Autocenter complete at line 10087, 6/8/2016 12:26:41 AM
The log says the autocenter error was -35.2px in RA, yet the frames post-flip are offset in RA by ~ 150 pixels. Attached is a screenshot of the last image preflip, and first image postflip, as an example. The red + indicates the same RA/Dec in both images :
As the log file shows, plate solving “worked”…it seems to have simply solved either to the wrong coordinates, or calculated the amount of error incorrectly, though the scale and image size being reported by PS2 in lines 10069-10071 are also correct, so I’s be surprised to learn the math was wrong.
Haven’t had this happen before, have been using PS2 on 3-4 previous images, flips went fine, so…pretty confused on this one.
Hm, that’s really interesting. The only thing I can think of is that something happened with guiding. After that sub it takes about a minute to get your guiding back under 1px. So maybe it locked onto a bright pixel and was drifting for a bit? May be worth checking out the PHD logs to see if it was issuing a correction in that RA direction for a bit.
Good call, Jared…you were on the right track I think.
Looks like the last center slew is approaching from west to east, and the worm backlash is taking up enough time that I’m getting about 10" of drift after SGP reports it’s centered. Combine that with what’s happening DURING centering that PS2 is able to solve against anyway, and that probably accounts for it.
Now to sort out why it hasn’t happened before.
Always something, isn’t it?
Thanks for the insight!
Oh God, I am SUCH an idiot. Took all of 3 minutes to find the problem.
I replaced my imaging laptop a while back…aaaaaaaaaaaaand the new profile has a mount settling time of 0.0s
This place is in serious need of a facepalming smiley right now…
