Captures of PLL LNB drifts over time. WIth & without mods.

  • Sorry for delayed answer. The LNB is a Octagon OTLSO model 1609. I use one of the F connectors to feed the RT320M chip with 25MHz from shack (for the more stable temperature and future use of a GPSDO). Forgot to eliminate the capacitors at the quartz pins. Will try today to measure the 56-th harmonic of D75F 25.000 , and compare-it with other TCXO less stable, with sdr# and a rtl-sdr stick with external 28.8MHz reference from Rohde&Schwarz generator.

  • Hello!

    After confirmation from Connor Winfield that it is an original, and the only difference between D75F 027.0 MHz and D75F 025.0 MHz is... 2MHz (HI), I did some new tests today and manage to keep the oscillator drift as low as posible, monitoring the 56 harmonic on 1,4GHz . First of all changed the capacitor between output and ground from 12pF to 15pF, insert another capacitor 10nF straight between D75F pins ground and +3,3V (allready have one on 1117 LDO) . "Bandaged" tcxo with kapton tape and insert-it inside small copper enclosure grounded with F connector as output. I am satisfied for now with the result. No more "chirp" present, no more sudden drift. Measured until now, about 500Hz in 4 hours.

    L.E. That is total drift, with rtl-sdr v3 (1ppm TXCO) tuned to 956MHz

    73 DE YO9HZN!

  • In addition to the time changes that can be made, I think the simplest is the one proposed by DF9NP, which provided a small box to have the 27 Mhz with external reference to 10 MHz. Another thing to say about the LNB and the their protection: the famous white lid. If you have the opportunity to make measurements, you will discover how disastrous these protections are. To avoid this problem, simply remove it and replace it with GoreTex, a material almost invisible to microwaves. This type of laminate is not readily available, but if you can it can be extracted from excess military fabrics.

    Modification of Octagon Oslo PLL LNB for External Frequency Reference

  • I have built a PCB to mount and filter the D75F TCXO (see twitter) using an idea from Remco on filtering the supply and with a crystal filter and buffer on the output for low phase noise. Unfortunately like several others here that TCXO does not seem to be suitable for SSB. It hunts up and down about a mean frequency. It is a lot more stable if kept in a sealed constant temperature environment - which rather defeats the point.

    Does anyone have a list of TCXOs are are or are not suitable?

    Note - personally I use a GPS reference. This is looking at options for those who don't want to spend €100 on one of those.


  • Does anyone have a list of TCXOs are are or are not suitable?

    Toyo TCXO TCO-986A1 not suitable for SSB

    See: Captures of PLL LNB drifts over time. WIth & without mods.

    Another cheap solution ist clearly the Software drift correction, no GPSDO or TCXO needed for NB SSB operation. A unmodified PLL LNB and a 0,5ppm RTL-SDR is enough.


    vy73 DB8TF

  • dg0opk I hope my post didn't offend you in any way. I never said that my source of information was your article about LNB mods. Just trying to find out why there is so much difference between TXCO from same series but not the same frequency.

    DD0KP I hope is not the case because i bought it from digi-key, part number CW654CT-ND Connor-Winfield #D75F-025.0M , paid 14.49$ + taxes...

    My intention is to help other hams not to spend money and (most important) time trying to use D75F 25MHz TCXO, like the one in the picture from my last post, with RT320M. It could be a both-side problem , TCXO+LNB, not figure-it out by now. I will try to send a e-mail to ConnorWinfield to get the whole picture.....

    73 de YO9HZN

    Mine is pretty useless too. Came from Digikey I think. Exactly the same symptoms. Putting it in an isolated thermally stable environment prevents this but I might as well use an OCXO in that case. Here is my PCB - perhaps it is not ideal but good enough I think.

  • Software Drift Correction locked on the lower EB works perfect!

    Same here. Without drift correction there is no chance of undisturbed QSO with my unmodified Octagon PLL LNB at this weather conditions (sun and clouds alternating). But with software drift correction all signals of the NB transponder are very stable. Perfect!

    At the moment my raspberry system is locked to Es'hail-1 beacon at 11199,800 MHz. This signal is very strong here.