Converting fixed bias to cathode bias

Discussion in 'Amps/Cabs Tech Corner: Amplifier, Cab & Speakers' started by CitizenCain, Feb 20, 2008.

  1. CitizenCain

    CitizenCain Member

    Messages:
    4,831
    Joined:
    Jan 10, 2004
    Location:
    Magdalena, NM, US
    I read myself in circles last night researching cathode biasing. Seems like maybe it's not as straight-forward a conversion as I thought?

    From what I read (what I remember at least) cathode biasing is fairly simple in a class A amp, but gets more complicated to do in a class AB amp?

    I have a Sundown amp that I'm using as a "live it and learn it" project. I converted the PCB into a hand-wired tag board. Now I'd like to investigate cathode biasing. The amp uses a 2x 6550 power section, fixed bias in class AB. How do I figure what size resistor to use if I wanted to convert to cathode bias? I read last night that with the tubes alternating cycles as they do, it's harder to run as a cathode biased amp...or something.

    Most of the sites I find when googling are hi-fi based. They are a lot more worried about uber-clean power and stuff. What's the scoop for a guitar amp? Straight resistor or bypassed with a cap? I'm just plain confused now.

    Any hints?
     
  2. CitizenCain

    CitizenCain Member

    Messages:
    4,831
    Joined:
    Jan 10, 2004
    Location:
    Magdalena, NM, US
    Thanks, John. That's basically what I had planned to do, but I started reading about how push-pull amps load the cathode resistor differently because of the tube cycling or something like that and just got confused. I can handle the calcs and zero in on the right sized resistor through measuring.

    I thought it was as straight-forward as that, and you seem to indicate that's true. I'm much relieved! :D
     
  3. CitizenCain

    CitizenCain Member

    Messages:
    4,831
    Joined:
    Jan 10, 2004
    Location:
    Magdalena, NM, US
    I think another thing that was throwing me off was that I was intially researching a SE conversion as well, and I started to intermingle the two concepts.

    Basically, I can just do the math to derive the static dissipation same as with fixed bias?
     
  4. THLH

    THLH Member

    Messages:
    156
    Joined:
    Jan 6, 2002
    Location:
    Flat Rock, MI
    It's kind of hit or miss. You'll need to get a bunch of resistors and find the ones that give you a usable range. Even doing the math is no guarantee that a resistor of that value will work. Depends on the tubes too, some run hotter than others.
     
  5. Wakarusa

    Wakarusa Member

    Messages:
    1,477
    Joined:
    May 7, 2004
    Location:
    Indianapolis, IN
    You could sit down with the tube characteristic curves, adjusted for the screen voltage you've actually got, the OT impedance, plate voltage, etc. as well as an assumption of where in the AB1 operating range you want to end up and calculate (both graphically and with pencil and paper) pretty much a spot-on value for the cathode resistor.

    But in your case, most of the work is already done for you :)

    Bias the amp using the existing fixed bias setup to wherever you think it sounds best within dissipation limits. Now measure the DC voltage on pin 5 of one of the output tubes. Next, using whatever method you've used to bias the amp (1 ohm resistor in the cathode circuit, Bias Rite, whatever), figure out the cathode current. The goal is to choose a cathode resistor that will develop the same voltage on the cathode that you measured on pin 5 for the cathode current you measured. After that, it's a simple application of Ohm's law.

    Let's say you got it all set up fixed bias and measured 42VDC at pin 5 and 50mA on the cathode. Ohm's law says I = E/R, so a quick algabraic manipulation gets you R = E/I. Since the cathode resistor will be shared between two tubes, the 'I' you plug in is double what you measured.

    42V/(2 * 0.05A) = 420ohm

    The closest standard value in this case is 430ohm, so that's what you'd use. Since we also know that Power (in watts) equals the product of voltage and current:

    42V * (2 * 0.05A) = 4.2W

    The resistor will need to dissipate 4.2 watts. For safety, you double that and buy the 10W part.

    And yes, you'll absolutely want to bypass it with a capacitor. A 22uF/63WVDC part ought to work fine. The bypass cap makes the cathode resistor look more like a current source (what you want here) and also has a strong influence on the frequency response of the stage.
     
