Local Charleston SDR hardware

Fri Jan 16 23:31:48 CST 2009

That's right Frank (about the few gates used).  You could do multi-band
receivers, etc in  the extra FPGA, like the QS-1R guys are doing.  The main
limitation is the 12-bit A/D.

I also think that down the road, a complete rig could be implemented in a
signle FPGA.  Just hook a D/A to another set of FPGA ports for the exciter.
Something like the "All-Digital SSB exciter for HF" in May/June 2008 QEX
comes to mind.  It uses the same D/A as the HPSDR Penelope (AD9744).

I think the Digilent board is a great test bed for learning FPGA.  It has a
lot of hardware support on the board itself, plus some option boards are
also availabble.  Make sure you get the larger gate-size FPGA, I think it's
1.2 million.  It is in a drop-down option when ordering the Nexys2.

I also got the cheaper A/D & D/A boards from Digilent to use with the
Nexys2.  The PmodDA2 has two 12-bit D/A converters that run upwards of
100ks/s for 19.95.  The PmodAD1 A/D board has two 12-bit A/D chips that can
sample up to 1Ms/s, for 21.95.

I'm wondering if these two would make a good basis for learning FPGA SDR
techniques at LF?

BTW, I've also purchased three FPGA books.

"FPGA Prototyping by VHDL examples, Xilinx Spartan-3 Version", by Pong P.
Chu, is essentially about how to code in VHDL for the FPGA on the Nexys2
board.  Some of the code can be used almost directly with the board.  Three
caveats:  1) it costs some $$; 2) it uses VHDL while John Schwacke, Phil
Covington, and HPSDR are all using the other language (Verilog); 3) it is
for general FPGA programming, not specifically SDR or DSP.  However, it is
good for background info about the Spartan-3 FPGA family.  I think the code
is downloadable.  ISBN: 978-0-470-18531-5. (c 2008)

"Digital Signal Processing with Field Programmable Gate Arrays", by
U.Meyer-Baese.  It is also somewhat costly.  But, it describes more SDR- and
DSP- specific programming, so it's a winner.  I don't understand it all, but
several lights did go on as I read it.  The code examples throughout the
book are in VHDL, but equivalent Verilog for every example is in an
appendix.  It also comes with a CD.  I recommend this one, if you are not
faint of heart, or easily bored.  ISBN: 978-3-540-72612-8.  There are older
versions of this book out there, make sure you get the Third edition (c
2007).

The third book that I have is "HDL Programming Fundamentals, VHDL and
Verilog", by Nazeih M. Botros.  I admit that I haven't looked at this book
as much as the other two.  Since it covers both FPGA "languages", I think of
it more as a reference.  This is more of a description of the languages, and
less of a coding cookbook.  It has a companion CD.  ISBN 10:1-58450-855-8.

Feel free to look for others, the above were some that I found interesting
after doing some research.  I purchased all three from Amazon, Half, or some
other place, at a reasonable discount.  If you get used, make sure to
request that the CD is also there (if the book has one).

I was hoping that more AMRAD people could get involved, and maybe we could
start monthly experiments, reported on in the newsletter and the web site.
This would fit right in with AMRADs goal of not just doing, but teaching.
We could start out with something VERY SIMPLE, such as blinking a few
lights.  Then, create a few audio tones, then decode touch tones (or
other?), then graduate to some low-freq RF.  Then, build a simple
transmitter (they are always easier than a receiver.  Then a simple
receiver.

73s
Terry
WB4JFI

> -----Original Message-----
> From: tacos-bounces+wb4jfi=amrad.org at amrad.org
> [mailto:tacos-bounces+wb4jfi=amrad.org at amrad.org]On Behalf Of Metavox
> LLC
> Sent: Friday, January 16, 2009 8:34 PM
> To: Terry Fox
> Cc: Tacos at Amrad.Org
> Subject: Re: Local Charleston SDR hardware
>
>
> Tacoistas,
>
> In addition to the attractive hardware there are two very attractive
> things from the web site about the FPGA Logic Design.
>
> First, there are only a small fraction of the available cells used.
> This mean lots of room for growth in complexity.  That growth can be
> used to increase the signal processing or perhaps a way to speed it up
> through more parallel processing.
>
> Second, "all FPGA code was implemented in Verilog and bit files were
> created using the free versions of the Xilinx development tools".  This
> means a cheap (as in free) way to learn how to program an FPGA to do
> some useful things.  As a learning platform, this is attractive.
>
> Is there any interest in several of us getting these things to play, er,
> make that learn with?  I am game but I sure would like some help.  I
> have never done an FPGA before but I recall a nice talk by Hal some time
> back. With more of us in the pool, we can help work out the hard spots.
>
> I will make may lab available, with tutoring, to do the surface mount
> assembly of the front end.
>
> Frank K0BRA
>
>
> Terry Fox wrote:
> > One of the people here in Charleston has been working on a
> less-expensive
> > direct-downconversion (DDC) receive solution for a while.  I have been
> > alluding to this project for a while.  At a recent meeting, he
> said that it
> > was OK to forward information.
> >
> > His name is John Schwacke, he is a PHD, and works at Medical Univ. of S.
> > Carolina (MUSC).  He has worked with Mel (W4MEL) in the past,
> back in DC.
> >
> > The hardware is based on a Digilent Nexsys II demo FPGA board
> (about $100),
> > and a sampling board of his design that plugs into the Nexsys II.  The
> > sampling board contains a simple (bypassable) 27MHz LP filter, a
> > variable-gain-amp (VGA), an 80Ms/s, 12-bit A/D, and an ASIC
> that does some
> > downsampling and VGA control.  I think there is also a
> bypassable attenuator
> > in the RF chain...  The PC interface is via USB.
> >
> > John is using GNURadio software under Linux to test the
> hardware.  He can
> > presently hear 2M repeaters (bypass the LPF), and signal
> generators on HF
> > (40M).  (yes, his built prototype is working)
> >
> > John has two more PC boards (reserved for Mel and me), and I am ordering
> > parts to populate these boards.  John says that he is willing
> to have more
> > boards made.  The estimated parts cost to build John's board is
> about $60.
> > So, at less than $200, you could build a DDC HF receiver.
> >
> > John has a web site showing this work:
> > http://sdrtrack.drupalcafe.com/
> >
> > John uses solder paste and an electric skillet to build these boards.
> >
> > The approach that John has taken is to use a less-expensive
> 12-bit A/D, and
> > preceed it with a VGA to reduce overload.  This is NOT as good
> as using a
> > 16-bit A/D, but is much less costly and easier to build.
> >
> > The info is still preliminary.
> >
> > Please check it out.
> >
> > Terry
> > WB4JFI
> >
> > _______________________________________________
> > Tacos mailing list
> > Tacos at amrad.org
> > http://www.amrad.org/mailman/listinfo/tacos
> >
> >
> > ------------------------------------------------------------------------
> >
> >
> > No virus found in this incoming message.
> > Checked by AVG - http://www.avg.com
> > Version: 8.0.176 / Virus Database: 270.10.8/1897 - Release
> Date: 1/16/2009 6:52 AM
> >
> _______________________________________________
> Tacos mailing list
> Tacos at amrad.org
> http://www.amrad.org/mailman/listinfo/tacos
>