so i've been doing FM sats for about two years now, mostly SO-50 and the TEVEL birds, and i always kind of ignored the linear transponder ones because they seemed way more complicated to get into. finally sat down last weekend and actually figured out AO-73 and i have to say i feel like an idiot for waiting this long.

the doppler correction stuff is less scary than i thought once you just watch the waterfall and get a feel for it. i'm running a pair of yagis on an az/el mount i built out of schedule 40 pipe and some surplus rotators, nothing fancy, and SDR# on the downlink with a FT-818 barefoot on the uplink. the tricky part for me was actually finding my own signal in the passband at first — i kept hunting around and then realizing i was way off. took me probably three passes before i heard myself clearly.

anyone else using gpredict for tracking and finding it accurate enough? i'm getting maybe a second or two of timing offset from the actual AOS compared to what it predicts, which seems fine but curious if others tweak their TLEs more frequently than i do. i pull new elements every couple days from celestrak.

yeah gpredict is pretty solid honestly, i've been using it for a few years and updating TLEs every 3-4 days is usually enough unless the bird has done a maneuver or something, which obviously doesnt apply to the amateur ones. the second or two offset you're seeing is probably just normal TLE age drift, nothing to worry about. if you start seeing consistent 5+ second errors that's when i'd look at your system clock sync more than the elements themselves.

the linear transponder mode thing really does open up a lot more operating. AO-73 is nice but if you havent tried XW-2 series or even FO-29 on a good day those are worth chasing. FO-29 is getting old and a bit unreliable but when it cooperates the passband is clean. also try running your uplink power down as low as you can while still copying yourself — the polite thing on a linear bird is to not stomp all over the passband and it'll actually help you work more stations too because they can hear you without you wiping out half the transponder.

the doppler thing messed me up for way longer than it should have. i was manually tuning which is fine but i kept overcorrecting and chasing my own signal around. eventually set up a rig control loop with gpredict and that helped a lot even though my radio's CAT implementation is a little janky. what rotators did you end up with on that pipe mount if you dont mind me asking, im trying to build something similar on the cheap