Question about FITS header RA/DEC info

Sequence Generator Plate Solving
1

Clarification question for the devs:

When SGP writes FITS header info for RA/DEC, what is the source of that data? Is it coming from the plate solver eg ASTAP or PS2, or is it coming from the mount driver (skyX, EQMOD, etc)?

Reason I ask is I’m trying to track down a problem im having in pixinsight where the image solver will not solve correctly using the RA/DEC in the raw or stacked FITS headers. I’ve had this issue for a while and generally ignore it, but I’ve got a hunch based on a post I read on cloudynights that ASTAP may be incorrectly using the RA/DEC from the initial slew and not the RA/DEC of the followup corrected slew/solve.

I do not have a permanent setup, and use EQMOD to control my skywatcher mount which I polar align each time I set up. I also use ASTAP solver. My initial slew on an object is always in the right general area, but always 5-10000 px total error off. The corrected slew after solving is generally dead on.

When i take these images into PI, image solver almost always fails. This wasnt always the case however, and I’ve only been dealing with this issue for the last 2 years or so. I also used to use platesolve2 as my solver, however I cant remember if the problem coincided with my switching to ASTAP. It’s been cloudy for months and I havent set up to test this theory. When I take the images into astrometry.net and blind solve them, they solve no problem, but the results RA/DEC is always a bit different than that in the raw images’ FITS.

Another added wrinkle is that when using a different setup (same camera, different scope and mount), in this case a paramount using the skyX ASCOM driver in SGP, and ASTAP as the plate solver, the images solve correctly. My theory here is that because the paramount is a permanent setup and uses a t-point model, its initial slew is always very close to the center of the FOV set in SGP, and therefore even if ASTAP is misbehaving somewhat, the error in the FITS header in those images is small enough that image solver in PI is still able to sort it out.

Anyway, any info on the inner workings of SGPs FITS handling of RA/DEC would be much appreciated. Thanks!

2

For headers, it always comes from the mount so it’s possible that they would deviate from reality in the case where the mount was not synced or the sync failed.

3

Ken, thanks. I suppose I will have to check and see if I have the mount driver set to ‘solve and sync’ or ‘slew and sync’ (cant remember its been a while since i configured it). That may be the issue, as I seem to recall changing it out of sync mode at some point as its not a permanent setup and the model would get messed up after a few nights and wouldnt solve…but as you say if it’s not syncing to update, it could absolutely be writing incorrect coords to the header. I’ll give it a try.