Check the May and June issues of QST from 1998, as well as the July/Aug
issue of QEX from 1999. The W3NQN designs are highly regarded and are
the basis for the excellent filters sold by Array Solutions. Tim, K3LR,
was highly involved in their development and evaluation. Another very
good article by W3NQN, this time on lowpass filters with additional
second harmonic filtering, is in the February 1999 issue of QST.
If you want to play with filter designs yourself, try downloading the
terrific program ELSIE, available at
http://www.tonnesoftware.com/elsie.html. The student version, with some
restrictions on complexity, is free but is sufficient to handle the
W3NQN designs. ELSIE is referenced in the W3NQN articles as being a DOS
program, but the version available now was rewritten for Windows. The
program is extremely versatile, easy to use, and gives you lots of nice
plots to evaluate. I don't really need higher complexity than the
student version will handle, but I'm thinking of buying the upgrade
version simply because I feel guilty freeloading such a fine program.
One of the things I did with ELSIE was to take one of the W3NQN low pass
designs form his February 1999 QST article and reconfigure it as a high
pass filter (you just have to switch the signs of the reactances and
shift the notch frequency). My intent was to build two high power (2KW)
filters --- a low pass 40m filter and a high pass 20m filter --- to
reduce interference between a 40m yagi and a triband yagi on the same
tower. I'd have two feedlines, one for the 40m antenna and one for the
tribander, and I wanted to try to avoid bandpass filters because they
typically have higher passband loss. Plus, of course, I'd only need the
two filters instead of one for each band since the pass loss of the 20m
filter remains low to above 30 MHz. I haven't built the filters yet
but ELSIE says it should work, assuming of course I use good quality
caps and replace the toroid inductors from the article with large
Jim Lux wrote:
> Prompted by the recent discussions of multiple feedlines.
> Is there a set of designs for such filters floating around? (doesn't
> have to be a great one, just something to give someone to analyze for
> a project). Probably a 10,15, and 20m filter would be sufficient.
> I glanced through the ARRL antenna book and didn't find any, and
> before I grind through a synthesis, I thought I'd ask if there's some
> "cookbook" design out there to play with.
> Jim, W6RMK
> TowerTalk mailing list
TowerTalk mailing list