All these tests were done with the roofing filters disabled. In fact with
the roofing filters enabled, no difference was found at all. The shape
factors of the DSP filters are so much sharper than the crystal filters,
that the crystals filters don't matter for this test. The crystal filters
will help with dynamic range and the IP3 intercept parameter.
Carl Moreschi N4PY
----- Original Message -----
From: "Bill Tippett" <btippett@alum.mit.edu>
To: <tentec@contesting.com>
Sent: Wednesday, May 07, 2003 7:58 AM
Subject: [TenTec] Orion Filters
> Hi Carl!
>
> I'm surprised at the variation in shape factors at very
> low bandwidths and believe you are seeing the combined effect
> of cascading the DSP filter with the roofing filter. Otherwise,
> I can think of no reason why the shape factor of the DSP filter
> alone would vary by bandwidth as it seems to. Remember that
> Orion automatically selects the next higher roofing filter
> whenever you select a DSP bandwidth. This could explain why
> the shape factor appears to improve as selected BW approaches
> each roofing filter BW (1.0, 2.4, 6.0 kHz) in your results.
>
> Could you try an experiment for us? I assume you have
> only the stock roofing filters (15 kHz, 6 kHz, 2.4 kHz and 1.0 kHz),
> with the automatic mode enabled.
>
> 1. FIrst disable automatic mode.
> 2. Set the DSP BW to 1000 Hz and enable the 1000 Hz roofing filter.
> 3. Record your measurements.
> 4. Repeat keeping the DSP BW fixed at 1000 Hz but successively
> enabling the 2.4, 6 and 15 kHz roofing filters (15 kHz might not be
> possible if it is only available for FM mode...will be interesting
> to learn if it will work for CW).
>
> I expect the 1000 Hz shape factor will degrade as you select the
> higher roofing filters and have less cascaded effect. If you select a
> DSP BW and enable a roofing filter much higher, you should be truly
> measuring the DSP shape factor alone, assuming the roofing filters
> center response is relatively flat.
>
> 73, Bill W4ZV
>
> _______________________________________________
> TenTec mailing list
> TenTec@contesting.com
> http://lists.contesting.com/mailman/listinfo/tentec
|