I'm really confused about bias

Discussion in 'Amps/Cabs Tech Corner: Amplifier, Cab & Speakers' started by scottypeel, May 14, 2003.


  1. scottypeel

    scottypeel Member

    Messages:
    140
    Joined:
    Mar 4, 2002
    Location:
    sw florida
    this issue just won't go away for me. Cesar Diaz told me to bias his amp at 52mA running classab ss rectifier. George Alessandro runs his class A amps at 60mA with 6V6's, and told me as long as I use good NOS tubes, I could go higher! I just bought a new amp that runs 4 output tubes with a GZ34 rectifier. The amp has a separate bias trim pot for each pair of tubes. A letter from the guy who built it says it can run 6V6 or EL34 (he doesn't mention 6l6) with a rebias. I started messing around with all my 6V6, EL34 and 5881 tubes, and with a little tweaking, I got a pair of EL34 and a pair of 6V6 within 3mA - sounded great! Just for fun, I replaced the EL34's with a pair of 5881's - the 5881's pull 42mA while the 6V6's pull 22mA - no hum - great tone. ???????? I also replaced the GZ34 with a 5Z4G/CV1863 - sounds great! I'd really like to hear some thoughts from Hogy, unquiet, mook, and James Peters, etc, and some of the other great builders, tinkerers, gearsluts and techs that hang out here.....please :)

    just to be clear, with this new amp I'm running a pair of 6V6 and a pair of EL34 at the same time - the pair of 6V6 pulling 22mA, and the pair of EL34 pulling 26mA - then put the 5881's in and they pull 42mA with the6V6's still running 22mA at the same time and it sounded great - much smoother breakup and great cleans - thanks in advance!
    p.s. - I dont know the plate voltage - I'm estimating between 350-400V?
     
  2. Well, there are two things to keep an eye on, the plate voltage (pretty important) and the screen voltage (really important). So long as your voltages are low, your current draw through the tubes can be high and not burn up. If the plate voltage is high, you can still run your tubes safely if the screen voltage is knocked down. If your screens are at high potential as are the plates, keep an eye on the draw...especially with wimpier tubes like mosty of the recent(ish) 6V6s.

    DS
     
  3. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    I held back from commenting here for a while, because my view is a little 'controversial'... but not too many other replies yet, so -


    You do absolutely need to know the plate voltage before the current measurements mean much. This is the whole problem with the 'biasing by current' method. Dexter is also right about the screen voltage, but in most common amp designs it is as (or very nearly as) high as the plate voltage.

    You also need to know what the designed operating point of the amp is - more complicated, and confused by the 'Class A' issue. Most amps that claim to be so aren't, and if you bias them as if they were (up to 100% max dissipation at idle), you will probably fry the tubes if as usual it's really a Class AB amp.

    Even so, those recommended currents sound very high. 60mA for 6V6s would require the plate voltage to be 200V or under for running within the maximum rating (12W) even in Class A. Also, if the amp is actually running in true Class A (which I believe this amp is), there's no advantage to pushing it higher - quite the opposite, it will clip the forward phase of the waveform too soon - this is a common problem with BF/SF Fender Champs and doesn't sound good, as well as unnecessarily stressing the tube.

    52mA with 6L6s in an amp with only 350V plate voltage is probably OK - it will most likely be operating near to Class A at that.

    The different tube types have different operating requirements though, so with a given bias supply voltage it is normal for the various types to run at different currents - that's the main reason you shouldn't just go substituting tube types without checking what's going on.


    I do not understand the current fashion for biasing amps as hot as possible. It does not improve the 'tone' to me after the point is reached where the crossover distortion is not a factor - usually just a little higher than when you can actually hear the 'notch' disappear. It does increase the gain slightly, and so can appear to make the amp sound fuller and 'better', but you can achieve the same thing by turning up a bit. If there is any tone change, to me it's for the worse - too mushy and saturated, although again that is really a symptom of the gain in the power stage.

    And without question, it shortens tube life.

    This to me is a total, pointless waste - especially as the good NOS tubes are not being made any more. I try to bias as cold as is possible for a good sound.

    That's if I adjust the bias at all. As I've said before, I agree with Randall Smith - if the amp is properly designed and you fit good tubes, you shouldn't need to adjust it. There is a reasonable range of correct bias (exact +/- 1mA settings are irrelevant in almost all cases), and anywhere in that range is fine.

    Setting to an exact number of mA without knowing a lot more about both the exact amp and tubes is like setting your tire pressures to an exact number of psi without knowing what type they are or what sort of car they're going on. Just being able to read a number with a meter doesn't mean much. Everyone seems to understand this with cars, why is it so difficult with tubes?

    NB, this is all my personal opinion and I expect to be disagreed with. But this approach works for me - it makes amps sound 'good' and work reliably.
     
