I thought about that, but I wasn't sure how the skirt selectivity would be.
Right now, even with a Murata CFWS-455H at the 2nd I.F. skirt selectivity is
very good. I think that this is because of the cascaded effects of the
diplexer
(15kHz BW), and the high I.F. filters.

Pete
Frank Dresser wrote in message
...
Have you considered using a single, less expensive IF filter and a Q
multiplier or Q multipliers to change it's shape and bandwidth?

The notch/peak of a Q-multiplier is a very sharp exaggeration of single
tuned circuit. Could a double tuned band pass circuit also be used?
Tuneable band pass notches at both the upper and lower sidebands with a
peak
in the middle would be very useful and would reduce the need for several
fixed filters.

I use a Heathkit Q-multiplier. When it's tuned to the center of the IF
frequency, advancing the regeneration control will smoothly sharpen the
peak
and progressively cut the side bands until it breaks into oscillation.
It's
like an IF frequency tone control! Pretty cool. Another good trick is
tuning in the sidband farthest from the interference. Then I bring the
carrier back up with the Q-multiplier. Or I can use the tunable notch to
get rid of hets.

Frank Dresser