Quick and Dirty Tube Matching

[ny_racer_xxx]

Does anybody have a quick and dirty way to match tubes? I assume I can put them in the amp and check current draw on each?

CR

[Phil_S]

Any fixed bias amp is a tube tester. You get test results for one set of parameters, but at least they are real life operating conditions.

You see, in a fixed bias amp, you can probably only control grid (negative bias) voltage. At a constant grid voltage, you will get plate and screen readings that are different for each tube. You may want to take readings at several (maybe 3) pre-determined grid voltages so you can plot a curve (even if just in your head). This is really a transconductance check. Transconductance (Gm), in non technical terms, is a measure of how much you can expect a tube to change its characteristics over a range of voltages. When you run a guitar signal through the amp, your power tubes fluctuate.

For example, if you get a decent match at two different grid voltages, and the transconductance is close, that is a better match than a static match at one grid voltage.

I have a pile of 6AQ5's I was trying to match, using a fixed bias VibroChamp I built. I'm not satisfied that the results are all that useful and eventually concluded the work had value, but was limited in application. The basis for conclusion was that I subsequently built a T-wreck xpress variant for 6AQ5's. I expected tubes matched from the VibroChamp test data to be matched in the Xpress. Since the VibroChamp runs Va ~250 and the Xpress ~335, I rejected some that I thought were pairs, as they were too far apart under different operating conditions. This is where the Gm number proved to be a useful pointer.

Here's my worksheet, for what it's worth. I'd welcome comments if anyone care to offer them.

[kd]

I'll just say this first, I don't know a thing about tube testers or how to tests tube. What if a person built a power section like an amp with adjustable fixed bias and installed a VVR on it to adjust B+ voltage, would you get more usable data from something like that? You could check the current draw at both high and lower voltage and maybe see more deviation on current draw and oossibly get a better match.

Ken

[zaphod_phil]

As many know, I have long maintained that closely matched tubes have no real benefit in guitar amps. Assymetry in PP power amps from "poorly balanced" PIs and/or "poorly matched" power tube yield tones that are richer in even-order harmonics. We also need to remember that the great vintage amps that we try to clone never used matched power tubes. The guy in the factory would just pull the next two (or four tubes) out of the box and plugg them into the amp. In my own A/B tests between closely matched and "ApproximatchedTM" tubes , the closely matched ones sounded quite noticeably more sterile.

So my conclusion is that all you need to do is to place a pair of power tubes in a fixed-biased amp, or a cathode-biased amp with separate cathodes, and calculate the cathode current. If they're in about the same ball park, that's all you need.

Obviously, for hi-fi, it's a whole different story....

[ny_racer_xxx]

Phil, by your definition is that a "mismatched" tube set sounds better?
OK, then it would be more important to quantify how much they are mismatched to get that sound.
I like the idea of building a rig just for the purpose of testing tubes.
Has anybody built such a rig?

CR

[Phil_S]

I like the idea of building a rig just for the purpose of testing tubes. Has anybody built such a rig?
CR

I know of at least a couple of tube dealers who have built their own rigs for testing and matching. It's my understanding that such a gizmo is a bit more overhead than the average Joe either wants to build or lay out the money for. Also, these are proprietary in nature, and I don't imagine plans would be freely shared. It is basically a power amp with the ability to control voltage on all parts of the tube and is multi-socketed for batch testing.

[ny_racer_xxx]

I think this basically the same dicussion as the super tube tester. http://18watt.com/modules.php?name=Foru ... 639#192541
I will post on there...

CR

[Colin_D]

As many know, I have long maintained that closely matched tubes have no real benefit in guitar amps. Assymetry in PP power amps from "poorly balanced" PIs and/or "poorly matched" power tube yield tones that are richer in even-order harmonics. We also need to remember that the great vintage amps that we try to clone never used matched power tubes. The guy in the factory would just pull the next two (or four tubes) out of the box and plugg them into the amp. In my own A/B tests between closely matched and "ApproximatchedTM" tubes , the closely matched ones sounded quite noticeably more sterile.

So my conclusion is that all you need to do is to place a pair of power tubes in a fixed-biased amp, or a cathode-biased amp with separate cathodes, and calculate the cathode current. If they're in about the same ball park, that's all you need.

Obviously, for hi-fi, it's a whole different story....

I was under the impression that if you don't use a matched pair of power tubes you'll lop off half their life expectancy right out of the box. Have I been misinformed? And also, I never heard of triode matching for PI tubes until I read about how an unbalanced PI is desirable (valvewizard site)
I use mostly new production tubes so I don't invest all that much money. I tried a set of NOS Mullard 6BQ5's in a TMB, and it sounded exactly the same to my ears as the new production crappers I had in it before. My point being, NOS lovers would seemingly want their precious investments to last as long as possible, wouldn't they? Isn't that the point of tube matching?

[DaveW]

Keep meaning to build this simple tester from Rikstone Amps. If anybody has can they let me know what they think.

Schematic

http://www.kolumbus.fi/risto.kivioja/Ri ... Tester.pdf

More info

http://www.kolumbus.fi/risto.kivioja/RikstoneAmps/

A more sophisticated tester is shown here

http://geek.scorpiorising.ca/GeeK_ZonE/ ... pic=1803.0

[zaphod_phil]

BTW before the mods start getting hit with e-mails about why we're allowing this non-18W thread to go through, I'm assuming what we're discussing here is in fact of direct relevance to 18W amps.

I was under the impression that if you don't use a matched pair of power tubes you'll lop off half their life expectancy right out of the box. Have I been misinformed?

I'm afraid you have, to a large extent IMHO. If your tubes were unmatched enough to damage each other, you would immediately know from a high leve l of hum, or one of the tubes tending to red-plate. The primary reason for using matched tubes is for fidelity. A good hi-fi designer will chose an operating point for the power tubes that minimizes generation of 3rd and other odd-order harmonics. Often the tubes will then generate quite a lot of even-order harmonics. So the designer will use closely-matched power tubes and well-balanced PI to eliminate those as well.

And also, I never heard of triode matching for PI tubes until I read about how an unbalanced PI is desirable (valvewizard site)

That's another rip off that some of the tube sellers promote. ideal for hi-fi, but useless in guitar amps.

I use mostly new production tubes so I don't invest all that much money.... My point being, NOS lovers would seemingly want their precious investments to last as long as possible, wouldn't they? Isn't that the point of tube matching?

I use almost nothing but NOS tubes myself (although I've never paid silly money for them). So I would be one of the first to worry about something like that.

[ny_racer_xxx]

Dave thanks for the link!!! That's a real nice and simple unit. Did you see they have an advance schem wich included 6L6's and 6V6's, just what I need!!!
I think if you made the bias ajustable, you could plot a curve of the plate currents better, maybe just 3 voltages?

CR