Re High Q filters


From: Mike Czuhajewski (
Date: Fri Oct 07 1994 - 04:38:13 EDT

Argh...forgot who started this, but he later sent me some component
values for a simple 3 element pi filter with loaded Qs of 1 through 5.
I ran them through simulations on Touchstone (some good software which
you buy instead of buying a new car; the cost is equal) and I
recognized the plots almost instantly. Waaaay back in the late 70's or
early 80's there was an article by W7ZOI in Ham Radio mag, titled
something like "The Ultraspherical Filter", which he also called a
peaked low pass filter (PLPF). I did a lot of simulations on some
ofthe circuits in this article, and they agreed closely with the
figures, and those are similar to the higher loaded-Q low pass filters
I looked at this time. With higher loaded Q, you get much better
harmonic suppression, but at the cost of, well, higher Q--sharper
component value selection required, and also, surprisingly, reduced
bandwidth. Read the article if at all possible--it's fascinating to
study. The bottom line is that you can end up with a low passfilter
which is not a true low pass filter but has substantial ripple; as you
go from DC to the cutoff, the insertion loss increases dramatically,
perhaps more than once, and ends up at a peak (low insertion loss) near
the cutoff, then plunges sharply. But the width of that final peak can
be pretty sharp, and you could easily end up having to tune the filter.
 It's too complicated to go into in a brief e-mail, but please read and
study the article if you can find it--truly fascinating and
interesting. (And one interesting circuit he shows, although
admittedly not terribly practical, is a single low pass filter which
has a variable capacitor in the center, and by tuning that cap you can
move the cutoff from, say, 7 MHz to 28 MHz! Of course the insertion
loss goes up quite a bit on the high ends, but still a fascinating
circuit. I built one, tested it on a network analyzer [$35,000 piece
of HP equipment to die for] and it worked as advertised, though with a
bit higher insertion loss than advertised.) 73 and Queue Our Pea DE

Mike Czuhajewski, user of the UniBoard System @
The WB3FFV Amateur Radio BBS - Located in Baltimore, Maryland USA
Supporting the Amateur Radio Hobby, and TCP/IP InterNetworking

Search QRP-L Archives

[ QRP-L Archive | ]
[ 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 ]


This archive was generated by hypermail 2b29 on Fri Jun 02 2000 - 11:27:02 EDT