  4. scottypeel

    scottypeel Member

    Messages:
    140
    Joined:
    Mar 4, 2002
    Location:
    sw florida
    unquiet, thanks for the info, and I agree. I've messed with bias on so many amps, and one of the reasons I'm confused about bias is that I've never noticed much of a tonal change with different bias settings within a 10mA range or so, so long as those bias settings are within a reasonable range - i.e. - not biasing EL34 at 5mA,etc.
    Also, I was confused because I tend to like most tubes biased relatively cold. For example, I ended up biasing the amp in question in this thread at 22mA with 4 5881's. It also sounds good at 42mA, but why burn the tube so hot??? They will last much longer at 22mA, and sound great.
    One of the only differnces I've ever noticed with biasing very hot is that a good slightly pushed clean tone is available - but when pushed into overdrive, that pushed sound becomes "overloaded" sounding, to my ear, and is very unpleasant, especially in the lower frequencies.
     
  5. Garey

    Garey Member

    Messages:
    722
    Joined:
    Jan 4, 2002
    Location:
    Northern California
    FWIW, George Alessandro's lower powered amps are true Class A.

    My Black N Tan has a plate voltage of 250v, I've used 6v6's, 6l6's, 5881's, EL34, EL37, KT66....all without a problem, even though idle current is around 60ma on the 6v6's and 90ma-100ma on the EL34, KT66 etc. I've never lost an output tube on this amp, even when cranked for a couple hours straight....and I haven't noticed any premature aging of tubes......unlike the Matchless amps I've had that claimed to be class A, but were just too damn hot and ate tubes up!

    In George's class AB amps...I've only owned a Redbone....the tubes are running a touch hot, but nothing extreme. I wouldn't use Siemens in there....but anything else is fine.

    Aloha, Garey
     
  6. scottypeel

    scottypeel Member

    Messages:
    140
    Joined:
    Mar 4, 2002
    Location:
    sw florida
    thanks hogy - very deep indeed for so early in the morning:)

    last issue:

    I've never, in my amateur trials and tribulations with NOS tubes, never ever seen a tube plate glow - at least I dont think I have. I always see what I think is the filament glow, which is the long rod running through the middle of the tube - but never see the large flat plates of any tube glow red - and I've run some pretty hot bias currents - does it glow right away at idle, or does it have to run hot for a period of time before it glows?
     
  7. tanawana

    tanawana Member

    Messages:
    285
    Joined:
    Jan 4, 2002
    Location:
    Youngstown, OH
    If I understand correctly, you can ----> Try turning out the lights, looking at the tubes and slowly move into the range till you see a slight glow on the tubes. This helps you see it easier. Be very careful I would suggest just in case.
     
  8. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    Yes, a dark room helps a lot.

    You shouldn't see any plate glow, especially at idle - in a Class AB amp, the power dissipation is higher under load (which is why you often see "70% of max dissipation" quoted as a good operating point - that's actually the upper end of the safe range in most Class AB amps).

    A true Class A amp (very rare, but having read Garey's measurements from his Alessandro I'm sure that is one) runs at the same power dissipation irrespective of signal level and so can be biased up to 100% of the max if you want. But that is still no guarantee of the correct operating point - you really need a scope to set that, since the only 'correct' setting for true Class A is when the waveform is exactly centered between cutoff and clipping.

    But hogy is very right... this is not hi-fi. 'Correct' may not mean much if it doesn't sound how you like it. Whatever sounds right to you is right, so long as you're not outside the safe operating zone (and even then I'm sure there are some people who will sacrifice reliability for 'tone' as they perceive it).

    Personally, I was very pleased to hear scottypeel's observations too - I don't feel so alone now! I do indeed prefer the sound of an amp biased towards the colder part of the range if there is any difference at all - the one exception is in a 'near Class A' (AKA hot-biased Class AB) amp like a Vox, when I want that semi-overdriven jangly crunch.

    But normally I find I like the sound best when the amp is running conservatively - cool bias, matched load, voltages within normal ranges, all that sort of stuff. This also gives the best reliability. Maybe a coincidence, maybe not...
     
  9. scottypeel

    scottypeel Member

    Messages:
    140
    Joined:
    Mar 4, 2002
    Location:
    sw florida
    exactly, the only time I've liked a hot bias point is for this sound - edgy clean is what I call it-

    also, IMHO, amps that are made right, i.e. Komet, Alessandro, and some others, sound best when operating at "normal" settings because they are designed correctly - I think maybe I like a colder bias point because it provides a slightly smoother more natural sounding overdrive when the amp is pushed - but again, only in amps done right - when I bias a reissue Marshall or other poorly made amp, the results aren't as predictable.
    Thanks for the input everybody - rock on!
     
  10. reaiken

    reaiken Member

    Messages:
    1,773
    Joined:
    Jan 4, 2002
    Location:
    Greenwood, SC

    Actually, in a true class A amp, the plate dissipation is highest at no signal, and *decreases* as signal is applied - essentially the power dissipated in the tube is transferred to dissipation in the load. It is the average plate *current* that stays the same, but since the instantaneous plate current and plate voltage are out of phase (into a resistive load), the product of the two is the highest at no signal (zero plus the idle point dissipation) and lowest at max signal (idle dissipation minus the product of plate voltage and current).

    This allows you to safely bias a class A amp at full dissipation, since it is guaranteed to be lower at full output or anywhere in between.

    As for class AB, I agree that biasing at max allowable is not always the best sounding spot. Max should be considered just that - a maximum, not the recommended bias point.

    Randall Aiken
     

Share This Page