How close should power tubes match up?

Discussion in 'Amps/Cabs Tech Corner: Amplifier, Cab & Speakers' started by jumpnblues, Aug 1, 2008.

  1. jumpnblues

    jumpnblues Member

    Messages:
    5,250
    Joined:
    Jan 1, 2007
    I have an opportunity to go through a former TV dealer's NOS power tube supply...it may be a back room treasure trove. He says he has lots of American brand tubes. One question I have is how close should power tubes match up? He has a tube tester and knows how to use it, so assuming he has some tubes I want, what steps should we take to assure I get really good results and therefore really good tubes? I'll be mainly interested in 6L6/5881 and 6V6 power tubes since that's what my amps require.
    I really have no idea what he has or if what he has will be of interest or useful. He says he has both power and preamp tubes. I don't know if he knows the value of what he has even if it will be of interest to me. I will not take advantage of him though.
    Anyway, back to my original question...How close should power tubes match? Anything else that should be of concern? Thanks.


    Tom
     
  2. SatelliteAmps

    SatelliteAmps Member

    Messages:
    6,185
    Joined:
    Oct 19, 2007
    Location:
    Normal Heights, Calif
    Within 5ma draw in an amp is good. As for how close in measurement depends on the tester. A lot of the old tube testers only test whether the filaments light up and whether there are any shorts. They don't put any load (B+ type) on the actual tube. If it is one of those, then you won't get great measurements to compare them.

    Just to throw out a few other things if I may. Most of the old TV's don't use the same tubes that we like to use in guitar amps. You might find a few power or preamp tubes, but usually it is only a very few. Might be some good rectifiers. Also, a lot of the old repair guys would just throw their pulls into the same box that they kept the originals in, so there might be some weak or dead tubes as well, even in NOS looking boxes.
     
  3. HipKitty

    HipKitty Silver Supporting Member

    Messages:
    2,636
    Joined:
    Jun 29, 2007
    Location:
    Pennsylvania Dutch Country
    Well said Adam!
     
  4. The Pup

    The Pup Supporting Member

    Messages:
    2,722
    Joined:
    Nov 5, 2005
    I like a little 3D swirl and depth...no Swiss-antiseptic efficiency for me.

    3ma-9ma spread is generally within my tolerance level (amp dependent of course).

    Note: Red Rhodes taught me to use an o-scope, signal generator and dummy load to check the cross-over notch of my AB2 amps nearly 40 years ago. For decades, no one knew what I was talking about...and to be honest, no one ever really needed to.

    If it sounds good...it is good.

    Think back to all of those famous recordings and performances; how many hinged upon how well balanced the performer's PI or output tubes were?
     
  5. jumpnblues

    jumpnblues Member

    Messages:
    5,250
    Joined:
    Jan 1, 2007

    Adam,


    Very useful info. Thank you.


    Tom
     
  6. SatelliteAmps

    SatelliteAmps Member

    Messages:
    6,185
    Joined:
    Oct 19, 2007
    Location:
    Normal Heights, Calif
  7. RedRock

    RedRock Member

    Messages:
    3,575
    Joined:
    Dec 26, 2006
    There was an article in the Vintage Guitar Magazine about 5 months ago.
    It said that there are a small number of persons who like the sound of
    intentionally mis-matched power tubes. Of course, this will create hum
    in the power section. I match them as closely as possible.
     
  8. dk123123dk

    dk123123dk Member

    Messages:
    3,887
    Joined:
    Sep 15, 2007
    Location:
    Mass
    I just read an interview from Caesar Diaz (SRV's guitar tech). He said that many of the original Fender amps didn't come with matched power tubes, so he wouldn't pay too much mind to matching.

