On 28 Oct 2002 at 14:10, George, W5YR wrote:
>
> Where this all falls apart is when an op wants to copy a low beatnote
> and yet use a fairly wide filter. I once had a fellow complain that
> his new Icom 756PRO was defective because when set for a 450 Hz offset
> and 800 Hz bandwidth, he could hear QRM from "the other side." No
> amount of explanation could convince him that such a setup was doomed
> from the start . . .
>
I'm not sure I understand your statement that this setup was doomed
from the start. Perhaps it was for the 'PRO, although I thought they
used the phasing method of sideband selection and generation.
My Pegasus uses the "Weaver Method" which is a type of phasing
method and is quite capable of setting the CW offset at 450 hz and
bandwidth at 800 hz or more without hearing anything on the other
side of zero beat. I just ran a test with a 20 over 9 carrier on a RX-
320, Pegasus and Argonaut V. In each case I set the offset at around
400 hz and bandwidth at 1000 hz. Also in each case, I heard a 20 over
9 signal on one side of zero beat and nothing on the other. The reason
for this is that the phasing method does not depend on the skirts of
the filter to reject signals on the opposite side of zero beat. It uses
the phase relationship between two phases of the signal and two phases
of the BFO, each shifted 90 degrees. When phasing was done with
discrete components, component tolerance and limitations on circuit
designs limited the suppression of the opposite sideband to something
around the 40 db range over a typical 2 to 3 khz bandwidth. Since DSP
is a math process, the tolerances of individual components goes away
and one has a near perfect single sideband detector or generator.
BTW, just for the fun of it, I took the bandwidth to 3 khz on the
Pegasus with a 400 hz CW offset and could hear the desired signal over
the entire 3 khz bandwidth and again nothing on the other side!
-73-
-Lee-
|