FLAT with a strange pattern, I can't properly correct the LIGHTS / SkyWatcher APO ESPRIT 120 / x0.77 Reducer SkayWatcher / QHYY 268 mono / Chroma 3nm SII 36mm and OIII - Forums

Topic starter

16 days ago

Hello, I have a problem I can't seem to solve. If anyone can shed some light on it, I'd appreciate it.

I'm testing a new setup: SkyWatcher APO ESPRIT 120 / x0.77 Reducer SkyWatcher / QHY 268 mono / Chroma Ha 5nm 36mm / Chroma SII 3nm 36mm / Chroma OIII 3nm 36mm.

Sunlight, light, and dark (S/H/O) exposures with a gain of 56 and an offset of 25 at 300 seconds produce a circular pattern that I've never had with other setups. Then, when I take flat exposures, I get the same pattern, and it only corrects the Ha exposures correctly; the SII and OIII exposures are not corrected properly.

I've tried changing the light source for the flat exposures (looking at a clear sky) with the ADUs (20000/25000), but it doesn't improve anything.

I've also tried increasing the value of the "Output pedestal (DN)" field between 100 and 500 in the "ImageCalibration" process, specifically in the "Output Files" section, but it didn't solve the problem. I've tried many different values and adjustments, but it either overcorrects or undercorrects. I haven't been able to find the solution, if there even is one.

If anyone has had a similar problem or knows of a solution, I would appreciate it.

Regards.

📷 Captura de pantalla 2026-02-23 212208.png📷 Captura de pantalla 2026-03-18 185620.png

Well Written Respectful

andrea tasselli

Frequent contributor

16 days ago

Dew on the sensor, most likely. What is your set cooling temperature?

Well Written Engaging

Jose Luis Bedmar

Topic starter

16 days ago

Hi Andrea, first of all, thank you for your interest.

The temperature is -10°C and the humidity is 35%.

Best regards.

Well Written Respectful

andrea tasselli

Frequent contributor

16 days ago

Then most likely isn't dew, too bad as that would have been easy to fix. If they always had that pattern and doesn't appear with broadband filters then the issue is with the filters.

Jose Luis Bedmar

Topic starter

16 days ago

The strange thing is that with that same SETUP and the zwo 294 nothing strange comes out, even though the sensor is smaller that is...

andrea tasselli

Frequent contributor

16 days ago

So, same everything BUT the camera and all the problems disappear? And only in NB? Does the peculiar pattern appears on the ASI294MM too but it is correctable via flats? Or just isn't there?

Jason dain

16 days ago

I had the same issue on my Esprit 100 and the Starizona reducer. Ended up sending it back as I couldn’t solve it.

Jose Luis Bedmar

Topic starter

16 days ago

Thanks for commenting Jason…

Harald Becher

16 days ago

I have the same issue with my newton, but not with the askar120. My problem seems to come from new LED streetlights (time of installation ist related to it).

Maybe you check neighborhood.

So it could be light pollution which is collected and so not correctable.

You can try skyflats which seems to help (at least in my situation)

Harald

Helpful Supportive

Jose Luis Bedmar

Topic starter

15 days ago

Hi Harald.

I've tried using Skyflat and it doesn't improve things. The pattern appears in both the Lights and Flats images taken with a flat sheet and with shots of the sky, so light pollution shouldn't be affecting it. From yesterday to today, I found something that might be the problem (I have to do a test this weekend). In the same setup, with an Astronomik Ha filter, the pattern doesn't appear (in the image of seven flat shots, the one in the center). The only difference is that this filter is 1mm thick and the others are 3mm, which gives me a clue. The Chroma filters protrude from the carousel, while the Astronomik remains below, not visible. This already indicates that the ones that do protrude have this pattern, and the one that doesn't (they're all on the filter wheel) doesn't show the strange pattern. I must have some residual light that's affecting the others from the side. I'm going to try covering the filters around their edges with something I can and see the effect.

Thank you for your comment.

Best regards.

Jose Luis.

Well Written Helpful Insightful Respectful Engaging

Michal Dubovsky

15 days ago

Hello,

have you used also dark frames? I had a similar issue with my setup, the problem was that I did not use darks (only lights and flats). The sensor had significant offset - the darks had average brightness of around 40% of the “empty” night sky brightness. The problem is that you definitely need to subtract this offset (i.e. use dark frames, or at least bias frames) before flat frame is applied.

Michal

Well Written Helpful Respectful Concise Engaging Supportive

Jose Luis Bedmar

Topic starter

15 days ago

Michal Dubovsky · Mar 19, 2026, 03:08 PM
Hello,
have you used also dark frames? I had a similar issue with my setup, the problem was that I did not use darks (only lights and flats). The sensor had significant offset - the darks had average brightness of around 40% of the “empty” night sky brightness. The problem is that you definitely need to subtract this offset (i.e. use dark frames, or at least bias frames) before flat frame is applied.
Michal

Hi, I just tried it manually, first calibrated with dark mode and then with flat mode, and it's still not working.

