FWIW, I have been using an RST-135 ever since the first production run from RainbowAstro, but have not run into this type of periodic (but not related to the gears' periods).
Granted, I have different power supplies for my instrumentation and my mount. The instrumentation is supplied by a 13 volt source (yes, not 12) and the mount is supplied from a homebrew DC uninterruptable power supply that provides 15V to the mount alone (the two grounds are strapped with heavy gauge bus).
If you have effectively turned off guiding, but monitoring it, and still see a 1100 second period, the problem is on the mount's end. I.e., the error is not likely caused by PHD2, but PHD2 is simply reacting to something causing the 1100 second anomaly.
I would try to power the mount by a large battery source (this will make sure the mount is not affected by whatever is coming through the power lines), but making sure the ground return of the mount power is connected to the ground return of the 12V supply for the rest of the instrumentation. Some people have multiple supplies but don't strap the ground returns, causing the USB ground pin to carry a lot of current -- don't do that for the health of the electronics. If that makes the problem go away, the problem is the power to the mount's end.
I use 15V for my mount because the RST-135 is rated for 11V to 16V operation, and their R&D manager (xuranus at Cloudy Nights, real name B.J. Jeong; who is also R&D manager at the parent company, RainbowRobotics) has confirmed that the mount works best at the higher voltage end.
Chen
