[hpsdr] The Basics (with updates)

[Lyle Johnson]

>> Are OZYMANDIAS, MERCURY and SASQUATCH going to use the Altera Cyclone II FPGA?
> 
> OZY will use a 208 pin QFP Cyclone II (EP2C5)
> 
> The MERCURY will probably use the same.
> 
> Don't know about SASQUATCH.

Sasquatch initially will use an Actel Flash-based FPGA.  Not merely to 
be different -- I'd normally use an Altera part for consistency with the 
rest of the project -- but because the board is really a 
prototyping/development board for the AMSAT (www.amsat.org.) Software 
Defined Transponder (SDX).  In space, RAM-based FPGAs, like the Altera, 
have to be constantly monitored to prevent single-event upsets (SEU) 
from reconfiguring the logic.  Flash based parts are not nearly as 
susceptible.

The AMSAT board would normally be "hosted" by a CAN (controller area 
network) control module which also supplies power.  For the ATLAS case, 
we merely install a DIN connector instead of the CAN module.

It will also have other features that are more suited to the space 
environment, like error detection and correction (EDAC) memory system, 
and dual Flash chips for booting.

Still, it will be an interesting and very capable board, and one which 
can be used for learning, which is really what this is all about.

And it still might be Altera FPGA based and not have EDAC memory.  A 
strictly HPSDR version.  Depends on the workload and the effort to 
convert the logic between device families.

Watch this space...

> A couple of other things to throw out there:
> 
> 1. GPIO board - I would like to see a general purpose I/O board that
> has buffered/protected industrial strength input and outputs.  Maybe
> an optical encoder interface on it also.  I would envision that this
> GP I/O board would communicate across the backplane using an
> SPI/Microwire interface since I2C is probably too slow for this.
> 
> 2. TX System - My goal is to put together a complete SDR RX/TX.  Right
> now we are building the RX components.  We need to start thinking
> about TX.  For sure I'd like the TX to be independent (can operate at
> the same time) from the RX system.  It is not clear if it should be a
> DDS-QSE type of system or a class E amp/PWM modulator type of system. 
> I am leaning towards the latter if possible.  Or maybe a DAC/DUC based
> system?  Lots to think about...

Sasquatch will have the logic needed to implement a highly efficient VHF 
transmitter using envelope elimination and restoration techniques. 
There is still some RF stuff required, of course, which is intended to 
hang off of some expansion headers, at least in the AMSAT version.

I'd like to see a DAC/DUC implementation, too, that could be driven over 
the backplane from Sasquatch.

> 3.  PROTO board - The PROTO board has already been mentioned.  This
> would be great for experimentation.
> 
> 4. DDS board - It would be nice to have one for at least general
> purpose signal generation for VNA type applications.  I would envision
> at least two sets of independent quadrature outputs.

Yes!  Using 99xx series parts and driven with a high frequency 
oscillator and not using the internal PLLS :-)

73,

Lyle KK7P


 1146671111.0