Amp Designer said this (about bias-rite/bias king, etc)

Discussion in 'Amps/Cabs Tech Corner: Amplifier, Cab & Speakers' started by kannibul, Jan 11, 2006.

  1. kannibul

    kannibul Member

    Messages:
    239
    Joined:
    Apr 27, 2004
    An Amp Designer said this:
    [person complaining about an amp not being able to be biased hot enough]...was using a bias king or something similar! These questionable tools are only any use in 50 watt amps that run on two output tubes. The readings on a 100 watt amp will be wrong, causing the amp to be biased well into the danger zone when using one of these tools! I heard that the person in question took his [amp] to a tech to have a resistor changed to enable him to bias the amp hotter. I would love to know if his amp still works? Its a good job that [the amp's] 100 watt transformers have a full half an inch more stack (both mains and output) than pretty much every other 100 watt amp out there (notice the weight of em?) which means its just gonna be tubes and fuses that this person will be buying at a rapid rate!

    Any merit to this?
     
  2. Blue Strat

    Blue Strat Member

    Messages:
    30,023
    Joined:
    Aug 2, 2002
    Location:
    Sterling, VA (not far from Washington DC)
    I can't think of any reason why a correctly functioning bias probe wouldn't work with ANY amp. How the readings are used is another question.
     
  3. rooster

    rooster Member

    Messages:
    2,123
    Joined:
    Mar 9, 2005
    Location:
    Arizona
    Not in my book, but I didn't write the book. I could be wrong. But I don't think so.

    rooster.
     
  4. fullerplast

    fullerplast Senior Member

    Messages:
    6,792
    Joined:
    May 5, 2003
    There's no merit to his comments, as they are written here. These probes all put a 1 ohm resistor in between the cathode and ground and the meter reads the voltage across that resistor. Since it is a one ohm resistance, there is a 1:1 correspondance between mV and mA.

    All these devices do is measure the current for each tube. It doesn't change anything in the circuit.
     
  5. conundrum

    conundrum Member

    Messages:
    695
    Joined:
    Sep 5, 2005
    Ade has an interesting way of constructing his sentences. I feel like he had the correct notion, he just couldn't type it out coherently on the Orange forum.
     
  6. kannibul

    kannibul Member

    Messages:
    239
    Joined:
    Apr 27, 2004
    Did you read my reply?
     
  7. donnyjaguar

    donnyjaguar Member

    Messages:
    4,242
    Joined:
    Nov 4, 2005
    Location:
    Canada
    I use a 10R resistor on each cathode myself. I'm not convinced that you can get an accurate current reading on a 1R. At 40mA you will measure 400mV across the 10R. That means the 10R is dissippating a whopping 16mW, which shouldn't eat into your sonic mojo.

    DJ
     
  8. fullerplast

    fullerplast Senior Member

    Messages:
    6,792
    Joined:
    May 5, 2003
    Why?
     
  9. kannibul

    kannibul Member

    Messages:
    239
    Joined:
    Apr 27, 2004
    I'd guess that if you have a crappy meter, that could be a problem...

    :)
     
  10. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    It might if you're talking about fitting it permanently into the amp. That's enough to introduce a bit of cathode bias, which might or might not be a good thing tone-wise.

    1 ohm is perfectly adequate to get a good current reading; the meter is plenty accurate enough at 40mV. Metering 400mV on a typical DMM means you have to use the 2000mV range not the 200mV, so you're back to exactly the same overall accuracy (even if you're able to read one more decimal place, this does not mean the reading is more accurate, BTW - just more precise, which is not the same thing). In any case, cathode readings are inherently inaccurate for biasing anyway because they include the screen current, typically a few mA which is about 10% of the total - but not known. An error of +/- 1mA at 40mA is only 2.5%.
     
  11. kannibul

    kannibul Member

    Messages:
    239
    Joined:
    Apr 27, 2004
    Mine doesn't really have a range - just stick it on and it tells you what you want to know with "Radio Shack" accuracy!
     
  12. Shea

    Shea Member

    Messages:
    540
    Joined:
    Sep 7, 2004
    That inaccuracy doesn't bother me, because I always do my final adjustment by ear anyway. I usually find a spot where it sounds juuuuuust right to me, and rarely is that spot right at 70% of max plate dissipation. Often it's a tad colder. I don't think there's anything magic about that 70% number.

    Of course, that "spot" I'm aiming for could be a moving target, because it might change when the amp warms up further or cools down, or if I take the amp to another building with slightly different wall voltage. But I like to aim for it anyway.

    Shea
     
  13. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    If it's auto-ranging it still has a 'basic accuracy' which is independent of the quantity being measured.

    Don't confuse precision with accuracy - they're not the same thing. Suppose you're measuring a current of 50mA (true value) and you have two meters, one which tells you that the current is .050A and the other tells you it is .0492A. The first one is less precise - one fewer significant figure - but more accurate.

    And you really don't need any more than 1% accuracy for any of this - the other circuit variances are at least that bad, some several times as much. Tubes aren't matched to that degree usually, and even the two sides of the OT may be more than 1% different... let alone not accounting for screen current. A 1%-tolerance 1-ohm resistor and a cheap DMM is more than adequate.