Thanks for your interest.

Best regards.

Well Written Respectful

Richard

15 days ago

Hey Jose,

I’ve got a slightly different setup (WO FLT120 / QHY268M / QHYCFW3 / Antlia 4.5nm SHO and LRGB) but I had a very similar, if not the same problem!

In short, the circles you see is from light entering the side of the filter and the darker shapes in the 10 o'clock 2 o'clock and 6 o'clock positions correlate with the screw positions used hold the filters in place.

See below comment on SGL where I show my flats looking like yours. I didn’t have trouble calibrating my lights, but I like to prevent any issue from happening in the first place.

https://stargazerslounge.com/topic/389405-qhy268m-qhy268c-users-thread-please-share-your-knowledge-and-tips/page/5/#findComment-4262282

To remedy this, remove the felt pad and screws provided with the filterwheel and use 3D printed masks which cover the perimeter of the filters and hold them in place (I can highly recommend BuckeyeStargazer). This prevents light entering the side of the filter and also does not show “shadowing” from the three screw positions per filter. Your flats will be nice and clean!

See the below post on SGL where I compare the flats with and without filter masks.

https://stargazerslounge.com/topic/389405-qhy268m-qhy268c-users-thread-please-share-your-knowledge-and-tips/page/5/#findComment-4264463

Hopefully cross posting to other forum is permitted, and this helps you with your scenario! Sorry that this does not fix your current data :(

Well Written Helpful Respectful Engaging Supportive

Marcin Cikała

15 days ago

Hi.

Do small test. Remove the reducer and take the flat. One is enough. Do the inspection and write what happens.

Regards, Marcin.

Riccardo Civati

15 days ago

Hi,

I had a similar problem years ago (zwo ASI 2600mm, chroma 5nm 36mm, TS 130 with flattener).

Basically, the flats generated with the OIII filter resulted in a very pronounced and sharp light disc (it wasn't the classic soft vignetting), while the other filters were perfect.

The problem, however, was that it wasn't visible on the lights frames, and so the flats ruined the lights.

I tried everything and finally got another flattener and solved the problem.

I don't know if this helps, but I wanted to share my experience.

Riccardo

Well Written Respectful Concise Engaging Supportive

Jose Luis Bedmar

Topic starter

15 days ago

Richard · Mar 19, 2026, 07:08 PM
Hey Jose,
I’ve got a slightly different setup (WO FLT120 / QHY268M / QHYCFW3 / Antlia 4.5nm SHO and LRGB) but I had a very similar, if not the same problem!
In short, the circles you see is from light entering the side of the filter and the darker shapes in the 10 o'clock 2 o'clock and 6 o'clock positions correlate with the screw positions used hold the filters in place.
See below comment on SGL where I show my flats looking like yours. I didn’t have trouble calibrating my lights, but I like to prevent any issue from happening in the first place.
https://stargazerslounge.com/topic/389405-qhy268m-qhy268c-users-thread-please-share-your-knowledge-and-tips/page/5/#findComment-4262282
To remedy this, remove the felt pad and screws provided with the filterwheel and use 3D printed masks which cover the perimeter of the filters and hold them in place (I can highly recommend BuckeyeStargazer). This prevents light entering the side of the filter and also does not show “shadowing” from the three screw positions per filter. Your flats will be nice and clean!
See the below post on SGL where I compare the flats with and without filter masks.
https://stargazerslounge.com/topic/389405-qhy268m-qhy268c-users-thread-please-share-your-knowledge-and-tips/page/5/#findComment-4264463
Hopefully cross posting to other forum is permitted, and this helps you with your scenario! Sorry that this does not fix your current data :(

Thank you so much for your comment, Richard.

I was heading in the same direction as you mentioned. Looking at my QHY filter wheel carousel, I have six 3mm Chroma filters and one 1mm Astronomik Ha filter. In one of the images I uploaded showing the seven flats of the seven filters, only the 1mm Astronomik, which doesn't protrude from the carousel, doesn't have that circular pattern. This leads me to believe it's exactly as you described. The other Chroma filters do protrude, some more than others, but they all show the pattern. So, I've downloaded some 3D printing files specifically for the QHY filter wheel and the Chroma filters (I have a 3D printer at home and I'm going to give it a try).

Thank you very much for your interest and for commenting. Best regards (I hope to try it soon and will post the results)...

