--- Gary Hoffman <ghoffman@spacetech.com> wrote:
> This is a heck of a good idea. Wonder if there is
> more info available ?
Since decided to give this shot just last night the
info bellow is pretty much all that there is. Although
I have been kicking this basic idea around for some
time now, just finally got an SDR-1000 recently to
actually try it with. I'm sure with some more thought
and hardware interfacing work the TX side of this DSP
IF could be done in a similar fashion as well.
> Wonder how this would work with other software
> defined radios ? Maybe some real cheap ones?
I have no doubt this could be done with just about any
other SDR hardware and software, in many ways the
SDR-1000 is actually overkill for this application, it
is however the easiest place to start such an
experiment today. All of the other possible approaches
that I'm aware of would require much more work to make
play at all (I tend to run out of patience and
ambition for such projects ;)).
The "Time Machine" board from Expanded Spectrum
Systems or the Kanga R2Pro direct conversion RX kit
could also be used as a starting point for a project
like this.
I've also been looking, watching, and waiting for
software to use for this kind of a project for some
time now. Some other software possibilities could be
SDRadio from I2PHD, it is only an RX and is rather
limited for what settings you have to work with, and
it is freeware though (for ham radio use only). Linrad
from SM5BSZ would be another possibility, it is quite
mature but requires Linux, it too it is freeware
(again for ham radio use only).
At one time I thought about using an RX320 (or even a
Pegasus) as a "DSP" IF to the Corsair II like I've
just done with the SDR-1000, I never actually tried it
though. My reluctance was because it would add several
more conversion stages, not reduce them. Also the
RX320's VHF 1st LO is comparatively noisy. One of my
main goals has been to avoid internal RX noise sources
such as phase noisy LO's, (or other oscillators) and
mixers; and also to specifically reduce the number of
mixers in the RX chain to a bare minimum. The Corsair
has an inherently quiet 1st LO, and the SDR-1000 is
based on a "Quadrature Sampling Detector" design which
is arguably not a mixer at all but is instead a
"sampling" device (the SDR-1000's RF to baseband/audio
is "direct conversion" regardless). So the Corsair II
/ SDR-1000 combination as I have it now configured
could be argued to be either a single or double
conversion RX depending on how you classify the QSD of
the SDR-1000.
Duane
N9DG
> ----- Original Message -----
> From: "Duane Grotophorst" <n9dg@yahoo.com>
> To: <tentec@contesting.com>
> Sent: Saturday, December 04, 2004 5:02 PM
> Subject: [TenTec] First DSP IF Corsair II?
>
>
> > I don't know if any one else has done this before
> but
> > I have created a DSP IF Corsair II, well sort of.
> The
> > experimental lash-up that I'm playing with took
> all of
> > 15 minutes to setup and get going. All that I did
> was
> > connect the antenna input of the Flex Radio
> SDR-1000
> > through a capacitor/tap to connector #73 of the
> Xtal
> > Filter Board (81252). This is the RX output side
> of
> > the 9 MHz 1st IF filter of the Corsair II. This
> first
> > IF filter then essentially becomes a 2.4 kHz
> roofing
> > filter for my little experiment, sort of like an
> Orion
> > :). The SDR-1000 was then tuned to 8.999800 MHz
> to
> > hear the IF signal coming through the Corsair II I
> 9
> > MHz IF filter. Tuning around the bands is
> accomplished
> > with the Corsair II's VFO, the SDR-1000 is run in
> USB
> > mode for all bands except 17M which is opposite,
> this
> > is necessary for the same reason you must use SB-R
> to
> > operate SSB on that band normally. With the
> SDR-1000
> > tuned to 8.999800 MHz the frequency display of the
> > Corsair II is exactly on when the audio pitch of
> the
> > RX signal is correct in both receivers. Since I'm
> only
> > tapping the 9Mhz signal off of the RX path of the
> > Corsair II I can actually listen to both the
> Corsair
> > II and SDR-1000 simultaneously; so making
> performance
> > comparisons is easy.
> >
> > Some impressions and observations:
> >
> > 1. Much quieter and cleaner recovered audio.
> > 2. Incredibly sharp "IF" filters in the SDR-1000,
> they
> > can be set to as little as 10 Hz and you can still
> > copy CW FB at this bandwidth, and no ringing.
> > 3. Automatic notch filter, the Corsair II did not
> have
> > that before, the SDR-1000 now provides it.
> > 4. The SDR-1000's DSP processing does add ~100 -
> 200
> > ms of delay to the audio (pretty much excludes
> ever
> > using QSK CW with this configuration).
> > 5. The Corsair II alone does have a slight edge in
> > absolute sensitivity from the tests I've done so
> far
> > (but I haven't calibrated the SDR-1000's settings
> very
> > carefully yet).
> >
> > The SDR-1000 is no slouch for dynamic range all by
> > itself but coupling it with the RX RF filtering
> and
> > low phase noise 1st LO and 2.4 KHz IF filter of
> the
> > Corsair II does make for an extremely pleasant
> > listening experience. By tapping into the first IF
> > output of the Corsair II I avoid all of the mixer
> > noise contributions by the additional mixers
> needed
> > for the Corsair II's PBT functions. It also
> bypasses
> > the Corsair II's product detector and all of its
> > low-level audio stages and AF filtering (all of
> which
> > are rather noisy by today's standards).
> >
> > Since the Corsair II is essentially running
> normally
> > it provides most of the AGC action to prevent
> > overloading the SDR-1000. Even so with 2-3 roughly
> > equal and strong CW signals inside the Corsair
> II's
> > 2.4 IF filter the SDR-1000 still does not suffer
> any
> > blocking effects that I can tell from my tuning
> about
> > the bands so far (I even tested it a bit during
> the
> > 160M contest last night). You can easily pick out
> > pretty weak signals between two strong signals
> <200 Hz
> > away, and you also get a panadapter display for a
> 2.4
> > KHz slice of the spectrum that is defined by the
> > Corsair II's 1st IF filter (so I can tell when
> those
> > strong signals are there).
> >
> > All in all a fun little experiment that did not
> take
> > long to set up, I will play with it a bit more to
> see
> > if it is truly useful from a practical standpoint
> or
> > not. Either way there are many other fun things to
> try
> > with SDR-1000 that I haven't gotten around to
> messing
> > with yet. This experiment has also confirmed my
> > suspicions about where most of the internal RX
> noise
> > of a Corsair II actually comes from.
> >
> > Duane
> > N9DG
__________________________________
Do you Yahoo!?
Meet the all-new My Yahoo! - Try it today!
http://my.yahoo.com
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
|