Cold bias in tube amps

Discussion in 'Amps and Cabs' started by Fuego, Jan 3, 2018.


  1. HaroldBrooks

    HaroldBrooks Member

    Messages:
    1,571
    Joined:
    Jan 18, 2015
    It's not down to just MA current readings, it's in conjunction with the plate voltages. You have to ensure you don't exceed the maximum plate watt dissipation under all operating conditions, if you do, you red plate and destroy the tube in rather short order. The plate has to shed the heat from the watts applied to it, and there's the critical limit. There is also a limit you can go with plate voltage.

    All that's in the tube data sheets that can be found here : http://tubedata.milbert.com/sheets69.html

    MA (converted to amps) x DC Volts = Watts (to be dissipated by the tube plate)
     
  2. Guinness Lad

    Guinness Lad Silver Supporting Member

    Messages:
    15,514
    Joined:
    Nov 11, 2004
    Location:
    On top a mountain of Chocolate Chips
    Anyone sets bias based upon a mathematical value is a chump.

    There is a range tubes can function within, not every tube sounds best the supposed magical value of 70% as claimed by some. Some sound better at 50 and other 70, depends upon the amp and what it's doing.

    When you turn up the current I have noticed this every time no matter what the amp - louder first, then the tone will thicken, if you keep going it tends to get harsh, then finally red plating.

    I generally like right before the thickening phase as the clean sounds far better to me.
     
    ledzep618 and GT100 like this.
  3. ken374

    ken374 Member

    Messages:
    6,096
    Joined:
    May 4, 2011
    Location:
    South East Michigan
    Right at 60% works best on my amp
     
  4. HotBluePlates

    HotBluePlates Member

    Messages:
    2,659
    Joined:
    Mar 27, 2017
    I wrote about all you need to know about biasing here & here. But if you want the short answer...

    If you bias to cold the amp will sound trashy during low volume playing (crossover distortion). Big signals will push a cold-biased amp out of the range where crossover distortion tends to happen. If you bias too hot, the tubes will redplate & try to melt when you play loud (or at idle if you bias way, way, way too hot).

    There's a large range in between those 2 extremes of acceptable performance. Biasing output tubes isn't too critical, though the designer may have had a particular bias in mind for maximum clean output power.

    That's bad luck! Maybe the Fryette is hard on tubes, but toss a coin in the air and it will land on 25 TGP threads saying how this- or that-brand new production tube is crap (and just as many saying that yesterday's awesome tube to crap and that yesterday's dog is now the only good tube).

    Since 1993, I have had exactly 1 output tube fail on me. I've had maybe 50-60 amps in that time, and currently own 11. The one failed tube happened sometime around 2003, IIRC. Plate-to-heater (pin 3 to pin 2) short-circuit in the tube. I use only old production American/European tubes, except when an amp I bought already has new tubes in them. When I feel like replacing, the new-production tubes get replaced with old production.

    Make your judgment call about your experience vs mine.

    Or the tubes are defective. I was once contacted by an amp manufacturer to help resolve a problem in their amps. The issue was there with every 12AX7 they installed in one socket of their amps. They assumed there must be a design problem. The real problem turned out to be a case of 100 new-production 12AX7's
    that all had the same defect!

    That's comparing apples to bourbon.

    The Fryette 2/90/2 uses KT88's, the Engl Invader 2 uses EL34's and the Engl Ironball uses EL84's. You can't compare the same tubes, to see if the issue is tube or amp related. Even from a single manufacturer, it's conceivable to have a defect in KT88's but not EL34/EL84, much less manufacturer-to-manufacturer.

    But if biasing "cold" allows the tubes to live, then what's the problem? As I posted in my past links, that's sometimes necessary by virtue of the amp's design.
     
    gearscrubs likes this.
  5. turtleboy3

    turtleboy3 Member

    Messages:
    544
    Joined:
    Sep 23, 2009
    Can you guys recommend a good book about maintenance and amp biasing?
    Im a newbie. Thanks
     
  6. GT100

    GT100 Member

    Messages:
    2,985
    Joined:
    Nov 6, 2009
    Location:
    Stygia
    Yes, I bias by ear and what the numbers/measurements (plate dissipation) say are safe for the the tube in question.
    "Hot biasing" isn't always better....

    Lloyd
     
    ken374 likes this.
  7. easyed

    easyed Supporting Member

    Messages:
    926
    Joined:
    May 14, 2017
    Location:
    Seattle
    I'll agree and add, set by the numbers to 70% and turn down till you feel it loses some of the tone or dynamics and then move it back up a bit. I'll call the 'start with the numbers then bias by ear'.

