Alignment issues from tilted coma corrector
Stacking program, astrometric (plate) solver, fits file viewer
Brought to you by:
han59
Menu
▾
▴
2 weeks ago I imaged M16 Eagle Nebula and captured Ha , Oiii and Sii data ( about 4 hours each ) over 5 nights with my 8" f5 carbon Newt , ZWO 2600MM camera and Antlia 3nm filters.
I did a stupid thing and retightened the coma corrector on one night which obviously caused some minor tilt ( at the time I was oblivious to the fact that this would cause tilt and ultimately double stars )
I stacked and aligned the individual data sets in ASTAP and then loaded into the Compose module in Startools.
After the first AutoDev stretch (image nspection ) I noticed I had double stars over most of the LHS of the frame . The stars in the RHS of the frame where not affected too much ( slightly oval ) and could be repaired using Startools Repair module.
I then checked each individual stack Ha , Oiii and Sii in Startools and the stars were pin point to edge of frame on all 3 filtered sets.
To salvage the image and not waist 16 hours of good data , I used the Heal module in Startools to remove and dim out the majority of the smaller faint to medium stars and just left the larger stars which could be corrected as best as possible ( a time consuming task )
Q: Is there any setting in ASTAP to nominate a master stacked image ( say Ha ) and then use that model to align the Oiii and Sii stacks. I could not find anything.
I just thought I would ask ?
Im not interested in AI correction.
Pi probably has a tool to use to correct misaligned stacked filtered data sets ?
I guess the moral to the story is once set up correctly , dont fiddle with anything !!
Cheers
Martin
Have images of a misaligned system is frustrating.
I understand you have now an align problem or a preference for using the Ha as the reference image.
To force one image as reference I think if you can select the stack option "calibration & align" then the first image in the list will be used a reference image. So if you first load select Ha in then the rest it will aligned to it resulting in files ending with _aligned. Calibration will be skipped if the images are already calibrated of darks and flats are missing.
Is this what you want?
Cheers, Han
Hi Han,
Ive been successfully stacking and aligning for years
This issue relates to one or even 2 of the stacked files ( I suspect its the Oiii ) my coma correctors plane shifted as a result of me tweaking the focuser thumbscrews just prior to a session which resulted in the star field being way out of alignment compared to the other 2 stacked filter sets.
When I use Calibration and Alignment with all 3 loaded ( Ha , Oiii and Sii ) the resultant aligned images with file _ aligned have double stars ( mostly the faint ones on the LHS of the image )
I suspect the coma correctors tilt has shifted the star field across the resultant stacked images. The individual Ha , Oiii and Sii stacked ( non aligned images ) have perfect pin point stars to edge of frame . The anomaly occurs after alignment as ASTAP probably cannot align a severley affected starfield caused by a coma corrector tilt.
See snapshot photo attached
This is the first time ive ever fiddled with the focuser in between filter sets.
My focuser is a stock GSO not an expensive Moonlite or other high end model
Does this describe my issue now ?
Martin
That is weird because no files are stacked during the alignment process. Are you sure there is a light included in the master dark or flat? Check the master dark and master flat.
If the double stars only occur after stacking then it is an alignment problem. Shifted and even images with a different FOV should be no problem for the alignment routine.
Can you share all image image and a master dark and flat so I can have a look myself?
Han
Han,
Apologies, I should have said " When I use my final process, Calibration and Alignment with all 3 ( Ha , Oiii and Sii ) stacked files and then load these aligned stacked fts files into Startools Compose , after an auto stretch I see double stars mostly across the LHS of the image"
I have noticed that the double stars are not the same size , one is slightly smaller than the other which can only mean that one star is say Ha and the other star is say Oiii. Also in Startools when i hit the colour module and process colour , the double stars are different colours across the whole image , one is red and one is blue.
I'm confident the coma corrector tilt has stretched the image in the corners by say 8 to 15 pixels due to the designed optics of the coma corrector which has caused the stars on one of the filter sets to have different stellar distances beween across the image from L to R . If you look at the image I sent of M16 and zoom in , the large star towards the centre has smeared or wider than normal diffraction spikes which looks like ASTAP has tried to align this star as best as possible but
left the stars spikes smeared a bit. This is not a shifted or different sized FOV but a stretched FOV on one or two of the filter sets caused by the coma corrector.
Maybe zoom in and around the attached M16 image to see what im talking about.
Im not sure if ASTAP's Alignment engine can correct Astigmatism between filter data sets ( either one or two filter sets out of three ) caused by a tilted TS GPU coma corrector ?
I could be totally wrong , so I will try to send you some Ha , Oiii and Sii lights plus a master dark and master Flats in Ha , Oiii and Sii for you to analyse
Thanks Han
Martin
Last edit: Martin Dowd 2025-07-01
Han,
Below is a link to the following files as follows -
95 x Ha Lights
92 x Oiii Lights
96 x Sii Lights
1 x Master Dark
1 x Master Flat Ha
1 x Master Flat Oiii
1 x Master Flat Sii
https://ufile.io/f/na63n
See if you come up with the same issue as me with the smaller faint double stars after composing an SHO linear stretch.
Thanks
Martin
If I measure the distortion I can see they are different distorted. The only thing what could help is to stack them using the SIP coefficients for correction. This is currently disabled because it slows down and makes no sense if all images have the same optical distortion. But I assume it will never be perfect.
Okay the stack of the individual filters looks good. I tried to merge them with the "image stitching method". This method is using the SIP coeficients but the result is not good. So the distortion is not to compensate with the 3th order SIP coefficients.
I tried it with the Mictrosoft image Composite Editor. No success
I think the only thing you could do is to put the three interim result in something like Photoshop or maybe Gimp in 3 layers and stretch the image layers such till the stars overlap.
Han
Han,
Thanks for testing and analysing
Much appreciated
I did eventually fashion an SHO image by removing and dimmingd stars
These TS GPU coma correctors are excellent ( best ive ever used including the TV Parracor ) but introduce tilt and they can ruin your work if using multiple filters with a mono set up. OSC not so bad .
The moral to the story is once set , dont fiddle
Thanks
Martin