Cathode bias checking techniques

Discussion in 'Glowing Bottle Tube Amp Forum' started by bluerolo, Dec 9, 2019.

  1. bluerolo

    bluerolo TDPRI Member

    Posts:
    34
    Joined:
    Jan 19, 2016
    Location:
    NJ by the sea
    Hoping for some help/opinions on how best to determine dissipation on a cathode bias amp. Working on a Harmony H415, 2x el84. I found a thread here that gave an easy to complete method, but then compared readings to other methods and came out with different results. Shooting for 100% dissipation at idle.

    Method 1 was:
    a. measure voltage from pin 3 el84 to ground...10v
    b. divide pin 3 voltage by the cathode resistor value...10/129 ohms = .0775ma
    c. measure plate voltage pin 7 to ground = 333v - cathode voltage of 10v = 323v
    d. 323v x .0775ma = 25.03W
    e. 25.03w/2 = 12.59w per tube...on a 12w max dissipation tube.

    Method 2 was Rob Robinette's website based calculator that used:
    a. Plate voltage pin 7 to ground = 333v, cathode voltage to ground = 10v
    a. plate to cathode v = 323
    b. plate current (which I assumed meant pin 7 el84 to pin 8 5Y3) = .033ma
    c. .033ma entered into his calculator = dissipation of 10.8w or 89%

    So as you can see if I'm trying to get 100% dissipation at idle for cathode biased amp I've got 2 conflicting measurements ...am I doing something wrong?
     
    Last edited: Dec 9, 2019
  2. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,268
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    Method 1 is missing a step.

    Once you've calculated cathode current (77.5 mA), you have to account for what portion of the cathode current goes through the plates and which portion goes through the screens.

    Rob's calculator automatically calculates something like 95% of the cathode current goes to the plates and 5% goes to the screens for some power tubes, and something like 90% of the cathode current goes to the plates and 10% goes to the screens in some other power tubes. These are estimates based on the tube data sheets of different tubes.

    77.5 mA for a pair of EL84s sharing the cathode resistor. Divide by 2.
    38.8 mA for a single EL84. That's the cathode current. Multiply by 90%.
    34.9 mA plate current. Multiply by plate-to-cathode voltage.
    11.3 W
    11.3 W / 12 W = 94%

    ^^^ that's my math. Here's what the calculator on Rob's site says with the same input:
    98%. Pretty close.

    Screenshot from 2019-12-09 17-37-57.png
     
    King Fan and rolandson like this.
  3. peteb

    peteb Friend of Leo's

    Posts:
    4,200
    Joined:
    Apr 25, 2003
    Location:
    Cascadia

    This sounds right, and you are very close. The plate current is 5-10% less than the cathode current.



    Every thin in method 2 sounds fine, except how did you determine that the plate current is 0.033? This should be 90-95% of the cathode current.
     
  4. bluerolo

    bluerolo TDPRI Member

    Posts:
    34
    Joined:
    Jan 19, 2016
    Location:
    NJ by the sea
    Thx for the replies...my assumption, although it's a layman assumption, would be the most accurate method would be by actually measuring plate current since given all associated components are the same any given tube will measure a different current...right? I measured plate current by probing between pin 7 el84 and pin 8 5y3. This method results were 89% ...is this the correct way to measure actual plate current?
     
  5. peteb

    peteb Friend of Leo's

    Posts:
    4,200
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    That sounds like the shunt method.

    You measured current between the plate and the rectifier tube? Bypassing the output transformer?


    This method is known to provide quick and accurate results. I have tried it and it does, but it may read a little low, but not low like you got.


    What you really want is plate current to get the plate dissipation, not the cathode current, like SNfoilhat said.




    I think it is nice to check cathode current and plate current to cross check.




    A good way to find pate current is measure the DC resistance of the OT primary when the amp is turned off then measure the DC voltage drop across the same when the amp is idling, divide per ohms law and find the current.
     
  6. bluerolo

    bluerolo TDPRI Member

    Posts:
    34
    Joined:
    Jan 19, 2016
    Location:
    NJ by the sea
    Isn't measuring from the plate (pin 7) to rectifier providing me with plate current...tube #1 measured 33ma tube #2 measured 33.5ma
     
  7. King Fan

    King Fan Friend of Leo's

    Posts:
    4,521
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    I don't know that amp, and I don't know the el84, but is there a reason you can't use Rob's cathode resistor voltage drop method? One measurement, or two if you want to get all fancy, and you're done.
     
    Last edited: Dec 10, 2019
  8. peteb

    peteb Friend of Leo's

    Posts:
    4,200
    Joined:
    Apr 25, 2003
    Location:
    Cascadia

    If the plate current isn’t 5-10% lower than the cathode current then something is wrong or the measurements are wrong.


    You could also check the screen current. Cathode current equals plate current plus screen current.



    Or, just do the transformer resistance method. It is the most trusted.
     
  9. gusfinley

    gusfinley Tele-Holic

    Posts:
    958
    Joined:
    Jun 11, 2014
    Location:
    SLC, UT
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.


  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.