    70 percent is generally considered safe maximum; it's not necessarily best. It's your tone and your ears.
     
  8. Badstrat

    Badstrat Member

    Messages:
    637
    Joined:
    Dec 7, 2013
    Location:
    Newcastle,Australia
    My gigging amp has a plate voltage of 640 Vdc so I bias it at about 50%. Sounds great.
     
  9. Fuego

    Fuego Member

    Messages:
    77
    Joined:
    Feb 8, 2010
    The tunes don't keep living. Just a little longer. The original tubes I had in there lasted over 10 years. As soon as I changed them, boom boom boom
     
  10. gtr777

    gtr777 Supporting Member

    Messages:
    1,301
    Joined:
    May 30, 2005
    Location:
    Long Island NY
    I always measure the plate voltage and then Bias to 60%. I have been told that some amp manufacturers purposely bias the amp cold to ensure the tubes make it through the warranty period but who knows....
     
    Last edited: Jan 7, 2018
    CallEOD and ken374 like this.
  11. ledzep618

    ledzep618 Supporting Member

    Messages:
    4,605
    Joined:
    Oct 17, 2011
    What amp is that? That seems very high.
     
  12. DonaldDemon

    DonaldDemon Member

    Messages:
    6,257
    Joined:
    Nov 13, 2007
    Location:
    Fear and Loathing, NJ
    What brand of KT88’s were you using? Not all of them can withstand the higher plate voltage. I use Shuguang Ruby V-STR for high plate voltage amps.

    When I bias amps I generally take plate voltage measurements first and then use that plus the max dissipation rating for the tube as per manufacturer to calculate the bias range of 60-70% dissipation at idle. I start at 60% and move up to 70% until I hear what I like, which is fullness of signal without muddiness. I know my particular amps well enough now where I don't have to do that but even with newer ones I usually end up in the 60-65% range.

    Two of my amps are fixed bias and are set to run cold, in which I trust the manufacturer because I wouldn't want them to be any more gritty than they already are.
     
  13. CallEOD

    CallEOD Member

    Messages:
    2,122
    Joined:
    Jul 24, 2015
    I got a new amp & retubed last week. I just rebiased now and it was a bit cold but my line voltage is down about 3 VAC since last time, Even tho i have a matched pair there is a 3ma differential between the two.
    That means when my line voltage swings 3 volts higher my amp will be biased a tad hot but only one tube which is better than both being cold.
    Still sounds Awesome.
     
  14. HotBluePlates

    HotBluePlates Member

    Messages:
    2,659
    Joined:
    Mar 27, 2017
    By your description, it's not the amp but the tubes you're using now.
    New tubes: boom, boom, boom.
    Old tubes: lasted 10 years.
    Amp: Didn't change.

    +1. Except Fryette's manual appears to specify use of Sovtek/New Sensor KT88's (the Sovtek, EH and SED KT88's mentioned in the manual are all sold by the same company, and might be manufactured in the same factory).
     
    Last edited: Jan 8, 2018
    DonaldDemon and CallEOD like this.
  15. Badstrat

    Badstrat Member

    Messages:
    637
    Joined:
    Dec 7, 2013
    Location:
    Newcastle,Australia
    It's a Baez 100 watter made in Newcastle Ausralia in the late 70s.
     
  16. Fuego

    Fuego Member

    Messages:
    77
    Joined:
    Feb 8, 2010
     
  17. HotBluePlates

    HotBluePlates Member

    Messages:
    2,659
    Joined:
    Mar 27, 2017
    Sorry, spell-correct mangled my post. I meant "Sovtek" in my prior post. So, what brand of KT88's were you using this time around when you had problems?
     
  18. EastCoastRocker

    EastCoastRocker Member

    Messages:
    1,233
    Joined:
    Sep 30, 2009
    I concur with the above. When I first got the DSL it was biased at 90mv. I played with a lot of settings and kept comming back to 80mv, where it has stayed ever since.
     
    CallEOD likes this.
  19. ken374

    ken374 Member

    Messages:
    6,096
    Joined:
    May 4, 2011
    Location:
    South East Michigan
    Im going by what my tech told me and what I hear is best.
     
  20. ken374

    ken374 Member

    Messages:
    6,096
    Joined:
    May 4, 2011
    Location:
    South East Michigan
    Friedman set there el34 at 32 480 PV
     

Share This Page