    If you're getting hung up about measuring idle current to any greater precision than that, then (with respect to anyone who thinks this may be aimed at them :)) they don't fully understand what they're doing. That doesn't mean bias metering is worthless at all, just that it's not a be-all-and-end-all method of adjustment. It's very useful for checking that an amp is in a safe operating range (like Shea says the often-used 70% figure is a maximum, not necessarily a target), for checking tube matching, and for returning an amp to a particular setting where it sounded good. It's also better to set current to the 'right' value (even if it isn't strictly accurate) than a total shot in the dark, which is what you get if you just measure the bias supply voltage like in the old days...

    I usually set bias by ear too BTW, then check afterwards that the current is OK.
     
  14. Wakarusa

    Wakarusa Member

    Messages:
    1,477
    Joined:
    May 7, 2004
    Location:
    Indianapolis, IN
    You and I must be using different tube spec sheets -- and I keep a pretty impressive variety of 'em. All of mine give a specific value for max plate dissipation. Additionally, some give a spec for max screen dissipation and some even give a max limit for cathode current.

    None of 'em have a dissipation rating for "the total dissipation of the tube". Not one.

    The 70% maximum value is typically based on plate dissipation and further assumes class AB1 bias conditions.
     
  15. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    ...and also some self-modulation of the signal, which is why I'm not sure it would necessarily be a good thing.

    Not arbitrarily, but I agree that you're not accurately biasing it. That's exactly my point as to why bias probes which measure total tube current at idle are not an accurate (or even precise :)) means of setting your amp, which is why using one to set your amp at 70% of max plate dissipation, or whatever other magic number, to the extent of getting worried about 1% errors in meter readings, is quite meaningless.

    The only way to bias an amp accurately is with a scope, but I'm not sure that's really the best way for a guitar amp. IMO by ear is the right way, with a current check for safety - they are musical instruments not hi-fi equipment and setting the 'correct' operating point is IMO secondary to simply making it sound the way you like it. After all, we like distortion, don't we :).

    As Todd says the tube ratings are for plate and screen separately, not combined. In practice it doesn't really matter, since by setting the cathode current to give a 'plate' dissipation of 70% while including the screen current, in reality it's a bit less. This is a good thing actually, since it builds in a safety margin (and most amps sound better a bit cooler than 70% anyway, contrary to popular belief). It just makes nonsense of the idea that setting to 1% is critical.
     
  16. kannibul

    kannibul Member

    Messages:
    239
    Joined:
    Apr 27, 2004
    I think of 70% as somewhat of a do-not-exceed number - whereas if you exceed it, understand you will have accellerated wear in comparison.
     
  17. Shea

    Shea Member

    Messages:
    540
    Joined:
    Sep 7, 2004
    Rope-A-Dope, John has merely pointed out that tube spec sheets give separate ratings for plate dissipation and screen dissipation. Do you disagree with that statement? Do you have a link to a spec sheet that doesn't list them separately?

    Here is one that does: http://www.triodeel.com/6l6gc_p1.gif

    See at the bottom?

    "GRID-No.2 INPUT........5 max. watts
    PLATE DISSIPATION...30 max. watts"

    Shea
     
  18. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    No. Self-modulation will introduce distortion (harmonic distortion, not clipping of course). Whether that's desirable, good-sounding distortion or not is purely subjective, but don't count on it. Some early SF Fender amps have resistors like this in the cathode path and are not thought to sound as good as the ones that don't.

    The ratings are quote as plate dissipation and screen dissipation. Like Todd, I've never seen a single tube data sheet where the total tube dissipation is quoted - it doesn't need to be since you can easily work it out by adding the two, and from a reliability point of view it doesn't matter - exceeding either (especially the screen rating, which is more critical) will cause trouble, so it's the individual ratings that are important not the combined one. Yes, the screen is certainly inside the plate, and that will be taken account of in the plate dissipation figure, since the screen rating is always a small fraction of the plate's - and the plate rating is not as critical anyway. You don't need to subtract the screen power from the plate (as you said in your earlier post) to stay under the limit.

    In any case, it really doesn't matter to this degree of accuracy. Running at 71% will not cause instant tube failure and running at 70% will not give infinite tube life. 10% either way will make a difference, but even then there is no sudden cutoff.
     
  19. Shea

    Shea Member

    Messages:
    540
    Joined:
    Sep 7, 2004
    I'm not sure what you mean by "self-modulation" here. When you have an unbypassed cathode resistor, the voltage at the cathode will swing up and down in phase with the signal at the grid, which results in a negative feedback effect that is known as "current feedback." If anything, they should reduce harmonic distortion.

    That's talked about here, starting at the last paragraph on the first page: http://www.tubecad.com/2005/February/04Feb2005.pdf

    Shea
     
  20. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    Certainly the voltage at the cathode will move in phase with the input signal strength, but since that alters the bias on the tube, does that then alter the plate current linearly and in phase too? What happens if there is already a conventional negative feedback loop via the OT? I'm not at all sure that the overall result will be to reduce distortion. I'll have to check up on the theory (for some reason that PDF wouldn't open for me), but I can defnintely say that those SF Fenders with cathode resistors (a much larger value than we're talking about here, certainly, but that doesn't mean the effect is non-existent with a smaller value) sound more compressed and rattier than they do without... 'worse' to most people. It sounds like 'more' distortion to me, although I could be wrong.
     

Share This Page