  6. VacuumVoodoo

    VacuumVoodoo Member

    Messages:
    1,567
    Joined:
    Jan 9, 2005
    Location:
    Gothenburg, Sweden
    The above is a good first approximation. In fixed bias configuration power tube cathodes sit at 0V (the few tens of mV on bias measuremnt 1 Ohm resistors don't matter). When reconfigured to cathode bias the cathodes will be lifted from 0V to couple tens of Volts above ground. This means that plate to cathode voltage will be lower by that potential - it's analogous to reducing B+ by the same amount so in effect your tubes will be running colder than in original fixed bias configuration. This effect is compensated to some extent by DC feedback action of bypassed cathode resistor but not always enough.

    As you probably want your amp to stay longer in class A before transitioning smoothly to B you need to run the tubes hotter, more like 80 to 85% of max plate dissipation. What you need to do is reduce the cathode resistor values.

    I would suggest a 330 Ohm 10W resistor in series with 100-150 Ohm 4W wire wound potentiometer wired as variable resistor. This will let you adjust idle current to a good sounding level.

    It's also good to keep in mind that most PP configuration require different Raa (plate-to-plate load imedance). Cathode biased power stage will want to see a higher Raa than a fixed bias one. If your OT has secondary winding with taps for different speaker impedances then connecting a speaker to correspondingly lower impedance output i.e. 8 Ohm speaker to 4 Ohm output may bring Raa closer to value required for proper operating point in a "deep" class AB (sometimes in the past denoted as "Ab").

    That was engineering, now, whatever sounds good is good as long as you stay within tubes safety limits.
     
  7. CitizenCain

    CitizenCain Member

    Messages:
    4,831
    Joined:
    Jan 10, 2004
    Location:
    Magdalena, NM, US
    Fantastic info guys! I can do this now! :D

    I'd just about given up on this project for now, thinking I had to study up a lot more. Well, actually I do have to study up a lot more, but that's beside the point [​IMG]

    I understand all of the above. This is gonna be great! Love the variable pot idea, and in fact I have a 250k 5w Clarostat pot here in my leftover parts bin.

    Thanks!
     
  8. phsyconoodler

    phsyconoodler Member

    Messages:
    4,351
    Joined:
    Jul 11, 2005
    Location:
    calgary canada
    I wouldn't be tempted to use the pot permanently,but just as a test to get the right size cathode resistor.What amp are you wanting to switch to cathode bias?
    Many amps are already cathode biased and a quick check of existing schematics can save a lot of trial and error here.
    Cathode biasing is far easier than fixed bias,which is very easy once you know the basics.I suggest getting a good book like Gerald Weber's books or Funk's or something that can help you easily understand what you are trying to do and the results you can expect.
     
  9. CitizenCain

    CitizenCain Member

    Messages:
    4,831
    Joined:
    Jan 10, 2004
    Location:
    Magdalena, NM, US
    phsyconoodler, I have a Sundown head that I've been using as a learning tool for tech work. I've been biasing my own amps for quite a while, all fixed bias though. Last year I decided I want to learn more about the inner workings, just for fun and my own education. I've taken the PCB out of the amp (it has some issues anyway) and converted the clean channel to a tagboard layout that I designed myself. Now I want to experiment with cathode biasing, hence this thread :D

    Here's the guts of my Sundown so far. I've got a hum problem I'm tracing down now. I think it's the filter caps that I reused trying to save a couple bucks. The amp was working beautifully for about 6 weeks, then one day just starting humming all the time. Checked out everything else I could think of, I'm going to replace the caps next week with new ones. After that, bias city [​IMG]

    [​IMG]
     
  10. VacuumVoodoo

    VacuumVoodoo Member

    Messages:
    1,567
    Joined:
    Jan 9, 2005
    Location:
    Gothenburg, Sweden
    Nice! Check your main chassie grounding point, old stuff can corrode and lose grounding.
     
  11. CitizenCain

    CitizenCain Member

    Messages:
    4,831
    Joined:
    Jan 10, 2004
    Location:
    Magdalena, NM, US
    Yep, checked the grounding points and everything else I could think of. Everything except the transformers and filter caps is new :D Nice clean lugs under a transformer bolt give solid, new ground points. The new caps only cost me $20 and I'll feel better if I replace them anyway. If it still hums after that, I'll start going into the circuit section by section to find out where it's coming from.
     

Share This Page