On 8/19/2013 10:28 PM, wb4...@knology.net wrote:
Warning: rather long reply follows.
If you don't care, or have your mind made up, please skip this email.
Fred asked for some education, so here goes... I'm not an expert, but
these are my personal ramblings from what I think that I have learned.
If you don't agree, and the moderator doesn’t like this discussion here,
maybe we can find another place to discuss? Or directly?
Hi Terry, thanks for the reply. I'm thinking I'm not the only one
seeking the education so I'll reply to the list. If that's
inappropriate, I'll take it off-list. I'm also not going to quote your
reply, those interested already have it.
So far, I've learned that 32-bit ADC's have apparently not become as
ubiquitous as I had surmised. I have looked at the I'net SDR at Twente
University in the Netherlands, and I understand the concept and the
desirability of digitizing directly from the entire HF spectrum for it.
That cool receiver wouldn't work for everyone if it didn't. I
understand DSP also, or at least most of the math behind it.
I want to focus my questions on a communications receiver and its
performance however, because that's what I'm certain I don't fully
understand. In this focus, I'm attempting to receive a single fairly
narrow-band HF signal in the presence of noise and other signals close
by. Let's say a 2KHz BW signal just for discussion.
Let's say I'm digitizing at the antenna connector everything up to
30MHz. We know that I have to sample at a rate at least twice the
highest frequency being sampled so I'll stipulate that we're sampling at
120MHz, [4x 30MHz] and that there exists some sort of low-pass filter
with a cutoff frequency of 30MHz, zero attenuation below, and 1,000dB
above in front of the ADC [trying to get rid of "discussion noise" here]
My K3 reads S9 with 50uV into the 50ohm connector which is 5e-11 watts,
or -73dBm. I have a KBPF3 [or whatever it's called] and I can find
non-ham signals in the .5-30MHz range that will put the S-meter at or
close to full-scale [60 over S9]. For the sake of discussion, let's
assume the S-meter is accurate and the strongest signal I can find is 60
over S9 or -13dBm. Were the LORAN-C station at Middletown CA still
operating, it would easily exceed anything I can find at HF, on my
service monitor it generally ran around -10dBm at night with my flagpole
as the antenna.
I'll assume that the 2KHz BW signal I want to hear is S2, or 42dB below
S9 or -115dBm which I don't think is unreasonable. I would calculate
then that the digitizing mechanism directly at the antenna connector
needs to have a minimum dynamic range of 102dBm or a voltage ratio of
125,893 [rounding up]. 2^18 gives me +/-131,072 steps so an 18-bit ADC
should work. Right so far?
In my K3, the .5-30MHz spectrum is filtered by the L-C BP filters prior
to the first mixer, significantly reducing the effect of far but very
strong signals on the mixer. The 8MHz 1st IF is then filtered again by,
in my case, a 2.7KHz xtal filter, so when this all gets to the ADC, it's
3KHz wide or so.
And thus my first question [slightly rephrased]:
"Ignoring the display possibilities of digitizing at the antenna
connector, and concentrating on a communications receiver attempting to
receive a 2KHz BW signal, what are the performance advantages of moving
the A->D conversion to the antenna connector?"
It seems like it should make sense to sample right at the antenna, if
you can digitize at an adequate rate and precision, but I just can't
make it work in my mind that subjecting the ADC to the full spectrum is
good if I'm a single-op and all I'm interested in is the new mult.
Follow-up question [watched too much 'West Wing' I guess]:
Is all this "full-SDR" based on getting conversion and DSP processing
rates high enough that it doesn't matter how far strong signals removed
from the desired signal? There is inevitable distortion that arises in
the analog amplifiers, filters, and mixers, sufficiently long
word-lengths in the DSP stages could minimize that.
That's enough questions for now.
73,
Fred K6DGW
- Northern California Contest Club
- CU in the 2013 Cal QSO Party 5-6 Oct 2013
- www.cqp.org
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:Elecraft@mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
