[time-nuts] Quadrature Phase Noise Measurement

[John Miles]

> I have searched the posts on this but am still unsure of how to transfer
the
> data to PN.exe. I can find no native interface for ADC samples.

Yes, that's the dreaded "exercise left to the student." :)  Currently there
aren't any good plug-and-play solutions for baseband acquisition in either
PN.EXE or TimeLab.  When I first started to look into PN measurement, I
planned to add support to PN.EXE (and later TimeLab) for quite a few
baseband digitizer platforms.  But as it turned out, I'm only just now
getting around to that area.

For what you're doing, you might check out Andrew Holme's experimental
direct-digital project at http://www.aholme.co.uk/PhaseNoise/Main.htm .
Andrew's digitizer isn't intended for use at baseband, but it will probably
be easier to adapt his HDL and C code for use with a mixer and LNA than it
will be to write something from scratch.  

Also consider a sound card with 192 kHz capability.  It would be a good idea
to look for a 192-kHz sound card whose input blocking capacitor can be
removed easily.  

Another option might be the Digilent Analog Discovery 2, which can be used
as a convenient open platform (or at least semi-open platform) for LF
applications.  Not too familiar with it myself, but I know it's become
popular in audio circles.  

The next step up from that would probably be the Red Pitaya.  Marco Gottardo
has a book called (literally) "FPGA to High Speed ADC Data Streaming," which
explores various data acquisition applications with the Red Pitaya including
baseband noise measurement.  I have a copy of this one that I haven't had
time to look over in depth.  The English is a bit rough in places, but the
content seems worthwhile for anyone contemplating a Zynq-based digitizer.

None of these options will bring data directly into PN.EXE, of course.  But
that's the easy part of the exercise compared to the general problem of data
acquisition.

-- john, KE5FX