Bill,
I understand your point. The only thing I am trying to say is when you say
audio DSP it makes it sound like the DSP can do no more than a good external
audio DSP unit like the timewave units. These DSP add devices are not in
the AGC loop and cannot do anything like the Orion's DSP.
The DSP in the Orion has a 100 db range all by itself. If the entire signal
path ahead of the DSP can handle a 100 db dynamic range, then the DSP alone
could provide a 100 db dynamic range. The roofing filters just add to this
by removing more unwanted signals on this path. The point is the DSP in the
Orion has more range and sharper filtering than anything we have seen before
in an amateur rig. I don't believe any manufacturer has achieved a 1.05 to
1 filter shape factor before.
By the way, UPS says my Orion should arrive tomorrow!
Carl Moreschi N4PY
Franklinton, NC
----- Original Message -----
From: "Bill Tippett" <btippett@alum.mit.edu>
To: <tentec@contesting.com>
Sent: Tuesday, May 06, 2003 12:41 AM
Subject: [TenTec] Hello Orion Users
> This discussion of semantics is missing the point. The
> point of this discussion was at what frequency/IF stage the DSP
> filter operates. The Orion/Main RX has 3 IF's at 9 MHz, 450 kHz
> and 14 kHz. My point was that DSP's can be made to operate at
> "audio frequencies" fairly easily, because ADC's, DSP's and DAC's
> are economical at audio frequencies (sub-20 kHz) but NOT at RF
> frequencies. I think we are picking at bones because the DSP
> clearly operates on the 14 kHz IF, NOT at 450 kHz or 9 MHz.
> It is impractical (i.e. very expensive) or impossible to get
> commercial 24 bit (high resolution necessary for dynamic range)
> ADC's and DAC's that would operate at 450 kHz...much less at 9 MHz
> where the analog roofing filter does the heavy lifting. Just to
> cite an example, we had an 8-bit 20 MHz ADC at HP in the early 90's
> that sold for around $8000. To do DSP at 9 MHz would require an ADC
> with 24 bit resolution at an 18 MHz sampling rate (twice the IF
> frequency for no information loss)...and Lord only knows how many
> GFLOPS in the DSP itself.
>
> I don't reallly care if we call 14 kHz Audio, IF or RF...it
> remains 14 kHz, and is done at the 3rd IF stage well past filtering
> necessary for good interfering signal rejection is done by the first
> IF roofing filter at 9 MHz. The roofing filter is primarily why we
> see IMDDR3 of 101 dB...NOT DSP filtering at 14 kHz. For an example
> of a purely analog system that even exceeds Orion's performance, check
> out W8JI's "heavy modified" R4-C using a 600 Hz Sherwood roofing
> filter below, and note the difference with the stock R4-C's (which
> had wide roofing filters) at 48 and 68 dB IMDDR3 respectively:
>
> http://www.w8ji.com/receiver_tests.htm
>
> The medium mod has IMDDR3 of 92 dB and the heavy mod has 118 dB
> and that's at 2 kHz spacing...not 5 kHz. Of course Orion might
> also do better if the optional roofing filters are used.
>
> 73, Bill W4ZV
>
>
>
>
>
>
>
>
>
> _______________________________________________
> TenTec mailing list
> TenTec@contesting.com
> http://lists.contesting.com/mailman/listinfo/tentec
|