so ive been messing around with this idea for a while now and finally started actually building it instead of just thinking about it. basically i have an old TV rotator that i repurposed for a yagi and the controller box is pretty dumb, just a pot and a motor drive, nothing fancy. i figured i could gut the controller and throw an arduino in there to handle the positioning automatically based on azimuth data from my logging software.

the plan is to use an AS5048A magnetic encoder on the mast to get real angle feedback, feed that into an uno (or maybe a nano, havent decided), and then have the arduino take serial commands from whatever is running on the PC. ive seen some people use hamlib for this and there's apparently a rotctld interface you can talk to but honestly the documentation is kind of all over the place and im not sure what the cleanest way to do it is.

has anyone actually done something like this from scratch? im comfortable enough with arduino code, done a few projects, but the hamlib integration part is new territory for me. also not sure if i should stick with the uno or move to a raspberry pi for this since then i could just run rotctld directly on the pi and skip the PC middleman entirely. open to ideas

yeah i did almost exactly this about two years ago with an old Channel Master rotator. ended up going the pi route and honestly it was the right call - running rotctld on the pi itself means you dont have to worry about the PC being on or any COM port nonsense, you just hit the pi over the network from whatever logging software you're using. WSJT-X talks to it fine, N1MM does too.

the AS5048A is a good choice for the encoder, i used a cheaper 12-bit one and it drifts a tiny bit in temperature extremes but nothing terrible. one thing i will say is put a big capacitor across the motor lines and maybe some snubber diodes because the motor noise was killing my encoder reads until i cleaned that up. took me an embarrasingly long time to figure that out. the SPI bus just goes nuts when the motor kicks on if you dont filter it.

for the software side there's a project called "Alfredo" and also ea4tx has some stuff, but honestly writing your own rotator controller code is not that bad once you understand the rotctld protocol, its pretty simple ASCII commands. good luck with it, these projects are actually really satisfying when they finally track a satellite pass correctly

curious what yagi you're running and what band, just because that changes how much positioning accuracy you actually need. for 2m EME you want to be pretty tight but for something like a 6m beam honestly even a few degrees of slop doesnt matter that much in practice.

i tried a similar thing a while back but i went a different direction and used one of those cheap BTS7960 motor driver modules instead of messing with the original control board. those things can handle way more current than you need and they're like 4 bucks on amazon. worked great until i fried one by wiring it backwards, but thats a me problem