so ive been messing around with AMSAT stuff for a few months now and i finally got a decent pass with AO-73 yesterday using my Arrow antenna and an IC-9700. got audio out of it which was exciting but i could not figure out how to stay on frequency to save my life. i read about doppler shift being like 10khz total across the pass but when i actually tried to tune for it manually i kept losing my own downlink signal.

my question is basically — do most people just let the tracking software handle the VFO correction automatically or is there some trick to doing it by ear? i was running gpredict linked to the radio via hamlib and it was doing something but honestly i couldnt tell if it was actually moving the frequency or if i had the port config wrong. the pass was only like 8 minutes and i spent the first 4 fumbling with everything.

also the inverting transponder thing caught me off guard. i knew about it intellectually but then in the moment i went the wrong direction tuning and that cost me another minute. any advice from people who have actually done this consistently would be really helpful

yeah gpredict with hamlib can be a bit finicky to set up honestly. the doppler correction does work once you have the right device string and baud rate configured but its not always obvious if its actually talking to the radio or just silently failing. what i usually do is watch the VFO on the 9700 screen during AOS and if the frequency isnt creeping you know somethings broken in the connection.

for the inverting transponder thing — that trips everyone up the first few times. just remember if you hear yourself going up in pitch you need to tune your uplink down, not up. after a few passes it becomes muscle memory but yeah in real time its counterintuitive. AO-73 is a good one to practice on, the transponder isnt super crowded compared to SO-50 or whatever. give yourself a few more passes before expecting to actually make a QSO, just finding your own downlink first is the real skill

i had the exact same problem when i started. honestly what helped me was doing a few passes where i didnt even try to work anyone, just tuned around the passband and listened to figure out how the doppler felt in real time. the math makes sense on paper but your hands dont know what to do until youve done it a bunch of times. also make sure gpredict has a fresh keplerian data set — if your TLEs are more than a few days old the pass timing and peak elevation can be off enough to mess with your doppler prediction timing too