[Premium-Rx] Cubic CDR-3280 Digital I + Q output

[Mack Rogers]

Now to the important question David. Hoew long will it take you to complete the work and how much will it cost me? LOL!!!

N4VGB
Mack Rogers




--- On Tue, 1/18/11, David I. Emery <die at dieconsulting.com> wrote:

> From: David I. Emery <die at dieconsulting.com>
> Subject: Re: [Premium-Rx] Cubic CDR-3280 Digital I + Q output
> To: "Mack Rogers" <n4vgb at yahoo.com>
> Cc: wombat5 at ntlworld.com, Premium-Rx at mailman.qth.net
> Date: Tuesday, January 18, 2011, 3:59 PM
> On Mon, Jan 17, 2011 at 11:52:36PM
> -0800, Mack Rogers wrote:
> 
> > Indeed that is the problem. This data format isn't
> something that a
> > soundcard would be able to process, since those
> process the I&Q on the
> > L&R audio input channels. As far as I know any of
> the commonly available
> > SDR programs could process the data, IF some way of
> inputting the data
> > stream to a PC could be found.
> 
>     Someone with reasonable modern hardware
> skills could pretty
> easily create some FPGA programming for one of many
> moderately priced
> development boards for FPGAs of the sort that include USB
> ports and
> import the bit stream through USB.   This
> obviously requires USB
> programming experience at the driver and chip level but
> lots of the code
> required is available in development kits for the chip
> involved.
> 
>     Another approach, which would also work,
> is to use the older
> Cypress FX2 chip along with a much smaller FPGA (or
> discrete logic if
> you like that) - one basically needs to clock the bits into
> a register
> (or for that solution 4 byte wide registers) using the
> clocks on the
> RACAL interface and the framing signal and then clock the
> registers into
> the FX2.    Cypress supplies most of the required
> software and driver
> code as part of a development kit.   There
> may even be some development
> boards with the FX2 on them along with the other required
> support
> glue... though there are now some large FPGAs with the
> equivalent of the
> FX2 in them in addition to a large number of programmable
> logic blocks.
> 
>     I might add that TenTec has a similar
> digital output on the
> RX-331 and RX-340 (no idea exactly HOW similar at a low
> level) and WJ in
> the 90s standardized on a parallel 16 bit wide bus and lots
> of their
> equipment from that era has the 50 pin parallel bus for
> outputting
> audio, I and Q and other things...
> 
>     The right FPGA based design could easily
> handle several of these
> different buses with reconfiguration of the FPGA by
> downloading
> different versions of the logic.   And USB
> 2.0 is reasonably future
> proof... not going away  in any real hurry.
> 
> 
> > I've been like a dog chasing his own tail on this
> situation for about
> > 4 years now. I dunno what the gov uses this format of
> I&Q data stream
> > for but it sure would be interesting to find out!
> 
>     All this output is is a better than
> (most) soundcards A/D which
> converts analog I and Q from some place internal to the
> radio to the
> same stream of 16 bit I and Q samples as a computer sound
> card would.  
> But  because there is no audio processing between the
> I and Q demods and
> the A/D the quality of the sampling is likely better than
> what one might
> get out of a typical soundcard with its audio oriented
> input circuit -
> and the actual sample rate clock may also be more stable
> and accurate.
> 
>     
> > To complicate the matter even further each active
> output on the J7
> > connector has an inverted output also.
> 
>     That means it is designed for
> differential transmission, which
> would be a very natural
> interface.   Probably more research will
> show
> this will work quite well with some particular differential
> receiver
> chip family. Also one would likely want to use twisted pair
> cables for
> these signals if one intends to run them over any distance,
> and
> differential driving is what is normal for twisted pairs.
> 
> 
> > Who knows, maybe one day I'll be scrounging in some
> surplus dealers
> > junk at a hamfest and there will be a strange looking
> box with Cubic
> > Communications on it that cures this problem for me.
> 
>     There may be some PCI cards that input
> this signal... probably
> highly specialized (and thus VERY expensive to buy new) but
> the sort of
> thing that shows up in the almost free whazit bins at
> fleas... only
> problem is recognizing them (and finding drivers and
> documentation).
> 
>     My USB solution suggestion would likely
> yield an open source
> design that would not require unobtanium based proprietary
> stuff... but
> does of course need some engineering talent to implement.
> 
> 
> 
> > 
> > N4VGB
> > Mack Rogers
> 
> -- 
>   Dave Emery N1PRE/AE, die at dieconsulting.com 
> DIE Consulting, Weston, Mass 02493
> "An empty zombie mind with a forlorn barely readable
> weatherbeaten
> 'For Rent' sign still vainly flapping outside on the weed
> encrusted pole - in 
> celebration of what could have been, but wasn't and is not
> to be now either."
> 
>