Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups
Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups
Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups

bias question for single ended amp

Discussion in 'Shock Brother's DIY Amps' started by calvin006, Mar 31, 2011.

  1. calvin006

    calvin006 TDPRI Member

    May 28, 2009
    Durham, NC
    I have a singled ended Rickenbacker M8E that I use as a practice amp, both for playing and for learning about amps. Lately I've been reading about power tube biasing, and I decided to take some measurements. Here's what I found in my Ric:

    Plate voltage to the power tube = 340V
    current drawn by the tube = 34mA

    First, I'm wondering if I took my measurements correctly, as 340V and 34mA are suspiciously similar numbers. I measured plate voltage by probing between pin 3 and ground. I measured current draw of the tube by probing between pin 3 of the power tube and the pin 2 on the 5Y3, which is where the red wire of my OT runs to.

    Assuming the measurements are correct, maximum plate dissipation for a 6V6 is 14 watts, so the tube is running pretty hot at 82% of the max. I believe this amp is cathode biased, though I'm not sure. I've attached a fairly accurate schematic for someone more knowledgeable to inspect.

    My questions: is the tube running too hot? The amps sounds fine, I'm just concerned about burning through power tubes. In order to lower the current draw, how do I determine the proper resister value to swap in? I'm assuming I would be changing that 270K resistor on pin 8?

    Attached Files:

  2. laird

    laird Tele-Holic

    Jul 8, 2009
    Palm Harbor, FL
    270K can't be the right value for the cathode resistor. 270 ohms @ 5 watts would be more like it.

    To determine the total current draw (plate plus screen), measure the DC resistance of the cathode resistor with the amp powered off. Then power up the amp and measure the voltage between the top of the resistor (pin 8) and ground. Let's say you read 270 ohms for the resistor and 12.1V at pin 8... divide the voltage by the cathode resistance to determine the current (in milliamps) passing through the tube:

    12.1v / 270 ohms = 45ma

    Next, take the plate voltage (340v) and subtract the cathode voltage to calculate the voltage passing through the tube... 327.9v in this example, let's just call it 328v. Multiply that voltage by the cathode current to determine the total wattage being dissipated across the plate AND the screen:

    328v * .045a = 14.76w

    With single-ended amps, the output tube runs the hardest when there is no signal passing through it. It's normal to see the idle dissipation very close to 100% of what the tube spec allows. The 6V6GT data sheet shows the plate max at 14w and the screen max at 2.2w. The 14.76w measured is a useable operating point, but having a little bit more buffer room would be nice. In this example, changing to a 330 ohm cathode resistor may cut that overall dissipation down just under 14w while raising the cathode voltage closer to 13v.


  3. calvin006

    calvin006 TDPRI Member

    May 28, 2009
    Durham, NC
    Wow, super helpful. You're right, the cathode resistor is 270 ohms. For some reason the schematic reads 270K, so I copied from that.

    I'll take those other readings you mentioned and see what I find. Thanks!

  4. Wally

    Wally Telefied Ad Free Member

    Mar 17, 2003
    Lubbock, TX
    FWIW, I have never measured a FEnder Champ that did not idle at considerably more than 100% plate dissipation. I did an experiment with one just because it sounded so good. When measured, it was WAY past 100% of plate dissipationat idle....alarmingly high, I thought; so I started subbing in larger resistances for the bias. By the time I got into the high 14 watts of plates dissipation, the amp sounded like puke. That was with a 820 ohm resistor. I took it back to the stock 470 ohms......and more watts of dissipation than I want to share with you. THe amp sounded great. IT had worked that way for 40 something years, and it is still working today. IT is the best sounding BF Champ that I have ever heard. I could have played it with maybe somewhere around 600ohms of resistance....but that was still dissipating much more than 100%....and the amp was getting 'clean'. The 820 ohms yielded a sound that was harsh and sterile...unusable, imho, at 14.8 watts of plate dissipation.
    While I was doing this experiment, I asked here at what point would the 6V6 redplate in a cathode biased circuit. I never got a definitive answer......all I know is that the point at which that tube was going to redplate was so far above that 14 watt spec that the spec was almost meaningless in that application.
    HEre's a clue as to how 'hot' that situation was. I compared it to a friend's BF Champ that sounded good....but not as good as this one sounded. His amp was dissipating about 18 watts with the stock 470 ohm resistance in the circuit. I suppose this is why some folks search with their ears for the 'right' tube in these cathode biased amps, right?

