[Gvvt]

Another can of worms...

I did a search and didn't find anything (recent) that specifically discussed the value of tube matching. Elsewhere I've seen arguments that it's a merchandising technique with little real-world benefit.

If there is a thread somewhere, could you please direct me to it?

I have no agenda; I'm just trying to learn about how tubes work.

And please remember, we all live in a post-partisan world now.

I know it's true - I read it on the internets.

 

[bobsmith]

The Tube FAQ sticky (http://www.head-fi.org/forums/f5/new...ewbies-119389/) has a section on tube matching. Says that matching in power tubes = usually good, but not generally required in other tubes.

 

[tomb]

Quote:

Originally Posted by bobsmith /img/forum/go_quote.gif The Tube FAQ sticky (http://www.head-fi.org/forums/f5/new...ewbies-119389/) has a section on tube matching. Says that matching in power tubes = usually good, but not generally required in other tubes.

Hmm ... I found the section a bit ambiguous, really. Most headphone amplifiers are synonymous with pre-amplifiers. In that respect, the pre-amp tubes are critical.

More than that, in many (if not most) headphone amps, the primary signal amplification is done in the pre-amp tubes. I would think matching there is very significant.

To the OP, what happens with unmatched tubes is that many times you can adjust them to have "centered" stereo, but the peaks will mostly be apparent only in one channel or the other. If the mismatch is truly large, it can be very irritating - almost like one side of the headphone is losing its connection.

 

[Skylab]

This has been covered in other thread - the importance of tube matching depends entirely on the circuit that the tube will be used in. In some circuits it won't matter one bit. In some it matters enormously. But there is no single answer to this question.

 

[BigTony]

There is no straight answer, but the fun with tubes is that you can experiment and find out, and generally at a fair price. I bought a few batches of misc. tubes that have sonically been magnificent, and then some matched within 0.1 that have sounded awful!

l admit though that i can balance l+r chanels via my pre-amp.

 

[Happy Camper]

http://www.head-fi.org/forums/f5/tube-matching-64222/

 

[Gvvt]

Happy Camper,

Thanks for the link. I don't think I would have had the patience to thrash back that far.

The quotes in the last post are particularly interesting (no indication of their source):

"Tube tester matching is NOT normally effective matching because most tube testers operate the tube far below the normal operating parameters used in the amplifier. Tubes that match at typical tube tester voltages will NOT usually be matched in the amplifier."

So - what are "Matched Pairs" matched on?

And, does the match hold, or does it change with use?

 

[sacd lover]

Check and see if your amp parallels both the 6922 and 6AS7. If your WA2 parallels the tubes triodes that will cancel out most of the differences between the tubes. I am 99% sure that atleast the 6AS7s are run paralleled. Matched tubes really arent necessary in this scenario.






Quote:

Originally Posted by Gvvt /img/forum/go_quote.gif Happy Camper, Thanks for the link. I don't think I would have had the patience to thrash back that far. The quotes in the last post are particularly interesting (no indication of their source): "Tube tester matching is NOT normally effective matching because most tube testers operate the tube far below the normal operating parameters used in the amplifier. Tubes that match at typical tube tester voltages will NOT usually be matched in the amplifier." So - what are "Matched Pairs" matched on? And, does the match hold, or does it change with use?

 

[ical]

Fyi,

http://thetubestore.com/powertubeinfo.html

 

[tomb]

Quote:

Originally Posted by sacd lover /img/forum/go_quote.gif Check and see if your amp parallels both the 6922 and 6AS7. If your WA2 parallels the tubes triodes that will cancel out most of the differences between the tubes. I am 99% sure that atleast the 6AS7s are run paralleled. Matched tubes really arent necessary in this scenario.

Is this really true?

For instance, if you have triodes in one dual triode tube that measure 85/90, but then a tube on the other channel has triodes that measure 120/100, is that going to cancel out?

Further, if those dual triodes are paralleled, does 85/90 = 87.5 and 120/100 = 110? Then would you have a tube on the left that was 87.5 and a tube on the right that was 110? Wouldn't that be an imbalance?

Seriously, I'm just asking.

 

[sacd lover]

Quote:

Originally Posted by tomb /img/forum/go_quote.gif Is this really true? For instance, if you have triodes in one dual triode tube that measure 85/90, but then a tube on the other channel has triodes that measure 120/100, is that going to cancel out? Further, if those dual triodes are paralleled, does 85/90 = 87.5 and 120/100 = 110? Then would you have a tube on the left that was 87.5 and a tube on the right that was 110? Wouldn't that be an imbalance? Seriously, I'm just asking.

If you take tubes that are obviously completely mismatched they are not going to work. You know that.

 

[tomb]

Quote:

Originally Posted by sacd lover /img/forum/go_quote.gif If you take tubes that are obviously completely mismatched they are not going to work. You know that.

The question is what constitutes an obvious mismatch in a circuit where the triodes within the same tube are paralleled? The answer is not so obvious.

So, I sent a note to Pete Millett. He tells me the Gm's add in a circuit where the triodes are in parallel (although he also admitted that it's been a topic of some debate). That was helpful information. Again, I was truly trying to find the answer - I've been unable to find it documented anywhere.

That means that a tube that is 105/95 (=200) should be the same as a tube that might be 120/80 (=200). In actuality, the triodes within the same tube are rarely that far off (120/80), but they always seem to be slightly different (105/95, etc.), never the same.

Hence, the essential question, how do you match four sets of triodes among two tubes that may all be different? Adding the Gm's answers the question and also explains why people refer to paralleling triodes as "averaging out the differences."

 

[mrarroyo]

One thing to remember is that tubes will change after operation. So if you took to tubes that were less than 5% apart when new and tested them again after 100 hours and then 500 hours you will be surprised by the change.

 

[Gvvt]

mrarroyo,

that's part of what I was thinking about, too: the effect over time.

And - how mismatched were the tubes we chase after - Western Electric, Tung-Sol, Telefunken, Amperex, etc - when they left the factory? Was that an issue? Were there acceptable and unacceptable margins of difference?

Just curious.

 

[PFKMan23]

Quote:

Originally Posted by Gvvt /img/forum/go_quote.gif mrarroyo, that's part of what I was thinking about, too: the effect over time. And - how mismatched were the tubes we chase after - Western Electric, Tung-Sol, Telefunken, Amperex, etc - when they left the factory? Was that an issue? Were there acceptable and unacceptable margins of difference? Just curious.

I'm sure that there was some sort of tolerance standard. However, given that in many cases, a fairly significant amount of time has passed, eg: 6sn7 VT231 was a WWII designation (so over 60 years ago), honestly I don't know that it would matter given that these tubes have been kept in varying conditions over the years.