After continued failures of autoguiding that all seemed to be cable- or connector- related, I tried a different approach.

My hypothesis was that the setup I had was fundamentally error-prone. The autoguider signal ran from the relay output on the camera to the guide port on the mount. This means that cable was in constant motion. The telephone jack type connectors used on both ends of that cable are unsuited for this – they don’t provide a positive lock, aren’t meant to be used in extreme temperatures, and certainly aren’t meant to resist motion.

So, I changed guider drivers. Instead of using the relays built in to the camera, I am using a GPUSB unit from ShoeString Astronomy. This with this setup, the small GPUSB control box is fastened permanently to the mount, and a short cable runs from the control box to the mount. Same connector, but no motion, so I hope it will remain more stable. I also thoroughly cleaned with mount-side connector with a small abrasive brush while changing the cable.

I originally thought the cable from the camera would be better since it meant one less wire running from the computer. However, that problem was solved by having only a single USB cable from the computer, running to a multi-port USB hub that is mounted in the space atop the pier. So adding the GPUSB device only mean an extra cable inside the pier platform, from the USB hub, not another cable to the computer.

Initial results: it seems to be working. I’m getting guiding on both axes again. Now I’ll need to spend some time calibrating backlash, aggressiveness, etc., again, because I messed those settings up with the previous guiding system, trying to make it work. I think this is going to be the winning approach, though.