> Interesting observation. Last week I posed this question
> on this reflector: "If one were to build a class "A" RF
> tube amplifier capable of producing an AM signal of 375
> watts carrier and 1500 watts PEP, what would be the plate
> dissipation required by the tube? (I know how stupid and
> inefficient this is, but I really would like an answer)"
> Of the 18 responders only two guys answered the question:
> The first guy said 700-800 watts and the second guy said
> 3000 watts. I'm not going to submit the same question to
> the Yahoo Amp Reflector, but based on the 18 responses,
> I'm not sure about submitting future questions to this one
> either.
To stay class A the amplifier always has to run at least
twice the peak envelope power output as static dissipation.
At least twice.
So for class A, the answer is indeed 3000 watts (or more)
standing dissipation in the PA. The useful RF output would
actually subtract from that, but you need some headroom so
the 3000 is a reasonable safe estimate under carrier
conditions. Wasted power actually decreases with audio.
If it is just linear and you can tolerate class AB somewhere
in the range near class B, you can do considerably better.
It also will not sound any different at all, and likely will
not have much more distortion (certainly none that could be
heard). In this case carrier efficiency is less than 35%
with 25% being a conservative estimate. That means 3/4 of
the carrier condition plate input power is heat. So carrier
dissipation is 3 times 375 watts or 1125 watts as a more or
less conservative estimate. The actual best you could ever
expect would be 750 watts, but that would be rare and also
not allow for any positive peaks exceeding 1500 watts.
During modulation peaks efficiency is 50% or more, so
dissipation would be 1500 watts or less. Since the duty
cycle is low on peaks a tube rated at 1200 watts dissipation
CCS would probably be a safe lower limit.
73 Tom
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|