Power Transformer sizing..

[dehughes]

Myself, I look at maximum plate current rating given a particular voltage (spec sheets) and see if that's in the ballpark of what I'm designing. As well, I compare this to known real-world ratings by examining existing amp designs and seeing what they've done in their power supplies. Between the two I can get a "spec-sheet" and "real-world" rating, and then just combine the two for what I'm needing.

I've yet to design a really "out there" amp, so I can get away with this. If I were designing something that had yet to be done, I'd have to rely on my math only (God help me...).

[Old Tele man]

...with Class-AB operation (its actaully for Class-B, but is close enough), the "mathematics" of plate voltage and current hinge on the ratio (half of PI/2):

PI/4 = 0.7853981... ~ 0.7854

...with Class-AB (and -B) operation each tube is operating "one-half of one-half" of the time, with one tube handling UPPER portion of signal and the other handling the LOWER portion of signal, and together, EACH tube working only half the time (alternatingly)...thusly:

½ of ½ = ¼

...so, if the average plate power dissipation (Ppd) rating of a 6V6GTA is 14W, then logically you would assume that each tube could be "driven" four times harder and still not exceed its 14W average rating:

(4X)*(14W)*(¼) = (4/4)*(14W) = 14W!

...this is 100% true, but because of how PP action 'works' the earlier PI/4 factor must be included:

Po = (PI/4)*(4*14W) = (PI)*(4/4)*(14W) = PI*14W ~ 3 x 14W

...thus, the maximum power you can expect from two tubes operating Class-AB (or -B) is PI-times their average rating (Ppd), or roughly 3-times-Ppd.

...hence, the "Rule-of-Thumb" is: Po(PP) = 3 x Ppd

where:
PI = constant: 3.1451592654...
Po(PP) = Output power for Push-Pull configuration, in Watts.
Ppd = Plate power dissipation for a tube, in Watts.

[dehughes]

Math! Aaaaahhh...(runs screaming from the room, leaving his cathode biased EL84 quad biased at 16w each and nearing meltdown...)

[mike_mccue]

Thanks to both. :-)

I have to admit I did not keep up with the math. Some of the coefficient naming confuses me.

best,
mike