Many of us have been sucked in by the “early adopter” syndrome. Just look at
the foolish long lines at the Apple Store when a new phone comes out...
I personally bought a new 2011 Nissan Leaf. After 36,000 very reliable miles
but four winters of freezing my you-know-what off going to work in winter, I
sold it for less than 1/6 of the purchase price…
I hope to not get similarly caught in Ham radio buying errors. I’ll struggle
through with my analog radio with IF DSP until it is proven what works and what
does not work in direct digital SDR’s.
As far as testing goes, within the context of ranking SDR radios, would it not
be a good start to first measure what was formerly known as “sensitivity” above
the noise floor, then apply a signal of 140 dB higher than that level and see
what happens in the receiver? Does it clip or does it continue to output a
desired signal that is linearly tracking the input signal? This could be done
with both in-band and out-of-band large signals to test both the signal dynamic
range in band and also the effectiveness of any band-limiting filters employed.
Then one could apply a known complex waveform (“standard voice", so to speak)
to a calibrated SSB transmitter and compare the Fourier analysis of the audio
coming out of a given receiver against the calibrated “standard voice” signal.
The closer the receiver output is to the known modulation waveform, the better
the radio.
This is not unlike the “disk compare” test used early on in the CDROM business
(with which I was involved in the mid 1990’s). A simple digital compare of the
final replicated CD against the glass master was a pretty good measure of the
overall manufacturing process. This laid the groundwork for determining
acceptable levels in tests known as BLER (Block Error Rate), Radial Noise, pit
jitter and so on.
Finally, one could apply something like “standard CW” to the same calibrated
transmitter and observe the quality of the decoded CW as a measure of a radio’s
ability to handle what amounts to a square wave (overshoot, rounding etc.).
To complete the test, one could then compare these performance measures against
the same measures when a condition exists in which the 140 dB above sensitivity
threshold interfering signals are applied. The less the test signals change,
the better the receiver.
I agree that analyzing the transmitter for spectral purity is equally
important, but this is a lot easier to do, since spectrum analyzers can pretty
well tell the story about the transmitter. The transmitter test, for what is
now called “synthesizer phase noise” could be performed just as it is now,
using the transmitted signal, to determine if the designer(s) did their jobs
well on the clock oscillator.
Just suggestions, not dogma…
Gary
W0DVN
> On Sep 11, 2016, at 1:08 PM, Jim Allen <jim.allen@longhornband.net> wrote:
>
> Be glad you didn't pay $1400 for an Apple IIP with 16k of memory, no mass
> storage or printer, no software other than BASIC and Little Brick Out, or
> monitor, back when you could buy a brand new Volvo for $4000!!!
>
> 73 Jim Allen W6OGC
>
> On Sun, Sep 11, 2016 at 10:47 AM, Greg S via TenTec <tentec@contesting.com>
> wrote:
>
>> ........
>>
>>
>> I am finding it hard to know when/where to jump into the SDR transceiver
>> fray..... Prices are being forced down, and technology is changing FAST on
>> some of the open source code. I bought an SDRplay, and have had a TON of
>> fun with it, so I am seriously considering the next level. It still hurts
>> to think about my 40" LCD, 60Hz, 720P Samsung "dumb" TV that I paid over
>> $1700 for in 2007, but our old TV died, and we jumped in where we thought
>> appropriate. (It still works perfectly, but draws 3 times the energy of our
>> 60" TV that cost 25% of that $1700!!)
>> I wouldn't want to be in Mr. Dishop's shoes in this market, but do hope
>> he comes up with a "winner" the first time around!!! Long Live TenTec!
>>
>> 73, Greg, KC8HXO
>>
>>
>> SNIPPED SNIPPED SNIPPED
>> IMO there is no longer any easy way to rank them based on one or two
>> specifications alone.
>>
>> Too many factors affect SDR receiver performance.
>>
>>
>>
>> I am in favor of removing all SDR radios from the list and placing them in
>> a
>> separate list, but I have no idea how we should do this.
>>
>> My best guess would be to use NPR testing such as Adam Farson is
>> conducting.
>>
>> See: http://www.ab4oj.com/test/docs/npr_test.pdf
>>
>>
>>
>> Adam has two lists, one for traditional heterodyning radios and another for
>> SDR radios.
>>
>> But this is not perfect either.
>>
>> As you see, the 7300 tops the list and as we all know, it has overload
>> problems due primarily to improper gain distribution throughout its front
>> end stages. Adam, Rob, and every test review I have read point out its
>> overload problem.
>>
>>
>>
>> Note that the two ANANs are ranked above the 6700, even though they do not
>> have dedicated BPF's for each ham band.
>>
>> I'm not sure how Adam is ranking the SDR radios. It is not only NPR
>> figures.
>>
>>
>>
>> At this point we are way outside (above) my pay grade. I have no idea
>> other
>> than to put these radios in the hands of contesters and let them report on
>> their experience.
>>
>>
>>
>> In summary, we have come a long ways since first reporting on IP3 in the
>> 1970s.
>>
>> We cannot compare SDR radios to heterodyning radios based on their IP3 or
>> DR3 results.
>>
>> At this point in time, I am not aware of any agreed method of testing and
>> reporting performance levels of SDR radios.
>>
>> There are ongoing discussions between Rob, Adam, Bob Allison (ARRL) and Ken
>> (ex Ten-Tec president) on how to do this.
>>
>>
>>
>> ON THE BRIGHT SIDE, this new technology has brought us great improvements
>> in
>> performance and as we will soon see, at an affordable price.
>>
>> The 7300 was just the beginning.
>>
>>
>>
>> 73
>>
>> Rick, DJ0IP
>>
>>
>>
>>
>>
>>
>> _______________________________________________
>> TenTec mailing list
>> TenTec@contesting.com
>> http://lists.contesting.com/mailman/listinfo/tentec
>>
> _______________________________________________
> TenTec mailing list
> TenTec@contesting.com
> http://lists.contesting.com/mailman/listinfo/tentec
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
|