Hello,
Yesterday, Walt K6WRU wrote "In analog filter design, you get to choose
between good phase performance and good amplitude performance. You can't
have both. If you want a flat passband and steep skirts, you'll get wonky
phase."
I suppose that is basically correct for the standard "off-the-shelf" filters
available to the ham community.
However, filters can be produced that correct the group-delay.
I have designed many such crystal-filters in my career spanning almost three
decades.
These filters were supplied to companies such as Harris for use in military
equipment.
A typical shape-factor 60dB/1dB would be less than 1.4.
The corrected group-delay would be quite flat over perhaps 80% of the 1dB
bandwidth.
(For many years, one of my filter designs was part of the display in the
lobby-area of Harris in Rochester NY; it may still be there for all I know.)
Additionally, about the "Bessel" design, I would look first to either a"
Gaussian to 6dB" or a "0.05 degree phase-ripple" approach. If I recall
correctly, either choice can give sharper skirts than the Bessel and still
have relatively flat group-delay. (I no longer have access to Zverev, my
copy of which by the way was full of errors, especially in the band-reject
section.)
K6WRU wrote "There is no free lunch, it's physics." So true.
Engineering is all about compromise; what do I have to give up to get what I
still want.
73 Jerry KM3K
______________________________________________________________
Elecraft mailing list
Home:
http://mailman.qth.net/mailman/listinfo/elecraftHelp:
http://mailman.qth.net/mmfaq.htmPost: mailto:
[hidden email]
This list hosted by:
http://www.qsl.netPlease help support this email list:
http://www.qsl.net/donate.htmlMessage delivered to
[hidden email]
Locked
