Re: SL settings for QRSS (was: Re: LF: RE: Todmorden receiver by Jack)

[Andy Talbot <andy.g4jnt@gmail.com>]

Someone, somewhere, must have designed a soundcard, or an A/D that looks to a PC like a soundcard to allow easy integration with much of this analysis software.   It really shouldn't be necessary to have to feed 1PPS references into the parallel stereo channel and do complicated software locking.

I did wonder if a PCM2900 USB soundcard chip could be made to do that, by clocking it with a locked 12MHz input.   But someone said it would suffer from USB packet slippage so I never persued the idea.  

I can transfer perfectly timed I/Q data , albeit from a slower 1kHz sampling rate, over a 115200 baud RS232 link implemented with USB COM ports, and it must be possible to do it faster.

In fact, USB RS422 COM ports using the FTDI232 device can be made to ackle at 3MBit/s - I've tested it, it works -   so transferring mono 16 bit data could allow 48kHz sampling along with some framing overhead and timing info using off the shelf hardware.

Andy  G4JNT

Virus-free. www.avg.com

On 6 January 2018 at 18:42, Wolfgang Büscher <dl4yhf@freenet.de> wrote:

Hello Jack and all,

you wrote:

>It comes back to using something like SL with its QRSS
capabilities. I wish I had a good file that worked at 8720kHz
for QRSS600 as an example. This would be a good start if nothing else.
<

It's not that difficult to set up. The key point in SL to get into the microhertz-resolution is the decimation (of the sampling rate), along with complex frequency to have a complex FFT spanning just a few Hz around the "frequency of interest". The effect is similar as a super-long FFT (concerning the frequency bin width, and thus the "gain"). I think on Stefan (DK7FC's) site you can find some configuration files for QRSSS.

Back in the "old days" of extremely slow Morse, I wrote this short how-to (for Spectrum Lab):
http://www.qsl.net/dl4yhf/speclab/qrss_quickstart.htm
  (it shows how to pick one of the VLF transmitters to stabilize the soundcard's sampling rate,
  a bit out of fashion since good GPSes with sync output became available, but anyway..)

To make the FFT window even longer / frequency bins even shorter than used in the default settings,
use the option 'complex input for the FFT with internal frequency shift', as shown at
http://www.qsl.net/dl4yhf/speclab/settings.htm#fft_size

And, last not least, there an explanation of FFT length, decimation, windowing functions, resulting resolution / effective receiver bandwidth versus dynamic range, etc:
http://www.qsl.net/dl4yhf/speclab/fftinfo.htm

All the best,
 Wolf DL4YHF