Gnuradio Problems with Reliably setting Gain for AIRSPY Mini

[Glen Langston]

Hello

I’ve been an Airspy user (player/researcher) for years
but in the last few months have been having trouble
with setting the GAIN for an AIRSPY mini 
using GnuRadio’s OSMOSDR interface.

I’m seeing the gain being erratically set.  Sometimes
the gain appears high, but not not seeing gaussian
distributed noise, and then sometimes low, not looking

like the gain is set on.   This is running the same python

code multiple times, with different apparent power levels

after initialization.

Is anyone else seeing this and is there a work a round?

This was not the usual behavior in previous years.

I’m wondering if there is a communications error
and/o packing/unpacking is not working correctly.

Note that very rarely I do get good gain values
and can confirm proper operations, but mostly not.

Two screen captures are attached, taken a few seconds
after each other, after starting and stopping the python
design in Gnuradio.

Again this was not a problem a while ago.   But after
days of testing I can not get this to clear up.

Any suggestions would be greatly appreciated.

Thanks!

Glen

Our Citizen Science Radio Astronomy designs are
at

http://www.github.com/WVURAIL/gr-radio_astro

if you’d like to test this yourself.

The OSMOSDR code has some use of the “sensitivity” and “linearity” arguments,
which I added for this example, but these do not seem to change anything.