On Mon, 2012-01-02 at 16:31 -0500, Gary Schafer wrote:
>There is no easy way to do audio clipping for SSB. The best way is to
> do RF clipping followed by a band pass filter.
The PM component can be ignored if the modulation bandwidth is very
small compared to the RF frequency (Always the case in SSB), in which
case the envelope is simply sqrt (I^2+ Q^2) which is trivial in a DSP
based radio, and given we are only interested in the peaks, abs(audio)
will come plenty close enough.
> The problem is that the audio waveform is not the same as the RF
> envelope waveform. It may look similar but operates entirely
> different.
Yep, the RF has a PM component, however the rf ENVELOPE is still
very close to sqrt (I^2+Q^2) as long as the modulation bandwidth is very
small compared to the RF frequency (It always is in our applications).
> Put a square wave audio signal into an SSB transmitter any you only
> get very large spikes out.
You get spikes in the frequency domain of course, just the same as if
you look at the frequency domain plot of the original square wave, but
in the time domain you just get pulses of RF with some PM that you
usually cannot see easily on a scope.
Consider the fourier transform of a square wave and you will see what I
mean.
Compression followed by any number of clippers works fine, RF clipping
is less audibly objectionable because most of the IMD falls outside the
clippers filter, but a pair of diodes will work.
73 Dan.
On Mon, 2012-01-02 at 16:31 -0500, Gary Schafer wrote:
> There is no easy way to do audio clipping for SSB. The best way is to do RF
> clipping followed by a band pass filter.
> The problem is that the audio waveform is not the same as the RF envelope
> waveform. It may look similar but operates entirely different.
>
> Put a square wave audio signal into an SSB transmitter any you only get very
> large spikes out.
>
> 73
> Gary K4FMX
>
> >
> > A clipper made from couple of diodes back to back across the audio line,
> > followed by a lowpass filter will get you most of the way there,
> > particularly if preceded by a compressor (THAT Corp do a very nice chip
> > that can be trivially configured as a compressor/limiter/gate).
> >
> > The filter will of course introduce some overshoot and ringing, so a few
> > stages may be indicated, but that is still only what a dozen or so
> > components.
> >
> > Or a pair of SL670s or similar and a 445Khz ceramic filter plus a few
> > diodes makes a dandy RF speech clipper/filter combo.
> >
> > I might draw one up and publish it somewhere.
> >
> > Personally I consider any ALC voltage to be an indication that something
> > is not set up right, you know the amps power rating and power gain, so
> > you know the required peak drive power, program that into the exciter
> > and leave the ALC line unhooked (My exciter does NOT overshoot, some do,
> > be careful).
> >
> > A reverse power sensor after the amp is a good idea if you go this way
> > as the amp now has no way to fold back the exciter drive power.
> >
> > 73, Dan.
> >
> >
> >
> > _______________________________________________
> > Amps mailing list
> > Amps@contesting.com
> > http://lists.contesting.com/mailman/listinfo/amps
>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|