Posts by SV1BDS


    The method I use is to measure the frequency offset that provides the Costas Loop Block of GNURADIO. But it needs proper scailing in order to measure Hz. I can measure at the same time the 750 QPSK beacon and if I want I can Tx at a free frequency and measure also mine. I Tx also BPSK similar to 750 signal but with random data. The values are compared to Doppler calculations. In previous tests Doppler calculations was outside GNURADIO. Now can included in it if desired.

    2E0ILY How many Hz is away the beacons? From my measurements a range of 30 Hz is due to Doppler effect and a few decades Hz may be exists due to QO100 Lo accuracy. Totally about 100 Hz max is observed recently. If your reception is more maybe LNB is not locked.

    In order to have as good accuracy as possible, a R trimmer must be used to set it exactly at 10 MHz. The R trimmer must be a multi turn one and be powered from a very good (without noise) source that provides 4-5 V. I want to make such a no GPSDO module and I plan to use an existing 4.096 V reference.


    Remco PA3FYM , some tests occurred today. I try a 100 Mhz Wenzel OCXO driving a 74HC74 /4 divider producing 25 MHz. This was cleared with 2 xtals. It was driving via a variable attenuator an SBB5089 MMIC amplifier with maximum output 20dBm enough to drive the twin Octagon LNB. A 0.3 to 0.4 MER increase was noticed. The performance was very near to an unmodified similar LNB. It seems possible to create a Lars GPSDO using 100 MHz OCXO. A /5 division at first /2 (50mhz) will produce 10 MHz.


    Maybe these values are possible as I am more near to the center of the satellite beam. Stations for example in Italy can confirm that. Always values are taken when NO other activity exists on WB. I use minitiune and DATV express receiver. On NB for example me and SV8QG can have max 40 dB SNR on CW beacons on SDR console using about 90cm dish sizes or slightly more.

    Update: all these values are with out POTY !!!


    Some more information:

    With 90cm dish an unmodified Octagon twin LNB gives about 10.7. A same one modified for GPSDO gives about 10. In my plans are to try a lower phase noise source of 25 MHz to see if the MER increase.


    I have an excellent pass at Fri 7/5 @ 7:13 UTC (10:13 Local).

    I will try to receive it using a 90 cm dish, a Octagon twin LNB with GPSDO to a Pluto with GPSDO. A GNURadio flowchart will be used to capture also IQ data for future decode of the signal. Now I try to write a Python script to calculate Doppler.


    If you prefer a band less than 70cm image rejection will be less than required (40 dB). This depends from the BPF of your upconverter. So for 2m it will be 2*144 MHz less aka 2400-2*144, relative near to the desired frequency.


    For driving a Pluto 10 MHz can be directly feed to it. There is no need to use 40 MHz. I use 10 MHz. Commands must be like:

    fw_setenv ad936x_ext_refclk "<25000000>"

    fw_setenv xo_correction 25000000

    pluto_reboot reset


    A setup was tested permiting two independent QSO at any frequency in the NB transponder using a single transmitter and one LNB. In practice it could be possible to be one in the NB and the other to the WB. The setup was with the IC735 and a Pluto. IMD products was not noticed. A typical usage could be FT8 & USB simultaneous QSOs. Typically using only one SDR any number of indipentet transceivers can operate as enough output RF power and CPU exists.


    This module depends on the hidapi module, which can be installed with the following:

    sudo apt-get install python-dev libusb-1.0-0-dev libudev-dev
    pip install --upgrade setuptools
    pip install hidapi

    udev rule

    In linux you may need to create a udev rule in order to use this without sudo privileges.

    For example, edit /etc/udev/rules.d/99-usbrelay.rules and add this line:

    SUBSYSTEM=="usb", ATTR{idVendor}=="16c0", ATTR{idProduct}=="05df", MODE="777"

    Then restart your computer or run this command:

    sudo udevadm control --reload-rule