[TenTec] SDR-1000 vs Orion vs SDR-X

[Bill Tippett]

K5KVH wrote:
  >It seems time to reaffirm that the SDR 1000 is a direct conversion radio and
can never, no matter how much software it has, equal the performance of a
Ten Tec receiver in the Orion.  You can't put selectivity after the front
end and fix problems inherent to the front end.  That is what the close by
signal sensitivity to overload is about with DC radios.

I replied:
          Stuart this is contrary to my understanding
and ARRL's Product Review as reported in October QST.
Could you be more specific about your comment and
compare appropriate ARRL measurements (IMD, BDR, etc)
from their Product Tests?

         For my own curiosity I checked the two QST reviews
of the SDR-1000 (1st = 16 bit soundcard in April 2005 QST,
2nd = 24 bit Delta 44 soundcard in October 2005 QST) with
the January 2004 review of Orion.  Generally the data
below is Preamp Off, 500 Hz, 14 MHz and 5 kHz spacing.
In the case of the second SDR-1000 review, I did use
a preamp setting to get MDS within the same ballpark,
and the review also notes that the measurements on
that unit are identical at 2 kHz spacing versus the
5 kHz results compared below:

Radio   MDS     BDR     IMD     2nd order intercept

Orion   128     130     92      63
SDR(1st)        127     93      86      69
SDR(2nd)        130     111     98      74

An ad hoc group of Flexers is working on an even better
soundcard (named the Janus project) than the Delta 44
used in SDR(2nd) above.  Initial results indicate MDS
of -160 dBm and dynamic range of 120 dB for Janus but
these may not translate directly to the measurements
above.  Typically direct conversion SDRs are most
dependent on soundcard (i.e. ADC and DAC) performance
(as can be seen in SDR(1st) vs SDR(2nd) above.  More
info on the Janus project:

Update 26 May 2006
The results of the ADC bake-off are in and the AK5394A is the clear 
winner for our particular application due to its flat noise floor at 192kps.
The measured figures of the prototype Janus using the AK5394A are:
Noise Floor = -160dBm (in an 11Hz FFT bandwidth) Dynamic Range = 
120dB ENOB = 20 bits
We are now at the final review stages of the Janus circuit and expect 
to be moving to prototype PCB manufacture very shortly.

http://hpsdr.org/wiki/index.php?title=JANUS

         Bottom line is I question Stuart's comment above
based on ARRL's actual measurements.  I feel the direct
conversion DSP approach Flex is using is already delivering
*better* RX performance than Orion and it is likely to
get even better when the Janus project is completed.

         SDR-X may be a very significant product because of:

1.  Excellent RX performance already demonstrated in SDR-1000.
2.  Modular configuration allowing multiple RX and TX
modules in the same mainframe (i.e true SO2R in one box).
3.  True phase-locked diversity with as many receivers
as you can afford (up to 64 modules total).
4.  New capabilities never seen before such as rejection
of inband interference correlated with a signal from a
second RX module (this is the most exciting to me).

         I agree with N9DG's comments that we may be on the
threshold of some very significant changes in the landscape
of our rigs and manufacturers.  Ten-Tec has done well by
being very responsive to performance oriented customers
and I have no doubt they will continue to do so, but the
competitive landscape may be changing at warp speed.

                                         73,  Bill  W4ZV

P.S.  For those who think Flex is a garage shop, keep in
mind that Gerald Youngblood K5SDR has started a number of
high-tech companies and is presently a hedge fund manager
(with easy access to venture capital markets).  I believe
Flex's real goal is commercial/government/military and
the SDR-1000 is just a warm up for the SDR-X.