Strange banding in O3, only with this target - Forums

Topic starter

Oct 6, 2022

Hello fellow astrophotographers,
I have been trying to understand this issue for a while now, and I hope our collective minds can help. I have been imaging with my setup for a while without any issues (RC8, Chroma 3nm OIII, 294MM bin2x2, gain 200, 0 degrees, 600s).

For this particular target "PK111 11.1", I am seeing a weird pattern (both with wbpp and dss). I cleared the light files in wbpp (reusing same flats, flat darks and darks) and added lights for another target (captured after PK111 11.1), the issue isn't there.

I tried integrating without calibration frames, issue is still there. I tried without flats only, still there. Without darks only, still there. I cleaned half of my worst images (total of 204 images, 10 minutes each), and still, the issue is there.

Ha for this target doesn't show this issue. O3 for other targets doesn't show this issue.

Any other ideas? I would appreciate it...

Helpful Engaging

andrea tasselli

Frequent contributor

Oct 6, 2022

If it is only this target then it must be related to the location of the object in the sky. Maybe the filter is tilting somewhat there?

Michel Makhlouta

Topic starter

Oct 6, 2022

andrea tasselli:
If it is only this target then it must be related to the location of the object in the sky. Maybe the filter is tilting somewhat there?

I will check this tonight and see if the filter is loose. But I doubt that's the case, I image from my balcony with a limited field of view, my telescope is always pointing somewhere between NE and N and can't image a target when it reaches above 65-70 degrees... pretty limited range of motion

Christoph Lichtblau

IOTD Staff

Oct 6, 2022

Dear Michel,

I had the same issue some time ago. It was caused by running patterns of my flat lamp, when the exposure time was too short.
It can not have something to do with the object. Try new flats please.

Cheers, Christoph

Concise

Michel Makhlouta

Topic starter

Oct 6, 2022

I found something new…

Stacking pre meridian flip: no issue seen
Stacking post meridian flip: issue is very visible

What could this mean I wonder… aside from the filter stability which I will check tonight

andrea tasselli

Frequent contributor

Oct 6, 2022

It doesn't need to be the filter in its seat itself, just something that flips that side of the meridian and obviously in the light path. BTW, how many bands can you count on the image?

Michel Makhlouta

Topic starter

Oct 6, 2022

the filter is not loose, I checked now, couldn't wait till tonight. I count vertically 28 but it's hard to be exact. It looks like horizontal bands are there too. Could it be stray lights? Given my balcon position, post meridian flip exposes it more to the neighboring building, while pre it is more to its back.

Exaggerated stretch:

andrea tasselli

Frequent contributor

Oct 6, 2022

Well, they seem pretty square to me so I'd reckon it doesn't matter whether is vertical or horizontal. Optical thickness associate with the banding is 0.68 so I'd reckon it gives around 2mm geometrical thickness, i.e., a filter. It could be stray light off to an angle and into the filter. It should happen in precious few locations when the right combos meet, IMHO. Not what I would call lucky…

Michel Makhlouta

Topic starter

Oct 6, 2022

andrea tasselli:
Well, they seem pretty square to me so I'd reckon it doesn't matter whether is vertical or horizontal. Optical thickness associate with the banding is 0.68 so I'd reckon it gives around 2mm geometrical thickness, i.e., a filter. It could be stray light off to an angle and into the filter. It should happen in precious few locations when the right combos meet, IMHO. Not what I would call lucky...

interesting, also the telescope sits behind the balcony fence, which is glass, not ideal. I want to go through some iterations to stack some images excluding near meridian... I will get back with the results... even previous attempts seem to mix pre/post meridian. Thanks for the support so far

Jonny Bravo

Oct 6, 2022

Did I read it right that you're imaging through a glass fence? I would be putting the blame there… it's probably some kind of polarized / tempered / reinforced glass. It could very well be that at this particular angle with that particular wavelength of light you're getting some kind of reflection / interference.

Well Written Insightful Respectful Concise Engaging

Michel Makhlouta

Topic starter

Oct 6, 2022

Jonny Bravo:
Did I read it right that you're imaging through a glass fence? I would be putting the blame there... it's probably some kind of polarized / tempered / reinforced glass. It could very well be that at this particular angle with that particular wavelength of light you're getting some kind of reflection / interference.

No, the fence is lower than the telescope, it's just a statement that its glass and not solid metal, so I have zero protection from stray lights coming from neighboring buildings.

Anyway, I did some more tests.

Pre flip: no issue
Post flip: issue

I will try to wake up during the weekend during the flip and observe other deltas... This became weirder now...

Dominik Weinbrenner

Oct 6, 2022 · Edited: Oct 6, 2022

Just a thought: register using bilinear and drizzle 2x afterwards.

I had something similar once and this workflow cleaned it up (hinted by Juan of PI fame).

Jonny Bravo

Oct 6, 2022

Michel Makhlouta:
Jonny Bravo:
Did I read it right that you're imaging through a glass fence? I would be putting the blame there... it's probably some kind of polarized / tempered / reinforced glass. It could very well be that at this particular angle with that particular wavelength of light you're getting some kind of reflection / interference.

No, the fence is lower than the telescope, it's just a statement that its glass and not solid metal, so I have zero protection from stray lights coming from neighboring buildings.

Anyway, I did some more tests.

Pre flip: no issue
Post flip: issue

I will try to wake up during the weekend during the flip and observe other deltas... This became weirder now...

Ahh. Sorry I misread that. Thanks for clarifying!

Michel Makhlouta

Topic starter

Oct 6, 2022

I removed the last hour from the post meridian flip files and stacked them. Here's the post meridian vs post meridian minus last hour of imaging:

It seems the problem is appearing at the last hour of imaging. I am doing a full stack without that hour to check. Then will investigate what's happening at the last hour.

Lorenzo Siciliano

Oct 6, 2022

Just a thought… It seems much like a moiré pattern… Maybe too little dithering and a very slight field rotation after meridian flip. When stacking it can create such a interference fringes.

GalacticRAVE

Oct 8, 2022

Those moiré type patterns can appear in connection with registration and noisy data (and slight rotations) combined with subsequent registration. Juan Canejero (the pixinsight grand master) had an article on the pixinsight forum on this many years ago, if I recollect correctly. That‘s why you see it post meridian flip, your image is probably slightly rotated compared to pre MF (and you register against an image pre MF - I am speculating here). Work arounds are: a) change your interpolation formula (sometimes that helps) or (b) integrate by drizzle (with n=1)

Matthias

Helpful Insightful Respectful

Tim Hawkes

Nov 3, 2022

I too have seen this pattern when signal to noise is poor. Agree with GalacticRAVE and Siciliano comments above
Tim

Roy Hagen

Nov 3, 2022

Tim Hawkes:
I too have seen this pattern when signal to noise is poor. Agree with GalacticRAVE and Siciliano comments above
Tim

I had this problem with my ASI071 in darker areas of the images with less signal.
Shot a new set of calibration frames with 10sec. flats and dark flats, and included a set of bias frames.
The problem seems to be gone
Roy

Rouz Astro

Nov 3, 2022

I have seen that at times.
Winsorized integration take care of it most of the time.
Its best to dither often, with large distance as well.

Hope that's helpful.

CS
Rouz

Chris Fellows

Nov 3, 2022

Check your flats..
CmF