Hi John
I think you can improve your mount performance.
Here's what I see, including suggestions if you are really aiming for the best possible:
- the guiding assistant recommended improving yoru focus on the guidestar, it's worth looking into.
- a polar alignment error of 2.4arcmin can be improved. Consider drift alignment
- the GA run (unguided) wasn't long enough to really characterize the raw mount performance. If you're interested more detail here, do successful calibraiton followed by a GA run of 30 min uninterrupted. That will better characterize it
= If you had run the guiding assistant backlash test (looks like you canceled out of it) it would have told you you have enough backlash that auto backlash compensation should be enabled (and it should) consider rerunning including backlash measurement, or if you want to wag it, enable that feature and start at 400ms: it will eventually figure out the correct value. This should be relatively easy to clean up
- on the RA axis, the picture is a bit more complicated. Here is an analysis of segment 13 of PHD2_GuideLog_2026-05-12_130354.txt the using some newer tools:
This shows raw period error in fainter pink, and residual (uncorrected) period error in green. Note the three numbers correlate to the top peaks listed at the bottom. you can see the effect of how guiding reduces these values to something quite manageable. However, it's about 0.50"
Even with relatively good guiding, your primary PE is still 0.53" on its own. Consider using PPEC at 360 seconds, aggressiveness of 80 on both reactive and predictive, and disable auto adjust period.
The next two periods are harmonics of your primary: 96 seconds is 0.52" RMS and 172 seconds is 0.33". The 172sec is probably and more acurately a 2x of your primary PE, and likely related to the worm (maybe worm preload, worm bearings, etc)
the 96 second is also at 0.52" but a bit more challenging because it's further down the drive train, probably in the gearbox/belt.
I mention these because if your goal is sub-0.5", you are going to at least address the first two, and probably the need to address all three of these.
Brian
>>>