Logs from last night:
https://openphdguiding.org/logs/dl/PHD2_logs_Trnf.zip
My mount has been very well behaved until last, typically giving a total rms error of between .4 to .3".
Last night I was seeing a 'heartbeat' of approximately 3 to 4 oscillations that would ramp up to approx 2"+ then disappear for what appears to possibly have been around 2 worm cycles and then repeat. The DEC and RA (relative to the PHD2 graph) moved together in opposite directions during these events which might indicate a return of the dreaded Paramount MyT mechanical crosstalk.
I had started out with a new (longer fl) guide scope that had a clamp style mount rather than rings, but the total RMS has grown to 1.21" within about 4 repeats of the cycle. Since it was new and the gs was a new installation, I went back to the previous gs setup to eliminate that as a cause since that had not been occurring.
The only other thing that changed is this is my first session in with this mount with ambient at freezing. Seeing appeard to be good, so I increased the guide frequency to 2 secs, increased RA aggr to 60 (had been 55), and reduced minmov on the RA and DEC to .17 and .19 respectively. I didn't want to do too much more aggresive tuning for fear of destablizing the periods of more or less straight line tracking/guiding.
That seemed to help mitigate/reduce the magnitude of the heartbeat oscillation, but did not eliminate it.
I haven't looked at the logs yet, having just pulled them from the mount this AM. I'd appreciate a review/analysis of logs to see what you think may be happening.
A couple of additional points.
Three calibration runs last night. The first two were with the new guide scope - very evident orthogonal issues on the DEC axis on the first run. The second run improved but appears to have a slight S curve. The third calibration was when I put the original guide scope back on - it appeared to have no orthogonal issues, at least visually.