[NLRS] VHF low pass filters?

[Douglas H Reed]

<http://www.calculatoredge.com/electronics/bw%20tee%20high%20pass.htm>
The above was just the first site I found. I've got an old DOS-based
program somewhere here that does a better job because it is
interactive and calculates a filter based on design requirements, and
also allows recalculation based on my modified values.

<http://www.wa4dsy.net/filter/filterdesign.html>
This is a better choice of program to use. Only it doesn't let you
tweak for standard values and calculate a new filter response based on
adjusted values. According to this program, a Chebyshev 9-pole
Highpass filter will do exactly what we want. A lesser filter will
work if you start higher in frequency, such as 7-pole and 92MHz
cutoff.

Another option is the Bandstop filter calculator on the above page.
Try 52MHz center and 25 or 30MHz bandwidth. The Chebyshev version of
the filter requires just 2 values of inductor and 1 or 2 values of
capacitor. The Bessel version has much better stop band but uses 6
different part values and some of the values are small enough that
stray capacitance would be an issue. If you will never go above
50.5MHz, just make 1 or 2 or 3 tuned circuits for 50.2MHz and it will
be good enough.

You will NOT get the calculated stop band rejection using SMT parts
because the inductor Q is too low, but the filter is something you can
build with a limited selection of parts or even wind the coils by hand
if you have a grid-dip oscillator or other test equipment to adjust
the resonant frequency of each LC pair. The Toko 5mm coils are nice to
work with.

My favorite filters were based on a slightly different design which
put nulls on frequencies I specified. The filter I eventually built
started life based on the design in the CSVHFS conference paper, but
then I tweaked it to use standard value components and put nulls at
50MHz and 28MHz. I'll have to see if I can find that again.

The minimum equipment to build the filter would be a grid-dip
oscillator. If you have a tunable signal generator, especially an old
analog tunable such as Eico, B&K, Knight, Heathkit, it can easily be
done. If all you have is a 6M receiver it will be more difficult. Of
course if you have a network analyzer or spec-an with tracking
generator, it is no problem at all....

What you need to do is start with the calculated component values for
the filter and calculate the resonant frequency of each pair. Then
build and tune each pair individually, and finally install them as
required in the original circuit. If you are going to wind coils,
start with the required capacitor values and adjust the inductor until
it resonates where it should. The important thing is the relative
values and the actual resonant frequency of each pair.

73, Doug Reed, N0NAS.