Triode in Class B or Beam Pentode in Class AB2

[Michael Koster]

6DQ5 looks like an excellent candidate for screen drive or g2/g1 scaled drive: decent gm (>10mA/V) and nice low mu (3.3) means you wouldn't need as much g2 drive voltage. About 130V P-P with scaled drive would achieve the 100W output given the conditions above.

 

[bear]

When setting bias, is there a rule of thumb on how far to stay away from plate dissipation? In other words, if plate dissipation is 30 watts, so I strive for 25 watts idling?

Look at the tube curve.

Half way - the center of the curve is the ideal point for maximum linearity...

Class A is 50% at the given operating condition.

Otherwise you can not drive it fully in direction or the other... so I am unclear why 70% would be good, unless you are willing to accept somewhat lower output and efficiency...

_-_-bear

PS. class AB is biased ON just a bit above Class B, which is almost OFF...

 

[rsumperl]

Greetings,
What are your thoughts on idling Ib so that the 24 watt plate dissipation is not exceeded.

 

[bear]

Sorry, I do not understand your question - or if it was aimed at me or everyone??

_-_-bear

 

[Tubelab_com]

So, if I wanted AB1 operation, with 600v on the plate and 150v on the screen, where would you set the no signal plate current? Plate dissipation is 24 watts. My gut feel is 22-27 ma/tube.

I have been working on a similar design with the intention of trying several different tubes including the 6DQ5, but this past week I was at work for 72 hours, so progress is zero. Not sure when I'll get back to it.

In all of these designs the trade off is tube life VS distortion and sound quality. As the plate voltage goes up the idle current must go down to avoid the red glow of death. There is an optimum point where more voltage doesn't make more power because you have to turn down the current.

In my purely non scientific method I use a variable power supply, adjustable bias, and several different load impedances, to find what works best for a given tube in a given circuit. I realize that the average experimenter doesn't have that flexibility.

My experiments with the red board reveal that 30 to 40 mA seems to afford the best sound quality on several different tubes, plate voltages and load impedance. In many cases you can't run that much due to plate voltage and dissipation limits. Sweep tubes are highly inconsistent from tube to tube, so a bias adjustment is mandantory. In this case I would start testing at 25 mA or so and tweak up from there. Find the glow point and back up to as least 75% of that dissipation value. My guess is around 30 mA, and this matches what someone else reported.