🚀 go-pugleaf

jgd@dixie.com (John De Armond) writes:

(with regard to using a stub as a notch filter:)

>Depends on the Q of the transmission line.  If very high quality coax,
>hardline or even better, air hardline is used for the stub, it will be 
>quite narrow.  My worry would be in making it too narrow for the 
>fairly wideband FM signal.  For 2-way work, where the bandwidth is
>much less, a piece of 1" CATV coax works just fine to notch out offending
>intereference.


Also depends on impedance levels.  The depth of notch will depend
on the Q (e.g., on the stub resistance at resonance, compared with
the impedance of the line the notch is being put in.  But off the
notch frequency, the attenuation will depend more on the impedances
of the lines involved.  A moderate Q or a high Q stub of the same
impedance line will each have nearly identical impedances 10% off
resonance, but will have far different impedances at resonance.

Also, most of the line loss in the stub at these frequencies will
be copper resistance loss, not dielectric, so it's much more the
diameter and length of the coax and the construction of the conductors
than the dielectric, assuming polyethelene or Teflon or air (or vacuum
;-) dielectric, that's important in optimizing Q.  Interestingly,
polyethelene dielectric will cause the inner conductor to be smaller
than air dielectric for a given impedance and outer conductor
diameter, thus raising the loss, _but_ the length needed for 1/4
wave at a given frequency will be less, so there will be less loss.
It's left as an exercise for the reader to figure out which factor
dominates.  (Hint:  does it vary with impedance?  With frequency?)