How much EL34 bias mismatch is acceptable?

Discussion in 'Amps/Cabs Tech Corner: Amplifier, Cab & Speakers' started by devbro, Feb 15, 2006.

  1. devbro

    devbro Member

    Messages:
    1,128
    Joined:
    Apr 17, 2004
    Location:
    Los Angeles
    I'm attempting my first bias on my EL34 Shiva. I purchased a dual bias tester, $20 DMM and a pair of SED winged "C" EL34's.

    I put the new tubes in, let them get real hot and the closest I can get them adjusting the bias pot is within 5mv of each other. One tube measures 30mv and the other 35mv.

    Is this much mismatch acceptable or should I tell the guy I bought the tubes from they're not a matched set?
     
  2. VintageJon

    VintageJon Member

    Messages:
    155
    Joined:
    Nov 3, 2005
    Location:
    Austin TX
    20% is the industry standard. For 35mA this would be 7mA difference and for 30mA this would be 6mA difference.
    Your pair is "matched" at 5mA or 16%...
    In single pair 20% is about what you can expect. I usually take 24 to 100 at a time in "matched quartets" and weed them out for 10%.

    Here I go for 10%, but I'm overly picky and I don't whine to supplier if they are within 20%. I do re-group them for 10% match, but I buy a lot of tubes!
    For a private user this is impractical, but being a business I sell a lot of tubes.

    I think you're OK..

    -Jon
     
  3. devbro

    devbro Member

    Messages:
    1,128
    Joined:
    Apr 17, 2004
    Location:
    Los Angeles
    Thanks John,
    Thats exactly the answer I was hoping for. One more question; what is the sonic benifit of closer matched tubes?
     
  4. brad347

    brad347 Member

    Messages:
    4,813
    Joined:
    Feb 8, 2006
    Location:
    Brooklyn, NY
    some people actually like the character of slightly mismatched tubes. Slightly harder on your output transformer though.
     
  5. VintageJon

    VintageJon Member

    Messages:
    155
    Joined:
    Nov 3, 2005
    Location:
    Austin TX
    brad347's point is valid.

    In answer to your question- I find closer match gets more headroom and more definition. The headroom is more obvious than the definition.

    I do a lot of amps for ''clean" players so I need that.

    -Jon
     
  6. samwheat

    samwheat Member

    Messages:
    231
    Joined:
    Aug 14, 2005
    Location:
    Scranton, PA
    Do you think Jimi worked with a Hickok and matched tubes?

    Non-matched tubes adds some color to the sound.

    Wasn't it Aspen Pittman who started all this matching ********?
     
  7. brad347

    brad347 Member

    Messages:
    4,813
    Joined:
    Feb 8, 2006
    Location:
    Brooklyn, NY
    closely matched tubes is great for hi-fi.

    So are telefunken 12AX7s but I don't want them in my guitar amp.

    That said, I do try to match my tubes/get matched tubes. Don't really know why.
     
  8. donnyjaguar

    donnyjaguar Member

    Messages:
    4,243
    Joined:
    Nov 4, 2005
    Location:
    Canada
    Having worked in electronic product design & manufacturing, I will say that there is a lot of component matching that goes on. Often times you see colored dots or letters on output devices and these are almost always identifying the characteristics of the device so that matched sets are used. In a manufacturing environment you match a lot of devices at once though, not a pair at a time. That said, the matching is probably a little looser than what some of the tube suppliers can offer. I don't really care too much about matched output tubes myself, but I try to avoid grossly mismatched tubes.

    DJ
     
  9. devbro

    devbro Member

    Messages:
    1,128
    Joined:
    Apr 17, 2004
    Location:
    Los Angeles
    Sounds like a mismatch of 5mv shouldn't be a problem. My Shiva has miles of headroom (85 watts) so I'm hoping I won't suffer in that area from the mismatch.

    Thanks to everyone for all your help:AOK
     
  10. VintageJon

    VintageJon Member

    Messages:
    155
    Joined:
    Nov 3, 2005
    Location:
    Austin TX
    No, Jimi played through whatever was there...

    Mismatch does add "color" and if you like that shade of distortion...

    Aspen's "matching" is a VERY loose system that I don't agree with at all.
    Tube matching has been around a lot longer than the GT trademark...


    I feel I should point-out that when you are running a mismatched pair you really need to set the bias for the tube that draws the MOST cathode current. This will at least avoid biasing this tube too hot. If one is pulling 35mA and the other 8mA you just have to accept it, or get a matched pair...

    -Jon
     
  11. brad347

    brad347 Member

    Messages:
    4,813
    Joined:
    Feb 8, 2006
    Location:
    Brooklyn, NY
    ...and i'll reiterate that if it's a class A/B amp and one tube is pulling 35ma and the other 8ma then it's going to be a bit harder on your precious output transformer.
     
