Time response IS very important in the REAL world where switch clicks,
lightning static, and power line noise accompanies the desired signals.
Its not all about separating bandwidth limited signals (voice, cw,
PSK31...) Filters with bad time response (typical of INRAD, YACOMWOOD,
and Collins filters with Tchebychev amplitude responses) ring
excessively to the point of incapacitation those receivers in the
presence of those noise sources. Filters with adequate time responses as
sold by TT, continue to receive between the clicks of light switches,
lightning, and line noise (which only occurs at the peaks of the local
AC power cycle).
The same linear phase response that handles data well also passes these
noise pulses (quite wideband) without causing nearly so much destructive
ringing.
The typical Tchebychev filter (such as the mechanical filter in an
S-Line) when hit with line noise fills in the time between peaks with
ringing. I've watched it with a scope, seeing individual pulses using
the AM IF transformers in my 75S3C and seeing a solid unmodulated
envelope with the same line noise through the SSB filter. Users of such
filters quit when lightning static occurs because each pulse is
converted into a long crash obscuring desired signals. Neither of these
happen with rounded response IF transformers, and barely occur with TT
filters. Lightning through a TT filter cascade, comes out as clicks, not
crashes. That's why good time response is of great benefit. TT
understands, I don't think INRAD understands, and I know Collins and
YACOMWOOD don't understand.
73, Jerry, K0CQ
--
Entire content copyright Dr. Gerald N. Johnson, electrical engineer.
Reproduction by permission only.
|