Difference % in Bias Method

Discussion in 'Amp Tech Center' started by ocduff, Mar 29, 2020.

  1. ocduff

    ocduff Tele-Meister

    Posts:
    175
    Joined:
    Nov 17, 2003
    Location:
    USA
    Has calculated the percent difference between cathode resistor method and transformer shunt method?

    I’ve heard that cathode resistor method always measures on the low side - question is, how low.

    I can’t do transformer shunt method due to not having a multimeter with 3 decimal places.

    Appreciate the input. I’ve got a ‘61 Super measuring 450v and 40ma on one 5881 and 46ma (80%) on the other with cathode resistor method.

    Definitely running the ragged edge but it just sounds so damn good clean.
     
  2. Paul G.

    Paul G. Friend of Leo's

    Posts:
    3,016
    Joined:
    Mar 17, 2003
    Location:
    Rhode Island
    The cathode resistor will read low by about 5-6% because it does not account for the grid current. Compensating for that, your hot tube is running at 75% of max dissipation. You'll have a slightly lower tube life, but I don't expect you to have any other issues. I'm assuming you're talking about modern 26-Watt 5881s. If you have 19-Watt tubes, that's another story.
     
    ocduff likes this.
  3. ocduff

    ocduff Tele-Meister

    Posts:
    175
    Joined:
    Nov 17, 2003
    Location:
    USA
    Thanks Paul - that’s the info I was looking for. Much appreciated.
     
  4. radiocaster

    radiocaster Poster Extraordinaire

    Posts:
    6,824
    Joined:
    Aug 18, 2015
    Location:
    europe
    You don't need three decimal places. Instead of using the chassis as ground reference, put the negative lead on the plate and the positive on the B+ to measure the voltage drop.
     
    Intubator and ocduff like this.
  5. Dacious

    Dacious Poster Extraordinaire

    Posts:
    8,201
    Joined:
    Mar 16, 2003
    Location:
    Godzone
    Did you try swapping tubes and rechecking voltage and current? The two halves of the transformer are not perfectly equal and you may find your 'hot' tube and 'cold' tube may match closer swapped.

    OTOH, if it sounds good why screw with it? 70% isn't red hot. As Paul noted life will be shortened slightly, maybe. Sov 5881 should be OK. You may find they drift closer with time.
     
    ocduff and Intubator like this.
  6. ocduff

    ocduff Tele-Meister

    Posts:
    175
    Joined:
    Nov 17, 2003
    Location:
    USA
    I was thinking of resistance method. I can do shunt as you said.

    Tubes are old stock Tung Sol 5881/6L6WGB.

    I have matched pairs but this particular slightly mismatched pair sounds awesome. I may lower the bias a notch. One of the tubes is original to amp - the reason it tests like new is the other original tube was so mismatched that it did all the work for 50 years.
     
  7. radiocaster

    radiocaster Poster Extraordinaire

    Posts:
    6,824
    Joined:
    Aug 18, 2015
    Location:
    europe
    That's what I said. Probably something like 80 ohms is close enough for measuring the transformer, in exactly the same place I said to measure the voltage. 80 ohms is close enough, for example, you don't need to have 79.347 ohms or something like that.
     
    SixgunElectric likes this.
  8. Wally

    Wally Telefied Ad Free Member

    Posts:
    35,277
    Joined:
    Mar 17, 2003
    Location:
    Lubbock, TX
    Rob Robinette says his comparison shows the two methods are within 1% of each other..iirc.
     
    King Fan and ocduff like this.
  9. ocduff

    ocduff Tele-Meister

    Posts:
    175
    Joined:
    Nov 17, 2003
    Location:
    USA
    I think the issue is measuring voltage drop, not resistance. If the meter doesn’t break voltage down into decimal places you end up with an unusably rough number.

    For shunt method - it’s my personal choice not to use it as a hobbyist who may only bias an amp twice a year. And, I slip too much.
     
    Paul G. likes this.
  10. ocduff

    ocduff Tele-Meister

    Posts:
    175
    Joined:
    Nov 17, 2003
    Location:
    USA
    Thanks for all who replied. Somewhere between 1-6% is my answer.

    And, if a 5881/6L6WGB tube is a 23-26 watt tube, respectively, based upon one’s source, I’m good to go.

    This amp sounds entirely FLAT biased at 50 percent. Lifeless. Cold. Sterile.

    I’m okay using up some old tubes a bit faster if it makes the amp.

