Calibration and downconversion

[PA1EJO]

I think there is a problem with applying a frequency calibration correction with the SDR receiver (a Zyncsdr / Libresdr) listening to the 70cm band (433.5 to 434 MHz) while the signal comes from a QO100 downconverter running between 10489.500 MHz and 10490.000 MHz. The required frequency shift is 10056 MHz, and this is implemented in the feed of the offset dish, and it is I specified in the downconverter section. When I start the SDR with the downconverter shift being selected then the frequency calibration correction does something strange, it applies the PPM correction to the 10 GHz signal which is not what I want, because, the error in the frequency is caused by the local oscillator within the SDR and not in the downconverter which is frequency locked to an external oscillator. You could call it a bug in SDR console, calibration corrections should be applied to the local oscillator in the SDR, the current implementation in SDR console suggests that we can control something beyond the scope of the SDR, and this is of course not the case.

[PA1EJO]

Checked it again, what I need to do is to first turn the downconverter off, go to the 10 MHz, do the manual calibration with a GNSS DO reference frequency, (it yields 0.23 ppm, pretty good actually), stop the SDR, go to select radio, turn the downconverter section on, and you see that the PPM correction in SDR console is correctly applied. So the SDR console software is confused when you change the calibration setting with a downconverter selected to on.

[PA2JSA]

I am using also the ZyncSDR / LibreSDR now for a couple of weeks, with SDR Console v3.1 as well as with SDR Console v3.4. I notice in Sdr Console v3.1 as well as v3.4 a spike in the Spectrum from out of the ZyncSDR, that spike is slowly going down within 5 to 10 seconds and than disappears. Afterwards this spike stay away. The spike is also visible in the spectrum without any transmission. It looks like a calibration spike or something like that. I even so noticed that the external 10 Mhz clock of the ZyncSDR is not supported by the modified Pluto firmware v0.37 and 0.38. TX and RX are working fine over the QO-100, however the internal 40 Mhz VCTCXO is not on frequency and drifts a little bit, due to the fact that providing the 10 Mhz External clock signal of the ZyncSDR is not supported. From the hardware diagrams I noticed that the ZyncSDR is comparing the external 10 Mhz frequency with the internal 40 Mhz VCTCXO in the FPGA and from there a DAC is controlling the 40 Mhz VCTCXO. the output of the 40 Mhz VCTCXO is split in a 40 Mhz signal going to the AD9363 as well as the FPGA. Support of the 10 Mhz external clock in the ZyncSDR is essential to use the ZyncSDR for QO-100 usage. So probably Simon when you test your ZyncSDR - LibreSDR, please have a look how support for the 10 Mhz external clock can be realized.