  12. VintageJon

    VintageJon Member

    Messages:
    155
    Joined:
    Nov 3, 2005
    Location:
    Austin TX
    What Brad said...

    In the 50's and 60's the USA-made 6L6's were more closely matched, even NOS, ANOS and UOS are today! It just wasn't a problem...
    UOS pair of GE I use for testing Surf Bugs are still 31Ma and 33mA!

    Modern production #'s are all over the map...

    Best match for Distributor-Matched is Ruby 6L6GCMSTR in my experience.
    I have pairs in several Super Reverbs that gig weekly and practice all the time that have been in there since 1997! (I can't tell them from new ones in comparitive testing!!!)

    -Jon
     
  13. Mickey_C

    Mickey_C The Original Racketeer Gold Supporting Member

    Messages:
    1,530
    Joined:
    Mar 28, 2003
    Location:
    Sedona, AZ
    Just install dual bias pots, and get on with life.
     
  14. Swarty

    Swarty Member

    Messages:
    1,143
    Joined:
    Dec 16, 2003
    Location:
    Michigan
    Why/how is a mismatched tube scenario bad for the output transformer?
     
  15. brad347

    brad347 Member

    Messages:
    4,813
    Joined:
    Feb 8, 2006
    Location:
    Brooklyn, NY
    This is mainly true in class A/B amps and the reason is this:

    A class A/B amp works, in simple terms, by dividing the signal in half, amplifying each half separately, and re-combining the two halves at the output transformer. Because of this, while one tube is getting full power, the other is getting less power, and they sort of see-saw back and forth... this type of operation is known as "push-pull" for obvious reasons. Well, in these types of amps, the reasons for matching your tubes are obvious... each half is amplified in a more similar manner to the other...

    This is also why, in a four-power-tube amp, you may use two matched pairs instead of a matched quartet, as long as they are arranged symmetrically (i.e. XYYX or YXXY, but NOT XXYY or XYXY)... each half of the signal is still getting amplified in a very similar manner.

    I'm not sure the exact mechanics of it, maybe someone else can chime in, but i've just always heard that with the two halves of the signal being very similar, it's easier on your output tranny when they are re-combined there.
     
  16. Swarty

    Swarty Member

    Messages:
    1,143
    Joined:
    Dec 16, 2003
    Location:
    Michigan
    I understand the rational for matched tubes. What I don't understand is how mismatched tubes can be harmful to an output transformer as you stated above, nor do I see how it could be related to class of operation. I'd just like an explanation as I don't think the transformer would really care if one side was running at 10mA and the other was running at 50mA. Is there a Randall Aiken in the house?
     
  17. brad347

    brad347 Member

    Messages:
    4,813
    Joined:
    Feb 8, 2006
    Location:
    Brooklyn, NY
    What I do know is that severely mismatched tubes can heat up the output transformer as much as 20 degrees hotter than it gets with a well-matched set. Someone else could probably explain why... try it with a thermometer!

    The heat can't be good, though many quality transformers i'm sure could handle it.

    I'd imagine that the vastly mismatched tubes would put unequal stress on different ends of the output transformer but I can't speculate with any detail beyond that.

    It's probably not too grave a concern except in the most extreme of cases.
     
  18. jbltwin1

    jbltwin1 Member

    Messages:
    269
    Joined:
    Jan 23, 2006
    Location:
    alton, il
    Jon brings up a great point(as usual) that kind of got glassed over a bit. I see too many guys who "bias" their amps with that test point crap on some amps that gives you a total cathode current for two tubes. This is a dangerous procedure in my opinion because you CAN have two grossly mismatched tubes. I go through this with the guys at the Music Man board all the time because the original set up procedure was to measure a voltage at a point in the cathode ground circuit that should equate to 50mV or 25 per tube IF THEY WERE EXACTLY BALANCED. Yeah, like that's going to happen. The new Marshall amps are the same way(like the 601). Don't (if you have an option) use the so called test point as they will only give you a total cathode current if that is the way you are testing current. Whether or not you use one of the bias adapter thingys or measure current with a one ohm resistor, do it per tube. Rant over.
     
  19. brad347

    brad347 Member

    Messages:
    4,813
    Joined:
    Feb 8, 2006
    Location:
    Brooklyn, NY

    Maybe i'm wrong! It's just always what I've heard. You don't think that the exterior temperature of the transformer reflects on anything? I'm learning...
     
  20. V846

    V846 Member

    Messages:
    322
    Joined:
    Dec 22, 2003
    Location:
    Enfield, CT

Share This Page