AB763 Static Dissipation

[samato]

I recently finished the build of my new amp which is a single channel AB763 Deluxe with (2) 6V6's & no effects.  For several days I've been running the bias at 35.5mVDC on V3 and 32.8mVDC on V4.  I was confused and thought this was okay for these tubes but now I know that this is way too hot.  With 445 VDC on pin 3 of each tube I was running this amp at around 30 watts and it should be more like 22 watts.  I know this now but I didn't do the static dissipation calculation until today.

It sounded really good so I left it that way for a while.  The only thing I really noticed was that the speaker I'm currently running it through couldn't handle this amp at max volume but since I don't know the specs of the speaker I figured it just wasn't rated high enough.  There was also some occasional popping sounds coming through.

Now I've re-biased based on what I read on the Lord Valve Bias FAQ page: http://www.duncanamps.com/technical/lvbias.html.  I'm running 454 VDC on pin 3 of each 6V6 with 25.7mVDC on V3 and 23mVDC on V4 which gives me about 11.7 watts and 10.4 watts for a total of around 22 watts.  The amp sounds even better now and the speaker can "pretty much" handle the power - it gets a little hairy at max volume.  I might re-bias a little colder but I think it's fine for now.

What I find odd is that the amp didn't really freak out when it was being run at 30 watts.  I guess these tubes wouldn't have lasted very long like that though.

No real question here, just thought it might draw some comments for discussion and learning purposes.

[bobmegantz]

Don't confuse watts dissipated quiescently in the tubes with output power watts.  Try measuring output power (using a dummy load) at both bias settings - you may find out that the power output is lower with the higher bias current.  Higher bias causes the tubes to operate in class A (that is, both tubes conducting current) more of the time, which sounds different.

If you bias too hot the tubes will dissipate more than their maximum rated value - they'll burn out faster.  Running 6V6's off of 450V at the plate is also above their maximum voltage rating (although Fender does it anyway...).  Anyway, the combination of high plate voltage and high bias current is risky.

[samato]

Well I can't say I completely understand all that stuff but I'm pretty sure the change I made today was in the right direction because the amp now sounds and feels more like a real Deluxe Reverb, which is what I want.

[PRR]

> 35.5mVDC ... With 445 VDC

Deduct 5% for cathode current to screen (does not count as plate dissipation).

> What I find odd is that the amp didn't really freak out when it was being run at

15 Watts in a tube which is rated 12W if you don't test, 14W if you do test. 7% over-diss is not instantly fatal, just shortens tube life. Modern "6V6" are nothing like the 1938 original (partly because Leo used many 6V6 after all other 6V6 markets died), will take even higher power without complaint, and give reasonable life.

However in this circuit there's no good reason to run them that hot, and as you say it may sound better idling easier.