Glad to hear you found the software useful!

Here, strf was used by the UVSQSat-NG team :
- at the beginning of the mission to optimize the preliminary TLEs so that the university could establish the most reliable communication possible,
- for monitoring purposes since refined TLEs have become available...

Thank you for the great work!

Unfortunately the surface of the Moon is also quite hard.

It looks like not just Satdump was used as the software for this project! I recognize the spectrogram plots. I hope STRF has been useful as well.

And Magritte fans 😀

Maybe the upper stage failure ended up tumbling the spacecraft fast enough to break it up?

Thanks for the write up. One thing I've not seen discussed is why the descent stage would've separated in Earth orbit. I assume the whole probe would've needed a cruise stage for the trip to Venus of which it would have to separate. Have you come across any information that might explain that?

After fixing this mistake, the frequency difference is on the order of 1 Hz. This is about 0.7 ppb, so it is within the accuracy that we could expect for the GPSDO at Bochum. So mystery solved!

Great work! This is why we put our IQ on the web :)
One remark on why Dwingeloo is less sensitive: we have a relay in front of the first LNA, because Dwingeloo also transmits in experiments like this one. Apart from that, we may also need a better LNA.

My solution for this has been to use SPKCPO to get the state vector of the center of Venus, and then use SXFORM and MXVG to convert surface positions on Venus to offset positions and velocities in the observer frame. This works but the delay/doppler needs to be computed separately for each position.

And thanks for the discussion of the SPICE functions you used. This is something I struggled with as well as it appears there is no SPICE function that'll compute the state vector between a surface position on a target body and a surface position on an observing body.