Need help with a long-standing problem

[jaydeepappas]

Warning: This is a long and detailed post!

Hi all. This is my first post here on Stargazers lounge. I have been on CloudyNights for some time and love the forum. I created a thread regarding this issue on CN but was never able to get the help I needed to fix the problem, and the thread has been dead for some time now. So I am turning to you guys to see if I can finally get this figured out. CN thread is here: https://www.cloudynights.com/topic/878303-image-calibration-issues-or-problems-with-my-data/

Basically, I got a new telescope - the AstroTech AT115EDT - and am having an issue with a strange artifact/gradient when using the scope. I only experience the issue when imaging in LRGB, not in narrowband. Here is an image of what I see after performing a dynamic background extraction in PixInsight:

 

Things I have tried so far:

- Eliminating light leaks. I did initially have a light leak from my ZWO filter wheel that I have since eliminated. I am pretty confident that I have no more light leaks, as I can take long exposure darks at high gain in a fully illuminated room without any noticeable difference in the ADU values.

- Ensuring that it is not related to calibration issues. While it is a faint, I can see it even in my individual sub frames - especially after performing a DBE. Even without any sort of processing though, during acquisition even, I can see a faint outline of the gradient from the ASIAir preview on my iPad so I am certain that the artifact is not generated from incorrect calibration. I can also see this ring in my flat frames.

- New reducer. I contacted AstroTech and they agreed to send me a new reducer since I noticed that the problem didn't happen when I imaged without the reducer. This did not fix the problem - however,  I think the fact that the problem is eliminated without the reducer may be a potential hint as to what is happening. I just don't quite know exactly what that hint would be

- New telescope. After lots of back and forth with AstroTech they agreed to send me a new telescope - very good customer service from them. Unfortunately this also did not solve the problem, so the issue _must_ be related to my equipment. What is also strange is I can not replicate this problem on my other telescope, the Redcat 71, with the same exact equipment. Potentially another hint...

- Removed the filter wheel and confirmed that the artifact still shows up without any filters - using only extensions to reach 55mm backfocus.

 

I am suspicious that it may be a reflection somewhere in the optical train, I just don't know where. I bought some adhesive flocking material and flocked the inside of the 16.5mm ZWO extension I am using. That also did not solve the problem. The dew shield on the scope is not reflective at all. It _does_ look a bit like there is a reflection when I look inside of the OTA of the telescope, however I can't imagine that AT would be selling telescopes that have reflective surfaces in the OTA, or that I somehow managed to get 2 defective telescopes.

 

As you can imagine I am pretty frustrated. I really don't know what else to do or try here. Maybe it is a byproduct of the environment I am in, high bortle zone (7/8 ?) and some stray light coming into my backyard from other sources. Maybe these are causing the problem I am seeing. I guess I could take all the equipment to a dark site and see if the problem persists... /shrug

 

Any help is greatly appreciated. It is also worth reading through the CN thread as there is also some information there. For what it's worth, I really like my AT115EDT - I just want to get everything working properly. I was still able to capture some nice data of M51 with the scope, and crop it heavily enough that the artifact is not in the image:

 

The relevant equipment is as follows:

- AT115EDT with dedicated 0.8x reducer/flattener

- 16.5mm ZWO extension tube

- ZWO 36mm 7x filter wheel

- Antlia VPRO LRGB filters, Antlia 3nm SHO filters

- ZWO ASI2600mm

 

Thanks to everyone who has made it this far and to those who are willing to help!

 

Edited to include a single light frame, flat, dark, and flat-dark.

Light_M51_60.0s_Bin1_L_gain0_20230620-213554_0.2C_0001.fit Flat_808.3ms_Bin1_L_gain0_20230620-235855_0.2C_0001.fit Dark_808.3ms_Bin1_S_gain0_20230621-000446_0.0C_0001.fit Dark_60.0s_Bin1_gain0_20230614-113619_0.0C_0018.fit

Edited July 5, 2023 by jaydeepappas
Include some FITS files.

[ONIKKINEN]

The initial thought i had in mind is a flats calibration issue, since this kind of ring artifact happens when your light frames and flat frames do not mechanically match (because for example focuser tilt/sag, unreliable collimation, image train stability issues between the lights and the flats). You did mention you have thought of calibration issues but this is still my guess. I used to have an unreliable focuser and an unreliably thin aluminum tube in my newtonian, and most of the time flats had this kind of ring artifact which went away after making sure everything is mechanically sound.