Well Written Respectful Engaging

Jose Luis Bedmar

Topic starter

15 days ago

Riccardo Civati · Mar 19, 2026, 08:25 PM
Hi,
I had a similar problem years ago (zwo ASI 2600mm, chroma 5nm 36mm, TS 130 with flattener).
Basically, the flats generated with the OIII filter resulted in a very pronounced and sharp light disc (it wasn't the classic soft vignetting), while the other filters were perfect.
The problem, however, was that it wasn't visible on the lights frames, and so the flats ruined the lights.
I tried everything and finally got another flattener and solved the problem.
I don't know if this helps, but I wanted to share my experience.
Riccardo

Hello Riccardo.

All the comments are very helpful to me. I really appreciate you sharing your experiences and taking the time to do so. I'll keep them in mind if I can't find another solution.

Best regards.

Well Written Respectful Supportive

Jose Luis Bedmar

Topic starter

15 days ago

Marcin Cikała · Mar 19, 2026, 08:02 PM
Hi.
Do small test. Remove the reducer and take the flat. One is enough. Do the inspection and write what happens.
Regards, Marcin.

Hi Marcin.

I'll definitely try it and see how it goes.

Thank you so much for your help.

Best regards.

Well Written Respectful

Marcin Cikała

15 days ago

In my RC16 f/8 I used a 2” 0.67x reducer. Rest of my setup was very similar to yours: 36mm unmounted filters, Asi 2600, working distance 55mm. Had very similar patterns (light circle). I was almost sure that this is due to the light leak (buffler problem) but discover that the reducer is responsible for issue.

It should be enough to take a rough flat and do the inspection with flat contour plot in PI or SAO DS9.

IMHO 36 mm unmounted filters have enough clear aperture to work properly with APS-C matrix with your scope.

Marcin.

Helpful

Jose Luis Bedmar

Topic starter

15 days ago

Marcin Cikała · Mar 19, 2026, 09:01 PM
In my RC16 f/8 I used a 2” 0.67x reducer. Rest of my setup was very similar to yours: 36mm unmounted filters, Asi 2600, working distance 55mm. Had very similar patterns (light circle). I was almost sure that this is due to the light leak (buffler problem) but discover that the reducer is responsible for issue.
It should be enough to take a rough flat and do the inspection with flat contour plot in PI or SAO DS9.
IMHO 36 mm unmounted filters have enough clear aperture to work properly with APS-C matrix with your scope.
Marcin.

Hi Marcin, the truth is that with the same Chroma Ha 36 filter, the pattern appears, but it calibrates perfectly, and there's no noticeable vignetting from the filter being just right. I agree with you; it's just right, but sufficient for the sensor, considering how close the filters are to the sensor. I've read a lot about this lately, and they all lead me to the same conclusion you mentioned: it's caused by the reducer/flattener. Although it can be improved quite a bit, and I could take advantage of the setup as “@Richard” mentioned earlier in this thread. I have the printed rings ready to assemble soon and see the result.

Thank you sincerely for all the information you're giving me. Best regards.

Respectful Supportive

Jose Luis Bedmar

Topic starter

13 days ago

Hello.

I think I've found the problem; at least the FLAT tests have improved, though I still need to test it with LIGHTS (time permitting).

The issue seems to be the height of the filters (3mm) and some light leakage from the rotator.

Because they protruded from the carousel and were only held in place with screws and washers, the protruding part of the filters, along with some residual light, was what was appearing in the LIGHTS and FLAT tests.

To solve the filter problem, I found a 3D printable design for custom washers for 36mm Chroma filters and the QHY CFW3 wheel. Apparently, someone else (whose generosity I greatly appreciate) had the same problem and designed these custom rings for 2mm and 3mm filters. I also found them for these filters on ZWO wheels (if anyone needs them, I can share the 3D printing files).

For the leak in the ZWO AAC rotator, I tried everything I could think of, every possible gap.

I'm including before and after pictures. I haven't been able to take any light exposures yet to check if I can calibrate them (even so, it looks promising since the flats, while not perfectly clean, were the same with my other setup and it corrected them perfectly). With these narrowband filters, the flats have always come out a bit odd but manageable.

Thank you all for your help, and to @Richard in particular, I want to say that he hit the bullseye, THANK YOU!!!

📷 CON ANILLAS 3D Y TODO TAPADO_FINAL.png📷 WhatsApp Image 2026-03-21 at 20.53.11 (4).jpg📷 WhatsApp Image 2026-03-21 at 20.53.11 (3).jpg📷 WhatsApp Image 2026-03-21 at 20.53.11 (1).jpg📷 WhatsApp Image 2026-03-21 at 20.53.11 (2).jpg📷 WhatsApp Image 2026-03-21 at 20.53.11.jpg

Helpful Respectful Engaging Supportive

Richard

13 days ago

Awesome, those flats are looking much better! It does make you wonder why filterwheel manufacturers don’t provide filter masks like these, should be very easy to manufacture.

I'm looking forward to seeing your new lights come through without calibration issues :)

Well Written Respectful Engaging Supportive