In article XNWLi.896$ht5.398@trnddc02,
"Earl Kiosterud" wrote:

"SFTV_troy" wrote in message
oups.com...

Earl Kiosterud wrote:

Synchronous AM demodulation uses a locally regenerated carrier,
fed along with the AM signal (upper or lower set of sidebands) to
a multiplier (modulator). The result is the audio. It replaces
the envelope (diode) detector usually used. You can think of it as
another superhet stage where the result, instead of another IF
frequency, is the baseband audio. That's because the local
oscillator is the same frequency as the carrier of the (IF)
signal, so the difference is zero. The sidebands wind up
translated to baseband audio instead of to another IF frequency.

There are advantages. Since one set of sidebands or the other can
be used, if there's a distant station 10KHz away, causing that AM
whistle, you just switch to the other set of sidebands, whichever
comes in the cleanest. Also, it doesn't depend on proper
amplitude and phase of both sets of sidebands to work properly, as
does the regular envelope detector, so it works better with
impaired signals.



I only understood about 75% of what your wrote, but if I understand
your meaning, this new receiving technique would not improve the
sound (it would still be limited from 100-6000 hertz), but would
only reduce interference.


Troy,

Well, the 6 KHz limit is due to the narrow bandwidth of the
receivers, not the detector used, or the stations. I think most AM
radios actually do much worse than that. AM radios are designed with
a limited bandpass because it gets noisy as the bandwidth goes up.
The AM band is a soup of distant stations, particularly at night, and
that's the source of much of the noise. AM radio stations in the US
are allowed up to 10 KHz audio. That's pretty listenable -- there's
only a little over a half octave to the 15 KHz limit of FM.

The synchronous detector, in addition to being able to use one set of
sidebands or the other, whichever is the best under the conditions,
is not subject to distortion from asymmetrical sidebands, such as
when there is fading, multipath, etc. There may be a non-flat audio
bandpass from those conditions, but a conventional detector will also
have distortion.

I just made a few empirical measurements on a receiver with digitally
adjustable filters and noted increased high end audio response out to
8K. 8 khz wide is not pleasing because most radio stations are
apparently boosting the high end. I usually set the bandwidth 4.4 khz
for best sound otherwise it is to sharp.

--
Telamon
Ventura, California