  5. calvin006

    calvin006 TDPRI Member

    May 28, 2009
    Durham, NC
    Good to hear that perspective, Wally. The Ric M8E is very similar to a champ and has some great small amp tone. If I have to burn through 6V6's for great sound, so be it.

    But I will measure the total current draw tonight as per laird and report back.

  6. tubeswell

    tubeswell Tele-Afflicted

    Jul 1, 2008
    If its a cathode biased 6V6, then set it to idle at full dissipation. It will sound good (and it won't do any harm) and its quite normal for cathode-biased SE.

    FWIW a 6V6GT is a 12W tetrode, whereas the 6V6GTA is sometimes claimed as a 14W pentode (e.g.: see GE's datasheet). However the JJ6V6S is a pentode and is quite different from a 6V6GT/A, and is supposedly at least 14W but I reckon it could do 16W-17W. The JJs plate structure looks very similar to a Russian (6L6 equivalent) 6P3S.

  7. JohnnyCrash

    JohnnyCrash Doctor of Teleocity

    Mar 12, 2005
    Fullerton, CA

    Interesting stuff.

    As far as biasing a SE, tubeswell explained it well... this is NOT a Class AB/Push-Pull, so the general "70% bias" suggestion does not apply as Wally demonstrated. Class A SE amps biased for 100% at idle is fine because as tubeswell said in his first post, they work hardest when at idle... it seems counterintuitive and "wrong."

    As far as the math working out to higher wattage than the advertised 5-8 watts, I'm still stumped. It could be the idea that old amp designers declared output as clean wattages (power before the onset of distortion), this is something I've heard and read, but do not know if it is true or not. It could also be that Class A/SE idled at 100% do not output 100% when doing their job, IDK.

    I'm sure an actual EE will step in to explain sooner or later :)

  8. printer2

    printer2 Poster Extraordinaire

    May 24, 2010
    No hard and fast rules on distortion % at output. Some rated at 5%, some at 10%. Mind you on a higher negative feedback am distortion remains low until clipping whereas tube amps generally distorted somewhat befor clipping.

    Also the 70% dissipation is for fixed bias amps not cathode biased (push-pull) which are sometimes run at 100%. On SE amps the plate at zero signal may be passing 100% of the current of full signal but the screens will not be at full dissipation. Their current will go up at full signal. You can think of the tube in Class A as absorbing full power at zero signal and as you increase the signal it diverts some of that power (about 25%) to the speaker (actually the output transformer diverts it to the speaker, the tube just goes on doing its thing).

  9. calvin006

    calvin006 TDPRI Member

    May 28, 2009
    Durham, NC
    Alright, here's what I measured:

    Cathode resistor = 340 ohms (what's with all my readings being 34 or 340?)
    Voltage b/t resistor and ground = 14v
    So 14v / 340ohms = 41ma passing through the tube
    328v * .041a = 13.366w across plate and screen

    That sounds about right from you fellas have said. Think I'll leave it alone. Though I'm still wondering why 34 and 340 keep coming up in my readings... pure coincidence or am I doing something wrong?

  10. tubeswell

    tubeswell Tele-Afflicted

    Jul 1, 2008
    The 41mA is tube current (i.e.: the sum of the plate current and the screen current), so if you subtract ~2mA for a typical idle screen current in a 6V6, that leaves 39mA (an assumed) plate current.

    If you've got 340VDC on the plate (as you said in the original post), and 14V on the cathode, then the plate-to-cathode voltage is (340-14 =) 326VDC.

    326V x 0.039A = 12.7W plate dissipation

IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.