Any ideas that come to mind are just wild guesses without seeing the raw data though. Can you post a raw light, a matching dark frame, darkflat frame (or bias, whichever you use) and a matching flat frame in raw format straight of the camera? You can attach .fits (or any other) files here on SGL just by dragging and dropping them below where you write a comment so no need to link something to an external file hosting site.

[Elp]

Have you got access to, or tried a different set of LRGB filters? I also assume the issue is not present when you try imaging without any filter present? Another thing to try with your existing LRGB filters is either screw it directly onto to the reducer/flattener if using one or try a singular filter drawer to eliminate the filter wheel out of the equation entirely.

Edited July 4, 2023 by Elp

[jaydeepappas]

3 hours ago, ONIKKINEN said:

The initial thought i had in mind is a flats calibration issue, since this kind of ring artifact happens when your light frames and flat frames do not mechanically match (because for example focuser tilt/sag, unreliable collimation, image train stability issues between the lights and the flats). You did mention you have thought of calibration issues but this is still my guess. I used to have an unreliable focuser and an unreliably thin aluminum tube in my newtonian, and most of the time flats had this kind of ring artifact which went away after making sure everything is mechanically sound.

Any ideas that come to mind are just wild guesses without seeing the raw data though. Can you post a raw light, a matching dark frame, darkflat frame (or bias, whichever you use) and a matching flat frame in raw format straight of the camera? You can attach .fits (or any other) files here on SGL just by dragging and dropping them below where you write a comment so no need to link something to an external file hosting site.

I initially also thought that it was a combination of focuser slop and tilt with my CAA, as I noticed that my vignetting was inconsistent between my flat and light frames with the new scope. The vignetting would move around depending on which direction the imaging train was leaning. I have since fixed this (mostly) by tightening up the focuser tension screws and removing the CAA from my imaging train, yet the gradient is still there. Maybe it is possible that my focuser still has a small amount of movement that could be causing this. The vignetting in my master lights for the luminance filter did not perfectly calibrate out, so that does point towards calibration issues.

I edited my original post to include a couple of FITS files - light, dark, flat, and flat-dark. One thing to note is that even without any calibration, the ring is still visible in a single light frame. If you perform a DBE or ABE you will notice the ring immediately, which is why I don't think it is a calibration issue. Unless maybe that ring is supposed to calibrate out..?

 

3 hours ago, Elp said:

Have you got access to, or tried a different set of LRGB filters? I also assume the issue is not present when you try imaging without any filter present? Another thing to try with your existing LRGB filters is either screw it directly onto to the reducer/flattener if using one or try a singular filter drawer to eliminate the filter wheel out of the equation entirely.

I should have mentioned this in the original post, but I have imaged without any filters or filter wheel and the problem remains - just with a couple extensions to reach backfocus. I have updated my original post to include this fact. This is another reason why I feel like it may be a reflection of some sort... maybe the narrowband filters are less prone to a reflection due to their tight band pass. But that is just a wild guess.

Edited July 5, 2023 by jaydeepappas

[Paul Puntin]

Your thoughts on an internal reflection are probably correct. I had a similar issue with my ES127. The problem for me was the first 2 inches of the focuser weren’t blackened correctly. This could be the issue anywhere in the imaging train. See this topic on Cloudynights in which I contributed . https://www.cloudynights.com/topic/868565-rings-in-image-not-the-obvious-causes-now-what/

 

First time posting, so hopefully the images upload ok. Not sure if the order will be correct but I think it’s pretty obvious which ones are before the blackening applied.

[sinbad40]

Probably makes me sound stupid, but have you tried adding an extended dew shield to what is already on, maybe some dark card or something just to extend it another few inches?

[AstroMuni]

14 hours ago, jaydeepappas said:

and am having an issue with a strange artifact/gradient when using the scope.

I have taken your attached FITS files and done a quick pixel math using SIRIL (lights/flats) and I cant see any artefact of the kind you are seeing. So maybe something not quite correctly configured in the software that you are using for calibrating and stacking.

This is your flat

This is your lights

And here is the Lights/Flats

Edited July 5, 2023 by AstroMuni

[ONIKKINEN]

11 hours ago, jaydeepappas said:

I initially also thought that it was a combination of focuser slop and tilt with my CAA, as I noticed that my vignetting was inconsistent between my flat and light frames with the new scope. The vignetting would move around depending on which direction the imaging train was leaning. I have since fixed this (mostly) by tightening up the focuser tension screws and removing the CAA from my imaging train, yet the gradient is still there. Maybe it is possible that my focuser still has a small amount of movement that could be causing this. The vignetting in my master lights for the luminance filter did not perfectly calibrate out, so that does point towards calibration issues.