    Even among my stash of 5881’s, they all sound different. Some tubes sound different than others.
     
  11. King Fan

    King Fan Friend of Leo's

    Posts:
    4,994
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Heh, I knew I was missing something. The cathode resistor method is for cathode-biased amps, right?

    But the rest of you were way ahead of me. We're talking about the *OT resistance method*, right. Slow today, my brain.

    So, yes. I agree the OT resistance method isn't very precise if your MM doesn't get down to decent decimals. So back when, I got one of those inexpensive Eurotubes bias probes that measure plate current; I can measure plate voltage myself. More recently, I just installed 1% 1ohm resistors between cathode and ground and measure current (as voltage) directly. So easy.
     
    ocduff likes this.
  12. ocduff

    ocduff Tele-Meister

    Posts:
    175
    Joined:
    Nov 17, 2003
    Location:
    USA
    Cathode resistor method works for cathode biased or fixed bias amps. I’ve fiddled with the 3 variations and it works best for me. Everyone has a favorite.

    I believe you’re describing cathode bias method - with the 1 ohm resistors from cathode to ground. Agreed. So simple. You don’t even have to remove the chassis - just slide it back 2 inches and you have access, with speaker connected.

    It’s like in fishing - there’s books that will give you all these fancy and complicated knots that the famous author likes best and says you MUST use but there’s far simpler knots that are easier to remember and work just as well.
     
    Last edited: Mar 30, 2020
  13. Wally

    Wally Telefied Ad Free Member

    Posts:
    35,277
    Joined:
    Mar 17, 2003
    Location:
    Lubbock, TX
    I would never suggest that someone use the transformer shunt method, bit it is the quickest method of all....works on both fixed and. Cathode bias circuits. And....unless one uses a probe system that reads plate voltage, one is going to be working on a live circuit at any rate. And...the shunt method is as accurate as any other. It is the only method I have ever used. Ommv...
     
  14. radiocaster

    radiocaster Poster Extraordinaire

    Posts:
    6,824
    Joined:
    Aug 18, 2015
    Location:
    europe
    Why? You're only going to get a couple of volts, like 5V. If you have one decimal place, you'll get something like 5.2. I don't see how getting a reading like 5.241V is going to get a way better bias, as even the voltage drop itself will not remain that precise while the amp is on.
     
  15. ocduff

    ocduff Tele-Meister

    Posts:
    175
    Joined:
    Nov 17, 2003
    Location:
    USA
    I think I may have misspoken in my first post. Apologies for the confusion.

    I’m going off some foggy memories here. I made a comment about decimal places in regards to shunt method, but I believe I meant resistance method.

    It’s not the one tube showing 5 volts and the other 5v, it’s because if you measure the difference between the two and the DVM reads 1 volt on one tube and 2 volts on the other (which is what I usually get), and you divide that by the OT resistances (which are usually fairly close - 235 ohms and 245 ohms), you get numbers that are vague at best and often, 50 percent mismatched, even when you know the tubes are closely matched via other methods.

    Like I said, I’m going off memory here.

    Now, had it broken things down into 1.235v 1.687v rather that rounding up and down to 1 and 2, respectively, then that would be something helpful to make a calculation that needs to be very accurate.

    THIS - is why I do cathode resistor method lol. I don’t know if I’m getting older or I just don’t have room in the old memory bank to keep old stuff. That, or I’ve accumulated so much in there...
     
    Last edited: Mar 31, 2020
    radiocaster likes this.
  16. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    7,784
    Joined:
    Dec 29, 2012
    Location:
    United States
    Measuring using a cathode resistor will read HIGH by about 5% because it will include screen current. Either subtract 5% or use my online calculator which does it for you.
     
    ocduff and Wally like this.
  17. sjoko

    sjoko TDPRI Member

    Posts:
    54
    Joined:
    Mar 4, 2020
    Location:
    The Netherlands , Amsterdam
    I made this little , handy tool for adjusting the bias myself .....
    All you need is an octal tube socket and an octal plug ,... 2 crocodile clamps and two pieces of wire...

    After adjusting the bias , you better can check the output sinus wave-form with a sinus wave generator and an oscilloscope as well .
    In case of cross-over distortion ( "silent buzz" ) , you're able to do a little fine tuning ( just a wee bit more idle current can solve that problem )

    CAM02308 (Medium).jpg CAM02309 (Medium).jpg
     
    Last edited: Apr 1, 2020
    ocduff likes this.
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.