QO-100 with IC-7300

  • Dear Ham Fellows,

    I've started to do some experiments with QO-100 using my IC-7300 transceiver. As it is my main base transceiver and I do not own the 9700 nor any VHF/UHF SSB gear, the main scope of this project was to build a transverter for this use case, yes NB. First of all I've built a downconverter to 28MHz and an upconverter using the same architecture. It quickly turned out that the upconversion is a real problem due to the image frequency on the 2.4GHz band (second mixing harmonic 28MHz away only 9-10 dB lower, so, useless...).


    Second attempt on which I'm currently working now:

    - Use the 6m band from the transceiver at 52MHz

    - Use two local oscillators; LO1 @Rx running at 687.5MHz, @Tx running at 659MHz and use LO2 @Tx only running 1689MHz. Therefore Tx is performed with dual conversion having the IF at 711MHz.

    - For 711MHz I found a very good helical filter with 3 cells and 5MHz BW. https://temwell.shoplineapp.co…ndpas-filter-tt6814a-711m

    - Results so far are good no spurs in the band of interest

    Some challenges encountered so far:

    - LO1 and LO2 are done with ADF4350 to keep low the phase noise the phase and frequency comparator (PFD frequency) runs at 12.5Mhz (of course that has the disadvantage of high 100kHz channel spacing but based on my measurements produced the lowest noise)

    - Next to implement external reference, GPS based, Leo Bodnar miniGPS (received today) clock feed references for the ADF's.


    Some problems which are unanswered and maybe you can share your experience and possible solutions for these:

    - Despite the fact that I acquired the Bullseye LNB, claiming very high stability, it is 20-25 KHz offset from the expected LNB's LO frequency. Secondly, it drifts very annoyingly and this is noticeable with FT8 reception. (no warm-up issue)

    - Okay I've found a method how the recenter to the correct frequency the LNB's LO (here: https://github.com/g4eml/Bulls…r/blob/main/Schematic.pdf) but what about drifting? I know has on output of 25MHz but to counteract at LO1 would mean to change the channel spacing to far lower which would have as consequence 20-30dB higher phase noise. How to steer this drifting?


    Thank you for reading, I highly appreciate any observation, regards and 73's by YO6OGJ, Endre

  • YO6OGJ

    Changed the title of the thread from “QO-100 with IC7-300” to “QO-100 with IC-7300”.
  • Hi


    You need to use signals; TRX: 29-29.5MHz, Generator PLL: 2371MHz

    Similar frequency signals (after a mixer):

    MIX: 2371-29: 2342MHz

    MIX: 2371+29: 2400MHz

    and generator: 2371MHz


    2371MHz attenuation: -34dB (MIXER ADE25MH)

    look pdf mixer ADE25MH: https://www.mouser.pl/datashee…0/ADE_25MH_2b-1700333.pdf


    2342MHz attenuation: filter SF2124E - 34dB (3x filter = -102dB)

    2400MHz attenuation: filter SF2124E - 2.62dB (3x filter = -7.86dB)

    2371MHz attenuation: filter SF2124E - 14dB (3x filter = -42dB + 34dB = -76dB)

    look pdf filter SF2124E: https://eu.mouser.com/datasheet/2/1100/SF2124E-3133342.pdf


    73

    Artur


  • Hello Artur,

    Congratulatulation for your work, looks very nice! Is this schematic/ layout publicly available?

    Checking the spec of the filter in deed is enogh steep to filter out the questioned products. Good point about the filter...

    Did you take measurements on PA's output, does it brings your expectations? I'm asking beacause from my calculations there are other harmonics and how bad are these:

    TxRx LNB LO1@Rx LO1@Tx Rx freq. calculated Tx freq. calculated Spurs:
    LO1-TxRx
    Spurs:
    LO1+2*TxRx
    Spurs:
    LO1+3*TxRx
    IF spurs:
    LO1-2*TxRx
    IF spurs:
    LO1-3*TxRx
    29.00 9,750.00 710.50 2,371.00 10,489.50 2,400.00 2,342.00 2,429.00 2,458.00 2,313.00 2,284.00
    29.50 9,750.00 710.50 2,371.00 10,490.00 2,400.50 2,341.50 2,430.00 2,459.50 2,312.00 2,282.50


    I see that you've used a 10MHz OCXO, I belive the PLL is on the backside of the board along with small uC. Also the keying is done with RF detector as I see, it is intended for FT817 or for some QRP rig?

    Nice work!

  • Hi


    The TRX has LPF filters at the PA output. Therefore, the 2x29MHz and 3x29MHz signals have small values. Therefore, these signals after the mixer should not be large. If you want to have attenuation of these frequencies replace one SF2124E with SF2173E filter.

    SF2173E -> 2429MHz att: -27dB

    Look: https://www.mouser.pl/datasheet/2/1100/sf2173e-1917627.pdf


    My VCOCXO is: 10MHz x 2,5(NB3N3020) =25MHz (signal for LNB)


    I use TRX the Home Made reansceiver because it has automatic frequency conversions.

    Look: https://www.youtube.com/watch?v=1CCs8Mc6Ix8

  • Thank you very much Artur! I'll have a detailed look. The LO+/-2IF, LO+/-3IF... are produced by the mixer unavoidably, that is why I was asking if you have performed measurements on the 2.4GHz output spurs. Your solution would save me from the dual conversion on Tx path.

    Thanks a lot, 73, YO6OGJ Endre