About ripple in a power supply:
> Ripple measurements and calcs have been RMS since day 1.
=====That is a true statement.
> Scopes which read peak to peak have
> been available since the stone age,
=====That is also a true statement.
> meters which will read true RMS of a
> non-sine waveform have not, and I suspect are
> still rare in most ham shacks even today.
=====And that is a true statement.
> Just saying "1% ripple" with no qualifiers leads to
> confusion.
=====Absolutely wrong! In that writer's mind, I would
suppose, but not in general. Certainly not my
mind.
What is being overlooked here is that the ripple
in the usual power supply is nearly sinusoidal,
with the higher-order harmonics rolling off at
least 6 dB per octave. (The dominant harmonic
will be the second harmonic of the line frequency.)
With a choke-input filter the harmonics will roll off
at 12 dB per octave. With a two-section filter
(choke input, shunt capacitor, another choke and
finally a second capacitor) they will roll off at about
24 dB per octave. If you look at the ripple
component you will see that you will not need a
fancy "true RMS" meter to measure the ripple
because the harmonic content is so low. One
could look at that nearly-sinusoidal ripple
component with a scope, divide the peak-to-peak
value by three to get the "RMS" (note the quotes)
and be done with it and get on with things more
imnportant. We are not needing precision here.
I used the number "three" because I would be
doing that in my head and I am must as lazy as
the next guy.
One item that could mess up these observations
is the presence of significant equivalent series
resistance in the capacitors. Then the pulses
of current through the capactors will result in
spikes of voltage across the capacitor. The
amplitude of those spikes will be dependent on
the value of that ESR. They will normally be quilte
narrow as viewed on the scope. In practice the
presence of such spikes is an abnormal situation.
Bill you are a bright and knowledgeable fellow
but I think you should back off on this one. :-)
- JimT W4ENE
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|