    Good luck on the tube score!

    dk
     
  9. rockon1

    rockon1 Member

    Messages:
    11,467
    Joined:
    Aug 12, 2005
    Location:
    Stamford CT
    Back in the day nobody matched power tubes.Bob
     
  10. 12ax7m

    12ax7m Member

    Messages:
    1
    Joined:
    Nov 25, 2007
    Location:
    NE Penna, USA
    Hi. I also read about some techs intentionally mismatching tubes to create unique tone. I check each individual tube first without touching the bias adjustment, then pair up the highest and lowest measuring tubes on one phase bank, then the two middle measuring tubes get paired up on the other phase bank. This seems to work best for me in providing the "smoothest" tone, IMO. Hope this helps. Dave :)
     
  11. Faustine Amps

    Faustine Amps Member

    Messages:
    423
    Joined:
    Feb 12, 2007
    Location:
    Southern California
    Hi Tom,

    Seems to be a lot of different opinions regarding tube matching. I prefer matched tubes and only use matched tubes (to within 10% on the idle current). I don't believe there is any special "mojo" to be had from using mismatched tubes. In fact, I can recall replacing tubes in my old Fender amps back in the seventies and being very dissapointed on a few occasions with the way the new tubes messed up my tone. I didn't know anything about tube matching or setting the bias (most guitarists didn't), I just took my tubes to the Radio and TV shop (or even the local drug store), stuck them on the emission tube tester, and replaced the ones that were in the "BAD" zone. If the new tubes sounded bad or produced more hum, I lived with it until I bought the next batch of tubes. It was a crapshoot!

    I've also sorted through a number of TV repairmen's tube stocks over the last ten years. They're getting harder to find these days as most of the TV dealers or repairmen tossed their tubes in the dumpster back in the 80's or 90's, but there are still some occasional treasures to be found (and I've been pretty lucky).

    So... to answer your original question, tube matching is a good idea, but not as easy as you might think. Even if you're able to test them on a tube tester, that's not going to find you any guaranteed "matched" tubes. If you use a basic "emission type" tester, you can test for shorts, emission, and maybe gas, but you can't match tubes with this type of tester. If you test them with a "transconductance type" tester, like one of the Hickok models, you can match them for transconductance, which is sort of like a measurement of the tube's gain (a simplified explanation, but close enough for this discussion). This is the only "matching" that you can do with a typical tube tester. If you can find 6L6 or 6V6 tubes that have a transconductance reading within a few hundred "micromhos" or Gm, then you can consider them matched (at least partially... read on).

    The second test to be done is even more important, but it requires plugging the tubes into a real amplifier and making "idle current" measurements. Some tube testers have a provision for testing "idle current" (like my Precise Model 111, for instance). But, since a tube tester will typically operate a tube at 150 to 200 volts and a typical tube amp operates power tubes at between 400 and 500 volts, the best way to match tubes is in a real amplifier operating under real conditions. To test for "idle current" in a real amp, you'll need some type of bias measuring device, like the bias meters that are available from Weber, Allessandro, T.A.D., and others. I won't go into that procedure as it's been covered many times elsewhere. In fact, there's a decent video tutorial at mojomusicalsupply.com covering the T.A.D. unit that is applicable to the other units as well.

    So, in short: Match the tubes for transconductance first with a good transconductance or mutual conductance tube tester, then match them for "idle plate current" in a real amp with one of the commercially available bias meters. I match the tubes to within 10%. So if you've got a pair drawing somewhere around 30 milliamps each, if they are within 3 mA they're matched to within 10%. If they're runing closer to 40mA, then 10% is 4mA, and so on. 5 mA is usually a pretty decent match (for 6L6GC) but if you're running a pair of 6V6 tubes at around 20 milliamps, then a difference of 5 milliamps is a 25% mismatch. Sorry to get so long winded. Not the simple answer you were hoping for?

    Happy hunting!

    Regards,
    Tim Gregoire
     
    Last edited: Aug 3, 2008
    twinrider1 likes this.
  12. The Pup

    The Pup Supporting Member

    Messages:
    2,722
    Joined:
    Nov 5, 2005
    When using an o-scope on AB2 amps, I get a "pure" or clean sine wave and then back off to settle into a very slight cross-over notch; just the way Ray Rhodes taught me nearly 40 years ago. Sounds the best to my ears.