I edited my original post to include a couple of FITS files - light, dark, flat, and flat-dark. One thing to note is that even without any calibration, the ring is still visible in a single light frame. If you perform a DBE or ABE you will notice the ring immediately, which is why I don't think it is a calibration issue. Unless maybe that ring is supposed to calibrate out..?

 

I should have mentioned this in the original post, but I have imaged without any filters or filter wheel and the problem remains - just with a couple extensions to reach backfocus. I have updated my original post to include this fact. This is another reason why I feel like it may be a reflection of some sort... maybe the narrowband filters are less prone to a reflection due to their tight band pass. But that is just a wild guess.

I reached the same conclusion as @AstroMuni above, calibration with this set of frames works as it should. Below a false colour rendering at an extreme stretch:

As we can see the image is very flat, apart from the sky gradient but this is unrelated to the issue. No ring artifact to be seen.

What you describe with the focuser tensioning and such is exactly what makes the ring thing happen, because at one camera orientation gravity will act on the imaging train differently than in another. You could test the stability of your whole system though, and its fairly easy to do so.

To do that, take a flat with the telescope horizontal to the east, take another with the scope horizontal to the west, then one to the zenith one to the south and north etc. Point is to get a flat in as many imaging train orientations as possible here. Then do a simple pixelmath division between any of the 2 flats (flat east / flat west for example), what you should get is a uniform gray mess with no clear gradient or any features to be seen. If you see a gradient or any features other than noise, you have a mechanical problem. This method reveals the smallest issues with your mechanical problems in the scope, give it a try?

[jaydeepappas]

Thank you to everyone who has replied so far. You all have given me a lot to think about and some new ways to approach this problem.

Since the issue doesn't show up when using narrowband filters I am currently shooting the crescent nebula, so that has been taking most of my time lately. It's nice to take a break from troubleshooting and actually start getting some more data again.

 

On 05/07/2023 at 01:58, Paul Puntin said:

Your thoughts on an internal reflection are probably correct. I had a similar issue with my ES127. The problem for me was the first 2 inches of the focuser weren’t blackened correctly. This could be the issue anywhere in the imaging train. See this topic on Cloudynights in which I contributed . https://www.cloudynights.com/topic/868565-rings-in-image-not-the-obvious-causes-now-what/

 

First time posting, so hopefully the images upload ok. Not sure if the order will be correct but I think it’s pretty obvious which ones are before the blackening applied.

This is something I am considering, although I am skeptical since I received a new scope from the same company which had the same issue. Unless they are sending out all of these telescopes with incorrectly blackened focusers, which I doubt. But maybe my specific equipment has some weird interaction with this focuser in general. I am going to attempt more accessible troubleshooting steps before I resort to flocking the focuser interior.

On 05/07/2023 at 05:36, sinbad40 said:

Probably makes me sound stupid, but have you tried adding an extended dew shield to what is already on, maybe some dark card or something just to extend it another few inches?

I actually haven't tried this, but it is a quick and easy way to rule something out. I will give it a go!

On 05/07/2023 at 05:43, AstroMuni said:

I have taken your attached FITS files and done a quick pixel math using SIRIL (lights/flats) and I cant see any artefact of the kind you are seeing. So maybe something not quite correctly configured in the software that you are using for calibrating and stacking.

This is your flat

This is your lights

And here is the Lights/Flats

This is perplexing to me. Since I have been shooting the crescent nebula, I haven't had any issues with artifacts using NB filters. I have no problems with calibration and my master lights turn out great. I have not changed any settings in WBPP since moving from LRGB to SHO. Is it possible to run a DBE or ABE on your resulting calibrated frame? I suspect that the ring is still there in your calibrated image but won't really show until a background extraction is done on it. At least this is what I've experienced so far.

On 05/07/2023 at 06:32, ONIKKINEN said:

I reached the same conclusion as @AstroMuni above, calibration with this set of frames works as it should. Below a false colour rendering at an extreme stretch:

As we can see the image is very flat, apart from the sky gradient but this is unrelated to the issue. No ring artifact to be seen.

What you describe with the focuser tensioning and such is exactly what makes the ring thing happen, because at one camera orientation gravity will act on the imaging train differently than in another. You could test the stability of your whole system though, and its fairly easy to do so.

