matched triode driver tubes

[dadm]

Sometimes it's worth it to spend a few extra bucks for a select tube.

A JJ EC81 (12AT7) balanced matched triode driver tube noticably improved

the sound of a recently re-tubed '66 bassman. Much better than the standard

new Russian 12AT7 I tried. It's easy to specify select tube options from most

good tube places. Might not make a bad amp sound good but makes a good

amp sound better. Brings out the best in your expensive power tubes.

[EMan]

How does an un-matched JJ compare to a matched JJ? The idea here is compare apples to apples.

[WrapAround]

Originally Posted by EMan

How does an un-matched JJ compare to a matched JJ? The idea here is compare apples to apples.

This is very true.

[Billm]

In my experience, the matched driver/phase inverter is a waste of money. Fortunately, it's only a dollar or so.

Nothing else is matched in the PI circuit or output circuit--the resistors are all +/- 5 percent and the caps are +/- 20 percent. The plate resistors are typically different sizes to compensate for the essential weirdness of the long-tailed PI design.

So the fresh JJ tube likely had far more effect than the fact that its triodes were matched.

[ThermionicScott]

I tend to buy matched triodes when I can, not because I can hear a difference, but like Bill said, it's not that much more. As an engineer, I like it when things are internally consistent.

- Scott

[EMan]

Well, Bill...that exactly my feelings as well. There's a lot of things post-PI that can affect the 'balance' of the Plate waveforms on the PI. Fundamentally, a long tail pair circuit isn't all that balanced to begin with.

Add to that what you mentioned, and the other plate load resistors (the bias feed Rs), the output tubes, and the OT primary balance, and I'd say that a pack of picks would be a better use of an extra buck .. :-)

OTOH, most of this amp 'stuff' is an individual thing, and there's nothing wrong with using a matched triode.

Bill, the different plate load resistors in them LTPIs are because the signal to the in-phase section is fed via the cathode, and has a different gain, than the grid input of the out-of-phase section. Notice too, that when Grid #2 is used for NFB, it's also tied to the cathodes to re-re-balance the circuit.

[ThermionicScott]

Originally Posted by EMan

Fundamentally, a long tail pair circuit isn't all that balanced to begin with.

I would take issue with that statement. Component tolerances aside, a well-designed long-tailed pair is very balanced. Richard Kuehnel demonstrates this in his excellent 5F6-A book.

Actually, a cathodyne PI also has excellent balance, but higher-wattage amps need the current-delivering capacity of the LTP.

- Scott

[EMan]

Scott - I agree that it is possible to make a decent balanced LTPI, although in the case in point...being a old guitar amp with not necessarily tight component specs.

I did a little simulation of the DR style PI here - http://home.texoma.net/~flhh/ltpi/ltpi.htm , if anyone's interested.

[Tim Swartz]

Mostly a waste of $$$ IMO... and here is the downside to the practice of selling matched/balanced preamp tubes... the "reqular" ones are now the rejects. And why buy a cruddy current production when a NOS 12AT7 can be had for what, $15? And it will probably outlive you!