I think what you stated below is what we in the Car Audio world Call "The
point of diminishing returns"
wrote in message
...
On Tue, 09 Dec 2003 16:07:37 -0600, 'Doc wrote:

The point being, if the power supply is not capable of
supplying the current required on SSB voice peaks, then the
power supply is too small to start with. The logical 'cure'
for a 'too small' power supply is a larger power supply. Any
thing else is a 'kloodge', a crutch to prop up a cripple.
A capacitor does supply some additional current to the
system if it's large enough, but even with very large capacitors
the supplied current is going to be very, very tiny in relation
to the total current draw. Something else you should remember
is that the power supply is also going to have to charge or
re-charge that capacitor when it is drained. That means that
there is no increase in total current in the system, in fact,
there is a decrease since capacitors are not 100% efficient
(they do have loss). This means that the current demand on the
power supply has increased and the 'hole' is just getting
deeper.
Capacitors 'work' in audio systems because they 'rob' current
from one part of the audio signal and 'deposit' it in another
part of the audio signal. If the purpose is to accentuate the
high frequencies, then the 'extra' power is robbed from the
lower frequencies (or visa-versa). The total average power of
the audio signal is not changed (increased), it's only re-
ditributed.
Some things with audio system do carry over in to RF power
amplification systems, but the use of capacitors in the way you
want to use them, isn't one of them...
'Doc

Another point is you can't equate SSB voice to music program
audio. There is the difference in compression. A direct ratio of the
capacitance needed applies to the peak to average ratio of the
power output. If your peak to average ratio in amplifier "A" is one
half of amplifier "B" then the capacitance needed for amplifier "A" is
twice as much as amplifier"B".

Music programming by default is suppose to be a true representation
of the audio. It requires a large dynamic range. The way this is done
is to not use much compression. Music programming has a very large
peak to average power ratio.


SSB voice is just the opposite. The preferred way to communicate with
SSB is to create a high order of intelligibility. This is done with
some sort of compression. Compression is desirable with a SSB voice
signal. A minimum of 6db of compression over music programming is
used.
The compression is normally quite a bit higher than 6db.

This all means that a SSB voice signal requires (6db) 4 times the
capacitance that a HI-FI audio signal requires. It is common practice
with audio amps to suggest a minimum of one farad for ever 1000 watts
of HI-FI audio, therefore a SSB voice signal would require four farads
for 1000 watts or in our case 6 farads for a 1500 watt dx1600.

1 farad cost $50, so we need $300 worth of capacitors in order to see
a difference. $300 dollars can better be spent on another battery or
two, or better yet a higher capacity alternator.

Bottom line on capacitors for SSB............Not worth the money.