DL folks - I seem to be having some issues with certain SBIG filters, where I get halos around bright stars that I can't figure out how to remove. These seem to be the worst with the Blue and Green SBIG 50mm square filters. I've provided four integrated images below. One is a zoomed in area of an RGB integrated master showing a great example of the problem, and the other three images are a full RGB integrated master, the green integrated master and the blue integrated master. These are apparent in single subs when you look closely, then become extremely troublesome in processing. Everything is focused and collimated. I've even ensured that I go through autofocus with MaximDL and PWI3 seperately - and have even done so by filter - to try and troubleshoot. Evan @ SRO took apart the image train and ensured that the filters were all seated properly. Any suggestions for this issue? (will post the images in a response to my own thread)
Here's a link to the dropbox folder with the four images. I've also posted the crop of the star example below. https://www.dropbox.com/scl/fo/1mkioywnp4pr49uodlxl4/h?dl=0&rlkey=9f45w0c4pflwpeyzun0adk5h9
In the event that you would like to see FITS files, I have uploaded them to the same dropbox folder. You'll also find the flats for LRGB in there (as well as the resulting integrated master flats). Note: I only uploaded one evening worth of light frames - but there are four more nights (36+ GB in total). Happy to share whatever you'd like to review. One other note. I chose this object for reference, but I've encountered this on lots of objects unfortunately, when using these SBIG filters.
Thanks, we'll look at the FITs files. What's the optical setup? e.g. what scope do you have, is there a focal reducer/field flatenner? Do you see this effect in the narrowband filters? (OIII, Ha, SII)?
If Evan has the equipment off the scope, could he flip over each of the problem filters? Then retest. My first impression is that there is a circular pattern around Gaia (about magnitude 8.9) in the NGC6960 shot. That's giving the "halo" effect. The anti-reflective coatings on the filters are a bit different on one side, and it may be the contributor. e.g. there a reflection happening, possibly internally in the filter, or between the coatings on the various optical surfaces. I'm still reviewing - that was a ton of sample data you sent over.
There's also a possible reflection of the diffraction spikes in the red and luminance filter images. Am not convinced yet. Was the AO in use?
(1) Data: I have plenty to share, and happy to do so. I have been trying to do multiple night sequences to really weed out mediocre data (which is probably good data over all but in comparison to the great after subframe selected out, I'm left with great eccentricity and FWHM). Also, I have only been shooting in High to try and figure out how to get great data with the AC4040. It's an extremely complicated camera and frustrates the heck out of most people who attempt to use it. I feel like I have to be one of DL's best use cases / user tests, considering I jumped in to the pool head first here. Obviously I'm in the weird neutral zone of 'am i reading the manual enough' or 'should I be pinging DL at this point'. (2) Optical setup: PlaneWave CDK 24, L-600, CFW7-STX 7 (SBig LRGBHaS2Oiii), PlaneWaver IRF90, StarChaser SC-3 / AO-X Bundle. No focal reducer or field flattener, but the CDK's are manufactured with a corrector lens cell that are coated for a flat field. (3) Adaptive Optics: AO was not in use for this sequence. However, I have been able to get AO working reasonably well within the last week or so (average pixel wander on a clear night is 0.10px or less for x & y). (4) Evan has reassembled the image train but I can definitely ask him to flip the filters. Should I ask him to flip just the Blue and Green or do you want me to test RGB? Not that it's terribly helpful but I have also uploaded the photo that Evan took when the filter wheel was disassembled (root directory "filter-wheel-inspection" - he opened it up on July 25th). I asked him to do it because I originally detected the halos in the integrated Green masters and thought that potentially the horrible corners I was seeing in my Green flat frames indicated a faulty filter. If you take a look at the integrated green master in the LRGB-Flats-FITS-and-Integrated folder you will see what I'm talking about. And while that may be an issue, when I detected it in Blue as well I thought that perhaps the whole set of filters might be defective. Noting that you are seeing it in Red ... I think I see what you're talking about as well, it just presents differently. (5) I have uploaded 6 more files. Two PNG files in the root directory that are integrated Luminance and Red masters. Also, in a folder called LRGB XISF Masters, you will find the four integrated LRGB masters which are in PixInsight's XSIF format and are not stretched. Not much processing done on them, just Crop, DBE, TGV Denoise, & Deconvolution. Doubt you want them, but I have the XSIF masters without those steps but I felt it made the issue easier to see for DL's review.
(Note: Sam Miller actually opened up the image train for me... they share the workload, but wanted to give credit where it was due)
Thanks for the info. We can't use XISF files, as they destroy the science data. We need FITS. I'll get back to you later, likely tomorrow.
What model of telescope is this? Does it have a corrector or reducer element directly in front of the camera?
@Doug you missed it: PlaneWave CDK 24, L-600, CFW7-STX 7 (SBig LRGBHaS2Oiii), PlaneWaver IRF90, StarChaser SC-3 / AO-X Bundle. No focal reducer or field flattener, but the CDK's are manufactured with a corrector lens cell that are coated for a flat field.
If one of the elements in the corrector is concave towards the camera, a reflection off the filter can bounce back and form an out-of-focus image on the sensor. It all depends on the curvature and spacing. Quite often when this sort of thing happens, the reflection is off-center for an off-axis star. Something to look at.
Doug - I can ping Matt at PlaneWave and ask him about this. Colin - while you & your team are reviewing - can you let me know if you would ask Evan/Sam to flip RGB or just GB to test?
Matt says that the corrector lens cell is seated into the back of the CDK24 at a fixed position and oriented at the factory. He hasn’t heard of a misaligned corrector lens assembly happening with their QA process etc.
I wasn't referring to a misaligned corrector; this was a design question. I was wondering if it had a concave element that faced the camera. If the filter were at the right distance, any reflection from it could bounce off the corrector lens and form an out-of-focus halo on the sensor. This would only be obvious on brighter stars.
We had a few bad nights of weather and/or smoke. I was able to get a clear night to test. Ok, so the RGB filters have been reversed and the problem persists. I found an area of the sky with a range of stars that we could use to evaluate. The FITS files (as well as updated flats, if you desire to try and process) are located in the same folder I shared, but in a sub folder called "09-14-22 Flip RGB Test". Here's what I'm seeing, especially still present in Green.
