable to very measurable deviations from intended operation, and not some magical property of a brand of tube. Fairy dust usually turns out to be just dust and a mismatch in circuit to substituted parts.
Try creating a simple test jig and plot the grid curves for some of the popular tubes and you will find they are all over the map and would not be called 12AX7 or 6L6 by an engineer in 1960. I play with a lot of tubes people bring to me and I check them on my curve tracer. Overall, the unit to unit consistency is not great at all even with tubes claimed to be "matched". But that is OK, it might even sound better this time. The problem comes in trying to reproduce that sound when those tubes wear out. It is not the brand but the individual tubes that vary. That is why I am not very agreeable with people who make wild sonic quality claims about a brand or series of tubes, they prove with every sentence that they are blowing smoke. Some do not realize it, others do it because a "golden" tube fetches 2-10 times as much.

So to the OP, save the extra tube you can use it in the future when one pops in the amp.

i really appreciate your response - and the knowledge embedded. Thanks.

[jrfrond]

I've been spot checking new power tubes on the Maxi-Matcher. I have found that many times, they are not even close, and if you burn them in and test them again, the original numbers are out the window anyway.

It was a bit different back in the day of excellent tube manufacturing. There was consistency, and NO ONE matched tubes back then. They went from the crate to the amp, and all was good. A lot to be said about the manufacturing du jour. I can't seem to recall when Aspen Pittmann started Groove Tubes, the first matched tubes I ever saw, but I believe it might have been while USA stock was still available.

[Tom Phillips]

I can't seem to recall when Aspen Pittmann started Groove Tubes, the first matched tubes I ever saw, but I believe it might have been while USA stock was still available.

I remember when Groove Tubes came on the scene in the 1970's. I believe that at first the entire business was based on NOS (Mostly USA manufactured) tubes. They put their own twist on matching methodology which continues to this day. Before that time you could special order matched tubes from suppliers. However, it was rarely done and I don't believe that there was a strict defined standard as to what "matched" meant. I've heard some great sounding amps that turned out to have very mismatched power tubes installed. Therefore, I don't worry about it much except for very high power clean bass amps such as the classic Ampeg SVT.