[hpsdr] Further thoughts on the Phoenix RX/TX architecture

[Ray Anderson]

To realize an "industrial strength" HF receiver I fully concur with Bob
and others' recent remarks emphasizing the need to integrate the A/D
(AK5394A) on the same board and close to the ISD. With the IP3, MDS and
Dynamic Range metrics that Bob has demonstrated on a test board and
which the recent proposals by Phil and Ahti have the potential for
achieving, the co-location of the A/D is necessary to achieve the
performance that is possible.

 

 

I do get the sense that their may be a bit of confusion regarding the
ultimate objectives of the Phoenix board compared to the Horton/Casmir
boards.

 

 Phoenix was intended to have "Good" performance (much better than
adequate, but less than world-class). The goal was to design a
relatively low-cost entry level integrated HF TX/RX board that will plug
into the ATLAS backplane and work with the Janus and OZY boards to
achieve a vehicle that allows one to get 'on-the-air' with a minimum of
fuss.

 

On the other hand, as I understand it, Horton is designed to exhibit
'state-of-the-art' performance on receive. Towards that end, the
integration of the A/D is mandatory. Casmir will exhibit similar
performance on the TX side perhaps utilizing advanced envelope
elimination/restoration techniques. Also, the Horton/Casmir pair are
intended to allow UHF/Microwave operation with appropriate outboard
circuitry.

 

 

With the above in mind, I can see a couple ways the Phoenix RX design
can go:

 

1)         As originally proposed, a robust ISD and maybe a preamp
feeding the Janus board which in turn feeds into OZY. Connect an antenna
to Phoenix and a PC running PowerSDR or DttSP to the OZY USB connector
and you are ready to go.....

 

2)         Alternatively, a robust ISD and maybe a preamp feeding an
integrated A/D, FPGA and FX2. This would allow true plug-and-play
operation with a single board without the need for Janus and OZY.
Connect an antenna to Phoenix and a PC to the integrated FX2 USB port
and you are good to go... 

 

Similar configurations can be envisioned for the TX part:

 

3)         A QSE fed from the PWM I/Q audio from Janus and OZY.

 

                           Or

 

4)         A QSE fed from I/Q audio from a yet-to-be-developed FX2,
FPGA, D/A chain located on the Phoenix board.

 

 

The versions utilizing the Janus/OZY combo would have the advantage of
utilizing the SASQUATCH demod/GUI capabilities when they become
available. It is conceivable that the integrated 'Janus/OZY' hybrid
functionality as described in 2) and 4) above could also work with
SASQUATCH, but that isn't clear to me at the moment.

 

 

At this point in the genesis of the Phoenix design we need to make some
early decisions on which fork to take: a design utilizing Janus/OZY with
'Good' performance, or a design using an integrated 'Janus/OZY' hybrid
interface that has the potential for 'Excellent' performance.

 

 

Comments/Discussion on the topic welcomed....

 

 

73 de Ray   WB6TPU

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/attachments/20060801/03165b5a/attachment-0003.htm>