> I was just reading some info on class A amplifiers.
> It seems that a typical triode class A1 amplifier (no grid
> current) the
> efficiency is in the range of 20% to 35%.
I was using the theoretical maximum. The practical maxmium
could be any value less than that (and proably will be).
> So I guess it depends on how the class A amp is configured
> as to the maximum
> efficiency it would produce. I have always used the figure
> in the area of
> 33% as a ball park value but obviously that can be
> different.
I haven't designed that many class A amplifiers where
efficiency was an issue, so I can't really contribute any
*typical* measurements like I can with class AB. Efficiency
would have to be some value less than 50% on modulation
peaks.
> Using a class A amp as an AM linear I don't think the 50%
> efficiency change
> rule applies as it does with a class B linear amplifier.
Let's think about that. Input power is constant. Efficiency
would be the standing dissipation minus the average RF
output power.
With 500 watts carrier power in the sidebands at 100%
modulation would be 250 watts. As I recall the formula is
Paudio= m^2*Pcar/2
so with 100% modulation and 500w carrier: 1^2 * 500/2
=250 watts in the sidebands. That's 750 watts when 100%
modulated, and 500 watts with just carrier. So efficiency
change would be a factor of 1.5 when going from carrier to
100% sinewave modulation. So my 50% to 25% assumption was
wrong, as is the "doesn't change" assumption. In one example
with 25% eff on carrier, the actual eff value would be 25%
carrier to 37.5% with 100% sinewave modulation.
I think this makes more sense.
We have to consider the average or effective power of the
signal, not the peak power, when considering heating. A
100% sinewave modulated carrier requires an additional 50%
of the carrier power to have 100% modulation. So the class A
amplifier runs cooler with modulation because input power is
constant and output power increases.
73 Tom
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|