[Lowfer] analog vs sdr vlf/lf comparison

[Douglas D. Williams]

On Mon, Nov 26, 2012 at 12:58 PM, JD <listread at lwca.org> wrote:

> Excellent report. Doug, with most interesting results.
>
> However, if I might point out one thing that seems to be a discrepancy...
> if
> I understand your setup correctly, the frequency stability comparison may
> not be entirely fair.  I believe you're comparing drift at 60 kHz in the
> Excalibur Pro against drift at 3060 kHz in the R-75.
>
> Now, that is valid in one sense, because it's the way many people would
> actually be using the two units in the real world.  But so far as checking
> the inherent stability of the two pieces of hardware against each other,
> it's a bit lopsided.  Sixty versus 3060 is rather like comparing apples to
> pumpkins...really big, Guiness record setting pumpkins. ;)
>
> (Practical example: the 100 millihertz drift I experienced during the
> capture of WMS last night, while warming the car interior back up from 45
> degrees F to about 75 degrees, was with the R-5000 tuned directly to 185.3
> kHz.  If I had been going through the upconverter Todd built for me, the
> change would have been just under 2 Hz even if the OCXO held the
> converter's
> frequency absolutely steady.)
>
> It might be useful to see what would happen if you put the splitter after
> the upconverter, and compared both receivers tuned to the same dial
> setting.
>


John, it took me a while to understand the differences that you are talking
about here, but I finally did (I think) understand them a s couple of
months ago. You are correct (in a way) in suggesting that my frequency
stability test was unfair. Whenever we choose to use an upconverter for
VLF/LF, we will, necessarily, multiply any frequency instability by
the frequency+drift inherent in the upconverter. I feel sure that if I
tested the (OCXO) modified R75 without the upconverter at WWVB's 60 kHz
frequency, I would record less (much less) frequency drift than the R75 +
the upconverter. However, as you and I both know, most analog shortwave
receivers (including the R75) are inadequate in VLF/LF reception sensivity,
as well as introducing numerous spurs into the VLF/LF spectrum. Long story
short.....most analog receiver manufacturers don't care about VLF/LF, and
thus we are left with the choice of dealing with their shortcomings, or
using a good quality upconverter.

But, as you mentioned, "real world" results are all that really matter, so
the only really "fair" test is to test the "entire receive systems" against
each other, both in sensivitity, stability, and IMD. Unfortunately, I don't
have sophisticated test equipment, or the time to devote to such things, so
the only thing I can offer is actual receive comparisons.


>
> By the way, I really like those two captures.  It's the first time I've
> actually seen the WWVB "station ID" this way (the blips from the +45 degree
> and -45 degree carrier phase shifts).
>

Heh, glad you like them. IMO, what better station to use for receiver
VLF/LF stability tests than WWVB?


>
> Another possible experiment suggests itself, if you haven't already done
> it:
> I'm making the assumption that the signal from the Clifton Labs antenna
> currently goes to the splitter, and then one output is through the preamp
> and/or attenuator into the Excalibur... while the other output is to an
> LPF,
> then the upconverter, then the ICOM.  I didn't notice mention of an LPF in
> your post, but I realize it's a necessity with this upconverter.  Anyway,
> what I'm wondering is, have you tried putting the LPF between the antenna
> and the splitter, so that the inputs to both receivers would be filtered?
> I'm curious whether keeping out-of-band signals out of the Excalibur Pro
> would reduce the risk of ADC clipping and enable you to run it with less
> attenuation, and thereby maybe give it an edge in sensitivity for detecting
> weak signals at LF.  Just a thought.
>
> John
>

Good point, and I see that I neglected to mention earlier that my receive
chain is as follows:

(1) Clifton Labs active antenna to...
(2) KIWA 500 kHz low pass filter to.....
(3) KIWA custom splitter to....
(4) AMRAD Converter + R75 and the Winradio Excalibur Pro

It turns out that the preamp is unnecessary. The Clifton Labs active
antenna provides plenty of signal.