On Sun, 09 Oct 2005 10:50:22 -0400, Dave Holford
said in rec.radio.scanner:

Al Klein wrote:

But I can assume that all filters currently in production are
symmetrical about the center frequency within 10% - because they all
are.

Certainly most current crystal filters

Mechanical filters, ceramic filters, lumped-constant filters, op-amp
filters ... any discrete piece of hardware that's supposed to pass a
band of frequencies in the IF of a modern (within the last 40 years or
so) receiver.

are symmetrical about the centre frequency,
as are most people symmetrical about the centre (externally at least, and I'm not
going to quibble about the orientation since I'm sure you know what I mean) - I
wrote 'carrier' frequency.

If you put the carrier (no need for tics - carrier whether suppressed
or not) at the center of the filter's bandpass, the filter is
symmetrical about the carrier - if you put the carrier outside the
bandpass of the filter are you going to claim that bandpass filters
don't pass signals? The filter is symmetrical about any signal
centered in the bandpass if the filter has a symmetrical bandpass - by
definition.

I'm not sure about current mechanicals (somewhere I have an old Collins filter
catalog with asymmetrical SSB curves).

Within 10%? None of mine were ever that distorted, although some of
the Collins filters were about the most asymmetrical filters I've ever
seen. They were designed that way. But the receivers were designed
to work with "asymmetrical" filters.

Just for fun I am currently listening to an LSB conversation on 3763 kHz with
a 7.5 kHz filter and I cannot hear any sign of the conversation on 3765 - but,
the noise is a bit higher due to the wider filter. I wouldn't normally do this
but just had to try it. Of course, if I select DSB or ISB I get both signals,
although the 3765 conversation is inverted and shifted in frequency.

With what receiver and which filter? What kind of detector? I can
design any number of receivers that will do what you claim, but that
doesn't negate the fact that the standard filtered superhet SSB
receiver will receive a signal on the "wrong" sideband if the filter
is wide enough and the detector is set correctly.

As I stated in my first posting, a Winradio.

Oh, you mean that a DSP is capable of not hearing the opposite
sideband. Nothing to do with filters. "Filtering" isn't always done
by "filters".

The rest of the paragraph is true if you are talking about 'the standard filtered
superhet SSB receiver' provided that it is also fitted with AM filters which can be
used in the SSB mode - not universally true. I can do it with my TS-450 but not my
IC-706 for example. Not all receivers have the ability to select mode and filter
independently.

And my Honda can't receive ISB either. If we're comparing fruits,
chickens don't count.