----- Original Message -----
Sent: Friday, February 28, 2014 3:44
PM
Subject: Re: LF: SL. 47 uHz vs. 16 uHz @
8270.005 Hz
Hi Peter and all,
Am 28.02.2014 16:21,
schrieb PA1SDB, Peter:
Hello VLF.
Just a experience message !
It seems that there is somewhere a perfect setting
in "Width of FFT-bin", but where ?
A good
question.. and not easily answered. I'd say "it all depends".
In this case,
it depends on the soundcard's oscillator stability, despite stabilisation from
the GPS sync pulse.
At 16 uHz frequency we seem to have reached a limit
(for this type of oscillator drift compensation).
The next logical step
would be building an A/D converter, or modifying a USB audio device to take
the samping clock directly off a low-noise 'rock-solid reference' (a good GPS
disciplined OCXO for example).
I have fragments for such a project laying
on the workbench (a low-cost Cortex-M3 microcontroller development kit) but no
time yet to put the bits (bytes) and pieces together.
At least, one would
know of the Ionosphere (stability of the path length) or whatever else is the
limiting factor.
GL and all the best,
Wolf
.
