> The 12 khz IF only moves in 2.5 khz steps. There would be no way to be
more
> accurate in the sweep display than 2.5 khz. You would end up with a plot
of
> vertical blocks, each 2.5 khz wide.
Not necessarily, the software could compensate. I'm envisioning a setup
where the software does frequency analysis of the incoming data (i.e. FFT) -
similar to what is done in a PSK31 "Waterfall".
Assume you wanted a 100 kHz Band Sweep from 7.0 to 7.1 MHz. Assume that the
roofing filter is 20 kHz wide.
Tune to 7.01 and sample the IF. That gives you the signals from 7.00 to
7.02.
Tune to 7.03: 7.02 to 7.04
Tune to 7.05: 7.04 to 7.06
Tune to 7.07: 7.06 to 7.08
Tune to 7.09: 7.08 to 7.10
In reality, you probably can't get a full 20 kHz from the IF, so you'll need
to take more samples with a smaller range. If you assume 10 kHz "good" scan,
you can do it with 10 steps: 7.005, 7.015, ... 7.095. You can control the
final resolution by the number of samples fed to the FFT. That in turn would
determine how fast you can scan. You need to dwell on a particular frequency
long enough to get the right number of samples. There's probably also some
settling time needed once the freqency change command is sent.
73,
Mark
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
|