N9NEO wrote:
Terry, Frank, et al.

Yes, an interesting article. Thanks for bringing it up to the front
burner again.

I don't understand his representation of the fading term as it has no
link to the actual fading frequency. I would think the distortion
would have to be created by the fade. The only frequencies discussed
are w_carrier, and w_audio.

Also his depiction of instantaneous fourier spectrum at the point of
deepest fade leaves me guessing. I would have liked to see him strobe
the fade in and out at a fixed frequency and see the spectrum output of
that. All in all a good article. I read it a few weeks ago briefly,
but haven't had the time to get into it. I'll for sure make the time
knowing there might be some meaningfull discussion here. Usually the
threads degenerate at warp speed.

My boss just bought a smokin nice spectrum analyzer that I might be
able to share with him.

And David, what was that comment about the Drake receivers? Do you
know what type of filters they use?


regards,
Bob
N9NEO


One big problem with recreating real world HF propagation is the random
nature of multipath. Lankford's sweeping RF notch recreates a simple
single fade, but can't
produce the multiple "comb filter" like effect I have noticed. I built
a simple HiFer
"beacon" that allows me to check a "real world" RF signal for just how
much
SN effected intelligibility. I found that many non obvious things had a
big effect.
One of the simplest source of nastiness is the post detection AF chain.
I had
doubted that minor things like the type of capacitor could "really make
any diference".
My whole goal was to improve my best radio as much as I could for as
little money
as possible. A synch detector is one improvement. But even though a
"good" synch
detector with a phasing fitler is supposed to reject non signal out of
band signals.
see :http://home.worldnet.att.net/~wa1sov/technical/sync_det.html
It does help, in some conditions.
I found the biggest jump in intelligibility for the least money came
from an "Improved
AM detector". See: http://www.amwindow.org/tech/htm/alowdisdet.htm
To qoute "There are no intermodulation terms to contend with. In fact,
functionaly,
this is identical to synchronous detection. The negative half wave
signal is derived
in a similar fashion."
For clear signals, with no or minimal fading this detector is cleaner
then any other
detector I have played with. With this detector, a decent 4KHz IF
filter, and using
offset tuning on AM with a 4KHZ AF LP fitler, the result is very close
in performance
a Synch detector. If you wish to follow my "saga", please look up the
various threads from last summer.


IF I were doing this project over I would likely go with the better
Kiwa fitler module and the improved AM detector. Along with an improved
AF chain. My wife's R2000 has a
4HKZ mechanical fiter in the AM-N/SSB slot, the improved AM detector
and a home
built 2W MOSFET audio chain. And she loves it.

Terry