To do that, take a flat with the telescope horizontal to the east, take another with the scope horizontal to the west, then one to the zenith one to the south and north etc. Point is to get a flat in as many imaging train orientations as possible here. Then do a simple pixelmath division between any of the 2 flats (flat east / flat west for example), what you should get is a uniform gray mess with no clear gradient or any features to be seen. If you see a gradient or any features other than noise, you have a mechanical problem. This method reveals the smallest issues with your mechanical problems in the scope, give it a try?

This is some great advice, taking the images at different angles and using pixelmath to verify the movement of the focuser. Something I did not think of, and a pretty conclusive way to rule out focuser movement. When I received the scope I had to remove the CAA because the vignetting was moving around my images so much depending on the angle my imaging train was. After removing the CAA I could tell in my flat frames that the vignetting was still moving a bit, and tightening the focuser screws seemed to have fixed that. However I did not run any of the test flats through pixelmath - I was just verifying the movement in the flats with my eyes. Which is definitely not as accurate as pixelmath

[AstroMuni]

4 hours ago, jaydeepappas said:

Is it possible to run a DBE or ABE on your resulting calibrated frame?

I dont use PI but did apply background extraction in Siril and no artefacts 😞

[dhickey]

On 10/07/2023 at 00:05, jaydeepappas said:

Thank you to everyone who has replied so far. You all have given me a lot to think about and some new ways to approach this problem.

Since the issue doesn't show up when using narrowband filters I am currently shooting the crescent nebula, so that has been taking most of my time lately. It's nice to take a break from troubleshooting and actually start getting some more data again.

 

This is something I am considering, although I am skeptical since I received a new scope from the same company which had the same issue. Unless they are sending out all of these telescopes with incorrectly blackened focusers, which I doubt. But maybe my specific equipment has some weird interaction with this focuser in general. I am going to attempt more accessible troubleshooting steps before I resort to flocking the focuser interior.

I actually haven't tried this, but it is a quick and easy way to rule something out. I will give it a go!

This is perplexing to me. Since I have been shooting the crescent nebula, I haven't had any issues with artifacts using NB filters. I have no problems with calibration and my master lights turn out great. I have not changed any settings in WBPP since moving from LRGB to SHO. Is it possible to run a DBE or ABE on your resulting calibrated frame? I suspect that the ring is still there in your calibrated image but won't really show until a background extraction is done on it. At least this is what I've experienced so far.

This is some great advice, taking the images at different angles and using pixelmath to verify the movement of the focuser. Something I did not think of, and a pretty conclusive way to rule out focuser movement. When I received the scope I had to remove the CAA because the vignetting was moving around my images so much depending on the angle my imaging train was. After removing the CAA I could tell in my flat frames that the vignetting was still moving a bit, and tightening the focuser screws seemed to have fixed that. However I did not run any of the test flats through pixelmath - I was just verifying the movement in the flats with my eyes. Which is definitely not as accurate as pixelmath

 

I was researching something else when this thread showed up in Google.  I don't know if you have since solved the issue, but I had something very similar with my AT115EDT that I recently resolved.  In my case, the equipment is the AT115EDT, MoonLite focuser, ASI294MM with 1.25" AstroDon filters, the AT2FF field flattener (I have the reducer too but haven't used it in a while), and a Pegasus Falcon rotator.  I saw the exact same ring as in your original post.  Long story short, after meticulously checking each part of the imaging train the culprit was the focuser.  Specifically, the adapter that threads into the 68mm drawtube on the MoonLite and adapts it down to 42mm threads.  It's a flat black surface, but applying flocking material from ScopeStuff eliminated the large ring in the images.  When I looked down the imaging train with the camera removed, I didn't initially see any reflections, but if I moved my eye slightly off-center, I could see it.  The flocking on that adapter eliminated it both visually looking in from where the camera is as well as in the images.  It was only present in the LRGB frames, the narrowband were fine.  I'm guessing because I'm under very light polluted skies (Bortle 7/8).

Along similar lines, I had an earlier issue where I saw arcs in the corners, so it looked like a very large reflection that was clipping the edge of the sensor.  That one turned out to be the filters.  AstroDon 1.25" filters, and again, the inside of the mounting ring was black but looked a little shiny on the LRGB ones.  I cut a thin piece of flocking material and carefully applied to the inside of the mounting ring.  The next time I imaged, those artifacts in the corners were gone.

Dave