I'm encountering a problem when I do a meridian flip with my telescope. After the meridian flip the location of the CCD FOV is not centred in the same location. This happens even after alignment when plate solving says the error is < 30".
It seems to me that the point to which plate solving syncs the telescope is not actually the centre of the CCD FOV. Thus when the image is rotated during a meridian flip the two FOVs don't overlap.
I have attached 2 pictures showing the problem below. Might anyone know what could be happening?
I am using an explore scientific iEXOS-100 as the mount.
My first reflex was to think of the pointing issue when the target location has no associated target. But I had such an observation yesterday night (with latest kstars/indi), and the meridian flip worked well (some 20px offset = 17 arc sec)....
I think it‘s more complex. My current suspicion is, that the first real position the slew reaches creates the problem and that the sync either does not happen (bug in kstars) or has no effect (bug in mount driver or below). But as long as I cannot reproduce it, I can’t investigate the problem.
Apologies to all for the long delay, today was the first chance I had to test the issue again.
I created an imaging session for the star Vega which also underwent an automatic meridian flip. Whilst it looks quite small, the star is about 150 pixels off where it should be, which with my setup corresponds to about 10'. However the plate solving says the frame is within the 45'' tolerance that I set in the settings, which is clearly not true. I am using the latest version of Kstars and the latest firmware for the mount (iExos-100). I have attached logs from the session.
Thank you very much once again
Last edit: 4 years 1 month ago by Giulio Crisanti. Reason: format fix