Correcting Atmospheric Dispersion - how to stack
Stacking program, astrometric (plate) solver, fits file viewer
Brought to you by:
han59
Menu
▾
▴
Hello Han
Firstly thanks for creating a great duo of programs in HNSKY and ASTAP, and for your input to CCdCiel. I am a relative newbie to AP and your programs have allowed me to derive much pleasure and intellectual stimulation in this fascinating hobby.
I have hit a problem which covers a number of possible areas - not sure how to classify it.
I took Lights of C/2022 E3(ZTF), then applied Darks, Flats and Dark-Flats.
The resulting image showed a lot of unexpected Atmospheric Dispersion, which I want to correct.
I researched possible options on the forum including
Calibration and Alignment only - https://sourceforge.net/p/astap-program/discussion/general/thread/bbb5a7c916/?limit=25#2b3a
Making an (L)RGB image from stacks - https://sourceforge.net/p/astap-program/discussion/general/thread/8b427709bc/?limit=25#1fef
When I tried aligning the (already calibrated) images by using the first option, the program asked me to fix a reference object. I could not find out how to do that, and I was not even sure that would be the right step.
So I tried extracting the B channel from the images, in preparation for aligning RGB.
What I have ended up with is a set of apparently blue images, but with a very odd spectrum which still contains RG, and a set of other images which seem to contain RG (and L?).
What I don't seem to have is the original (L)RGB images.
Unfortunately I had ignored the basic step of keeping a backup before processing.
I attach
1 The stacked and calibrated image -
C2022 E3 (ZTF), 2023-02-13, 20x10L, LX90, ZWO CCD ASI294MC Pro_stacked.fits
OR C2022 E3 (ZTF), 2023-02-13, 20x10L, LX90, ZWO CCD ASI294MC Pro_average_stacked.
(sorry I have lost track which it is)
2 An image from the 'extracted B' set -
C-2022E3(ZTF)_20230213_204426_TB
3 An image from the 'originals after B extracted' set -
C-2022E3(ZTF)_20230213_204426.fits
Is there any way forward or back from where I have ended up, and what should I have done / should I do in the future?
I want to be able to start again from at least a set of calibrated fits files with full (L)RGB, but preferably the original uncalibrated fits files.
Because I suspect that I should align RGB before subtracting Darks.
Hoping for any advice you can give.
PS I seem to be able to post only one attachment at a time, so will follow this post with the others, four posts in total.
Tks
R
C2022 E3 (ZTF), 2023-02-13, 20x10L, LX90, ZWO CCD ASI294MC Pro_average_stacked is too large to post.
C-2022E3(ZTF)_20230213_204426_TB
C-2022E3(ZTF)_20230213_204426.fits
C-2022E3(ZTF)_20230213_204426.fits
So I can see that this thread has been viewed a number of times, but no answers.
I did a bit more digging and found this
https://sourceforge.net/p/astap-program/discussion/general/thread/51b633e3ea/?limit=25#7cc4
Which is close to what i want to do - either recombine the separated B and RG(L) images or continue to produce the R G channels and then stack.
But the thread does not finish, as cytan found another way to do what he wanted.
Following that thread, I have put the 'original minus blue' and 'blue' files into ASTAP as lights, selected 'Calibration and Alignment only', then pressed go. I am asked to select an object in each image on which to align. I double click an image which opens in the viewer, then select an object with the mouse. So far so good, I get a red circle round the object.
The thread says that the file name should turn green. It doesn't. I tried to save the file, and am asked to confirm whether I should overwrite.
So far I have not ticked yes.
Am I doing the right thing so far, and should I continue by ticking yes?
Tks
R
Sorry for the late response. I did not get automatic emails from new positing anymore.
The dark/flats should be applied before any alignment.
Dispersion is only a few arc seconds for an elevation of 15 degrees. You images where taken at 52 degrees elevation/altitude it should be nihil. It must be a stack problem.
Lets try to fix the normal stacking. Is the option "colour smooth" in tab "stack method"check marked?
Can you share some of the original files by a link?
Han
Hi Han and thanks for the reply - hopefully this post will reach you in a timely fashion.
I suspect there is AD despite the latitude. As far as I recall, I did not use 'colour smooth' - and when I open ASTAP just now the box is not ticked ..
Here is a dropbox link to my 'original' files https://www.dropbox.com/sh/x0aa5iiyuokk9it/AAAw2IRHMaChGDhZsIvbBBPla?dl=0
As I mentioned in the OP, I processed my original files without keeping backups - too enthusiastic!
In the link you will find a folder called 'ASTAP'
In that is a folder called LIGHTS-COPY-AFTER BLUE 'EXTRACT'. This contains two sets of images. Both sets are lights with darks and dark-flats applied by ASTAP.
I had the blue balance set too high.
The 'blue' folder contains an output of some function within ASTAP= I'm afraid I have totally forgotten what I did to get these images. The goal was to separate channels, correct intensity, and then re-stack,, but as per previous posts I am not at all sure that would be the right way to do it.
The 'RG' folder contains the files after the 'blue' function was applied. I don't know whether these are supposed to contain RGB, or just RG, but as far as I can see they contain RGB.
I have now corrected the gain values in the ZWO driver, and have added two additional image sets (1) the Moon and (2) M42 collected with the new channel balance settings but without any calibration. The Moon also contains some AD in my view.
A caveat to all images is that although I use an SCT. so there should be no CA, it is 20yrs old, and the focus, collimation and sensor orthogonality are possible sources of distortion.
Interested to see what you find. Please ignore the 'extras' if you like - they are only there to see if in fact there is any AD in the unstacked images. I appreciate it may be difficult to tell, especially with M42 which is uncalibrated.
Best, R
Hi Richard.
I did read your post. I will have a look tomorrow. Today I was very busy with other things.
Cheers, Han
Yes there is a weird shift between red and blue. It is in the whole image and under angle of maybe 40 degrees.
It is also visible in the Moon craters.
I have no explanation. If it happens again, I would just go through the typical checks like collimation. Maybe visual check the stars through an eyepiece.
Is there are correcting lens in the SCT?
By the way, do you get any email from SourceForge? I still don't get any emails about new postings on the forum.
Han
Thanks for taking a look.
There is no correcting lens in the SCT. I may try running the Moon images through PIPP's RGB align function, though there is something similar in GIMP which I am learning slowly.
I did get a notification from SF for your post, because I had clicked on 'follow topic' or whatever it is called. It is weird that you don't, considering that it is your Forum!
Perhaps they think you would be overwhelmed!
Did you discover whether the RG comet images contain enough information to obtain another stack? Or would I have to align with the B images?
Best
R
I didn't try to combine separated colours. OSC images should work straight out of the box. This shift is abnormal. It is better to find the cause then to try to compensate it.
What happen after a meridian flip? Is blue and red shift still the same.
You could try to image some faraway trees of houses and see how that turns out.
Tks for the ideas - no clear skies for a week or more so will try on a long terrestrial view - I may be able to get some streetlights ~1/2km distant, or a daylight shot to trees >1km. We'll see. Can also try with the scope flipped - good idea.
My feeling is still that this is AD. Will let you know how the tests work out.
Best R