    For an audiophile application, I would try to get the sine wave as near perfect as possible. For a guitar amplifier...a little character and coloration at the cost of some efficiency is perfectly fine .

    I also agree, that a delta of 10% is a good target.
     
  13. Faustine Amps

    Faustine Amps Member

    Messages:
    423
    Joined:
    Feb 12, 2007
    Location:
    Southern California
    So you knew Red Rhodes? He was "da man" when it came to troubleshooting and designing amps. I know an engineer / physicist at JPL (who is now retired, but still consulting for JPL) who was a very good friend of Red Rhodes. When he recently moved, he let me go through all of his vintage electronics and test equipment and he had some of Red's old test bench gear (some of it custom designed) which is now sitting in my own shop. He had a lot of crazy tales to tell too. Do you know John South?

    Regards,
    Tim Gregoire
     
  14. The Pup

    The Pup Supporting Member

    Messages:
    2,722
    Joined:
    Nov 5, 2005
    His name sounds familiar. I also work with a guy named Ray Rhodes...sorry Red.

    There's a guy in LA County who has all of Red's benchstock and a great deal of his old test equipment. I can't remember his name...I have it at home (I'm away on business).

    If I remember correctly, Red played a mean lap steel too!

    Red taught me everything I know about amp basics...so much of which I have forgotten...I went into the military for my career; I couldn't stand relying on musicians for my next meal. :)

    He was a good dude. RIP Red.
     
  15. Blue Strat

    Blue Strat Member

    Messages:
    30,036
    Joined:
    Aug 2, 2002
    Location:
    Sterling, VA (not far from Washington DC)

    ..but overkill when biasing 6V6s at the correct 20mA.

    5mA is a good benchmark but higher often works fine.
     
  16. Blue Strat

    Blue Strat Member

    Messages:
    30,036
    Joined:
    Aug 2, 2002
    Location:
    Sterling, VA (not far from Washington DC)

    ...unless they're matched for mutual conductance as well as idle current.
     
  17. mark norwine

    mark norwine Member

    Messages:
    16,169
    Joined:
    Jul 25, 2005
    Location:
    NJ...GSP135
    Bingo
     
  18. Faustine Amps

    Faustine Amps Member

    Messages:
    423
    Joined:
    Feb 12, 2007
    Location:
    Southern California
    Really? "If it sounds good... it is good".

    Is that the "mantra" that we want to promote? If that's the case, then why bias at all? I could toss some new 6L6s in my amp and if they sound good, why check the bias? They could very well be drawing 60 mA apiece but they sound great! I'll worry about repairs down the road.;) Actually, the tubes would probably just wear out fast, I doubt it would cause any damage, but I guess you get my drift.

    I know my 10% matching requirement is pretty anal by most amp tech's standards. That's alright, it's just my own thing. I work for perfectionists in my day job, so I probably swing too far toward the "perfectionist" camp. I have been known to use tubes that are up to 7 or 8 milliamps out of "match" (in my personal amps), but please don't let that get around.

    Regards,
    Tim Gregoire
     
    Last edited: Aug 6, 2008
  19. Swarty

    Swarty Member

    Messages:
    1,143
    Joined:
    Dec 16, 2003
    Location:
    Michigan
    Amen. They are likely to drift apart anyways. At some point (30%?) some hum will start to creep in. The more tubes are mismatched the more emphasis on even-order harmonics for whatever that's worth. Of course the fundamental is compromised which may be fine for greasy leads, but not so good for bass or tight rhythm.
     
  20. Blue Strat

    Blue Strat Member

    Messages:
    30,036
    Joined:
    Aug 2, 2002
    Location:
    Sterling, VA (not far from Washington DC)
    Using mutual conductance AND plate current at a plate voltage somewhat close to the amps real plate current/voltage removes the "arbitrary-ness".
     

Share This Page