> But is the requirement ripple reduction, or the ability of
> the HV supply to hold up for short peaks?
> The capacitor releases 0.5 .C.Ve2 (one half C times V
> squared) joules where V is the drop in the HV. That's
> also equal to coulombs/sec, which is I times t.
> So you can equate these to figure how much capacitance is
> needed for a given AC supply impedance and frequency for
> an amount of ripple. But if you're talking of allowing a
> different voltage drop over a 50 millisecond speech peak,
> you get a different answer, and that's where a big
> capacitor holds up better. It also can show up a lot
> better on a 2 tone test as the tones get closer together
> in frequency, although I think we all agree that a 2 tone
> test is not the be-all and end-all of test methods. But in
> terms of ripple, the bigger the ESR, the sooner you get to
> the stage of diminishing returns.
If the supply is that loose, you better use a battery.
The problem isn't only a 50mS time constant for the lowest
speech tones, the problem is a cycle per second time
constant for the syllabic amplitude gaps and level
variations.
Increasing the electrolytic size can actually make an amp
less linear under some speech conditions if the supply is
loose from a very poor transformer. The cure is to not use a
high ESR transformer on SSB.
73 Tom
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|