Cathode Biasing Questions

[guitarboy58]

Having just built a Hoffman Bias Tester this weekend I decided to try it on my cathode biased 6V6 push-pull amp.  Now I have questions and concerns....  The tubes are matched by the manufacturer and share a common resistor and by-pass cap.  I see very different currents between the two tubes.  With the volume at zero, one measures around 32mA (which is in the ballpark I expected) while the other is around 60mA (which I didn't expect and don't understand).  The amp sounds great and does not appear to be overheating.  Nothing has changed with the amp for 2 years or more.  So my questions are:  Is this normal?  If not, do I balance it by going to different bias resistors for each tube or introduce a balancing circuit? Am I losing a tube?  Or did I just do something wrong?  I've used other bias testers on fixed bias amps and had no issues, and, yes, I did check the wiring of the tester before I used it, so I'm hoping there's a logical answer that doesnt involve my sanity.   Thanks.

[AZJimC]

If you swap the tube positions, does the excessive current draw follow the tube, or stay with that hole?

[sluckey]

32mA and 60mA is not matched at all. If you switch positions with the tubes, does the 60mA reading always stay with the same tube?

Tell us more about your bias tester. Does it have dual sockets so you can check both tubes at the same time? Are you measuring actual milliamps or do you have a 1Ω resistor across the banana plug for your meter so you can measure millivolts? What voltmeter? Describe your procedure.

Give us some more info about your amp without the bias tester plugged in. What is the plate voltage? What is the cathode voltage? What is the value of your cathode resistor?

[guitarboy58]

Thanks for the quick replies.  1. The current follows the tube.  So I agree they are not matched....60mA verses 35mA.  2.  Unless you guys see a critical need to pull the amp apart and check the exact voltages I'll just post what I have in my notes from the original build.  The transformers: Power is 360AC center tapped.  The B+ voltage at the output transformer CT is 421V.  419V on the other side of the choke feeding 470 ohm screen resistors.  6V6 GT tubes cathode biased with a 250 ohm resistor and a 100uF capacitor. The tubes were thought to be matched, but clearly are not.  The power amp section is loosely based on a Fender Deluxe. 
The Bias Probe is the Hoffman kit.  I'm feeding it into a Fluke 8060A multimeter (fully calibrated but the protective fuse in the current circuit is slightly undersized, read that as 1A, not the full 2A fuse it will handle).  As I said, the amp is giving me no cause for alarm and has been perfect for the past 2 years.  I was just testing out the probe and now this has moved to academic questions to further my education.  I've never tested a cathode biased amp for bias current.  The results were surprising and I'm looking for a better understanding of what's up. 
Assuming there's no serious damage issues here, is it possible the imbalance may actually be improving the tone (perhaps increasing 2nd order harmonics)?????

[AZJimC]

Thanks for the quick replies.  1. The current follows the tube.  So I agree they are not matched....60mA verses 35mA. 

Agreed.

The Bias Probe is the Hoffman kit.  I'm feeding it into a Fluke 8060A multimeter (fully calibrated but the protective fuse in the current circuit is slightly undersized, read that as 1A, not the full 2A fuse it will handle).

Am I reading that you're using the amperage inputs on your meter? The whole idea behind the bias tester is that the 1ohm resistors position themselves between the cathodes and ground, or in this case, the cathode resistor, and allows reading the milliamps as millivolts. You want to read the millivolt  drop on that resistor. In your case, reading the voltage on cathodes directly, and calculating the voltage dropped across the 250ohm is the most direct and reliable way to read the current drawn by the tubes, collectively. At this point, I'm not sure your readings aare reliable.

Assuming there's no serious damage issues here, is it possible the imbalance may actually be improving the tone (perhaps increasing 2nd order harmonics)?????

Anything is possible, however if your 6v6 is actually drawing 60ma at idle, it will be short lived to say the least. There are much better (read safer, more reliable)  ways to get that imbalance if you feel it suits your taste.


Another note, I recently found a pair of 6L6's doing pretty much the same thing, and the run away tube was doing near nothing in contribution to the output. upon replacing the bad tube, my headroom improved and the amp was restored to it's original volume.

[guitarboy58]

If I understand the function of the Hoffman Bias probe: the current flowing through pin 8, the cathode, is measured directly by breaking the signal path and routing it through the current measuring inputs of the multimeter.  There's no voltage/resistance calculation required on my part....the meter does it for me, so it's effectively a direct measurement of bias current.  You don't need the 1ohm resistors because the appropriate shunt is inside the meter.
Everything about the tester seems to be working.  My future plans are to place a second socket assembly and a switch on the unit so I don't have to pull hot tubes to get the readings for a pair.
Having said that, I think I should grab a new set of tubes to validate things.

[AZJimC]

If I understand the function of the Hoffman Bias probe: the current flowing through pin 8, the cathode, is measured directly by breaking the signal path and routing it through the current measuring inputs of the multimeter.  There's no voltage/resistance calculation required on my part....the meter does it for me, so it's effectively a direct measurement of bias current.  You don't need the 1ohm resistors because the appropriate shunt is inside the meter.
Everything about the tester seems to be working.  My future plans are to place a second socket assembly and a switch on the unit so I don't have to pull hot tubes to get the readings for a pair.
Having said that, I think I should grab a new set of tubes to validate things.

I stand corrected, I failed to look at the diagrams and instructions for Doug's design. I am accustomed to the type with the 1ohm resistor built in, however you're absolutely right, and I see your readings are likely correct. A replacement pair is a good step, and will likely correct the runaway condition.

[sluckey]

Using a bias tester is a great way to tell how much current is flowing thru a tube. But that's only part of the picture. The whole picture is about how much power the tube plate is dissipating. For that, you must measure the plate voltage also. And in the case of a cathode biased amp you also need to know the cathode voltage. That's the only way to get a good sense of how the tube is operating, whether it's operating safely or about to burn up.

Just doing a little math on the info you provided, you have one tube running 35mA and the other running 60mA. That's a total of 95ma flowing thru that 250Ω resistor, which means there is 23.75V on the cathode. I'm gonna use your 421V (at OT CT) as the plate voltage, so the real plate voltage is 421-23.75=397.25v. Multiply 397.25 x .035=13.9W, and that's fine for one tube. But 397.25x.06=23.8W and that's way too much for the other tube. You may not be able to see red plates in that tube (maybe if you turn off the room lights) but that tube is cooking.

[guitarboy58]

Thanks guys.  Had I not wanted to verify the tester I would have never known (until I lost a tube).  I'll pick up a new pair and see how things look.  I'll be back with more questions if it's still unbalanced.