Help - Vertical bands - Flats is the culprit - Can someone help me understand what's happening ?

Deepan Vishal · Nov 23, 2025 at 03:37 PM

Hi,

Recently I am noticing some weird band like structures in my images.
My setup - QHY600PH-M + Takahashi Epsilon 180ED + Astronomik Filters.

I Captured RGB images in QHY 2CMS-1 Readout mode at 56 gain for broadband images. (mistake)
After noticing these weird structures, I decided to narrow down the issue and finally identified the problem - The Flats calibration.

Here are two images -
Left image calibrated with a flat frame (same gain and offset settings) long back.
Right image calibrated with a flat frame after my recent imaging session. (same gain and offset)
I can not use the previous flat frames since the orientation of camera setup changed and there are vignetting and dustmotes that can not be calibrated out.

📷 Flats.jpg
So the only difference between left and right frame is the flats. The other potential reason is the settings in processing.

Experts of Pixinsight and Astrobin, can you please help me what might be causing this?

Thanks in Advance!
Deepan

For a trouble shooting experiment, you could rotate each individual flat frame 90 degrees and apply Canon Debanding at about .75 strength. Then you’d need to rotate them back to their original orientation and save out each result. After that, you could use these new flats in your stacking flow and see if that calibrates it out. I have to do a similar process for darks and dark flats for my SVBony SV405CC which is known to have banding issues due to the electronics.

Now, for the root cause, are you using a flats panel, taking sky flats, or a diffused light pad?

Deepan Vishal · Nov 23, 2025 at 04:57 PM

bigCatAstro · Nov 23, 2025, 04:41 PM

For a trouble shooting experiment, you could rotate each individual flat frame 90 degrees and apply Canon Debanding at about .75 strength. Then you’d need to rotate them back to their original orientation and save out each result. After that, you could use these new flats in your stacking flow and see if that calibrates it out. I have to do a similar process for darks and dark flats for my SVBony SV405CC which is known to have banding issues due to the electronics.

Now, for the root cause, are you using a flats panel, taking sky flats, or a diffused light pad?

Thanks for the message. I am using a diffused light pad.

These bandings don’t appear straight and doesn’t seem to be coming from electronics.
I did try the canon banding script in pixinsight. It helps to an extent, but I was trying to get to the root of the problem to curb it.

The method that I was detailing for debanding requires you to deband each frame individually since the pattern is random. You won’t be able to fix it entirely by applying it to the end result only.

Regarding the root cause, have you had a chance to look at your filters? I wonder if there’s something going on with them. Are you taking flats for each filter used?

Also, does this camera have the photo mode read-out capability?

Brian Puhl · Nov 23, 2025, 06:02 PM

Did you take new bias frames as well?

I actually had a similar issue when I switched to 2CMS mode, and I never figured it out so I just went back to normal.

HI Brian,

I never take bias frames. Only Dark flats to calibrate the flats.
That is how I have imaged with other IMX455 sensors.

Is my understanding wrong? Do the QHY IMX 455 needs bias frame calibration as well?


And Yes, that is what I am considering. going back to regular photgraphic reacdout model.

John Hayes · Nov 23, 2025, 07:01 PM

Deepan Vishal · Nov 23, 2025, 06:05 PM

HI Brian,

I never take bias frames. Only Dark flats to calibrate the flats.
That is how I have imaged with other IMX455 sensors.

Is my understanding wrong? Do the QHY IMX 455 needs bias frame calibration as well?


And Yes, that is what I am considering. going back to regular photgraphic reacdout model.

It’s not camera mode. How are you taking your flats and what exposures are you using? Do you see this with every filter? It sounds like you are using flat darks. What exposure time are you using for those?

John

Hi John,

Thanks for the message.

I am using a diffused light panel to take flats. Worked perfect until this encounter.
I see this with LRGB filters. My Ha integration looks clean surprisingly!
I am matching the exposure time with the flats.
And usually all the flats are sub 1 seconds (could that be a problem?)

Another development - I stacked a bunch of files without flats or darks, I see those vertical lines after removing the vignette using graxpert.

This leaves with the suspicion on the readout modes.
Did I mess up by using 2CMS-1 which is a high gain mode on broadband images?

John Hayes · Nov 24, 2025, 03:04 AM

Deepan Vishal · Nov 23, 2025, 11:18 PM

Hi John,

Thanks for the message.

I am using a diffused light panel to take flats. Worked perfect until this encounter.
I see this with LRGB filters. My Ha integration looks clean surprisingly!
I am matching the exposure time with the flats.
And usually all the flats are sub 1 seconds (could that be a problem?)

Another development - I stacked a bunch of files without flats or darks, I see those vertical lines after removing the vignette using graxpert.

This leaves with the suspicion on the readout modes.
Did I mess up by using 2CMS-1 which is a high gain mode on broadband images?

First, as a matter of good practice, you should adjust the flat illumination so that you can use 2 second or longer exposures. I personally try to use 3s or longer. Second, unless you are using very long exposures for your flats, you don’t need to use flat darks. I personally never use flat darks with my IMX455 cameras and I’ve never seen a problem even with up to 500s exposures.

Have you tried adjusting the USB speed in the driver? That may be the source of your problem. How long is your cable and have you cleaned the connectors? From what you are describing, it sounds like it might be due to some electrical issue. I run the same camera in the original 2CMS mode and I’ve never seen this sort of problem so something funny is going on here. Again, it shouldn’t be the mode—unless something is wrong with the camera.

John

Hi John,

Thanks a lot.
Since I was not sure of what the problem was, I tried several combinations of debugging. Narrowed down to two potential problems -

1. My pixinsight WBPP settings
2. Multi night images

This images was captured over few nights. Though I take care to preprocess each nights data separately, I have inadvertently preprocessed multiple nights in a single stack instead of stacking them separately.

Then, I reset the WBPP pixinsight settings and integrated them again.
And the problem is now gone!

Thanks again for your help!

📷 Problem_fixed.jpg

Deepan Vishal · Nov 28, 2025, 04:57 PM

John Hayes · Nov 24, 2025, 03:04 AM

Deepan Vishal · Nov 23, 2025, 11:18 PM

Hi John,

Thanks for the message.

I am using a diffused light panel to take flats. Worked perfect until this encounter.
I see this with LRGB filters. My Ha integration looks clean surprisingly!
I am matching the exposure time with the flats.
And usually all the flats are sub 1 seconds (could that be a problem?)

Another development - I stacked a bunch of files without flats or darks, I see those vertical lines after removing the vignette using graxpert.

This leaves with the suspicion on the readout modes.
Did I mess up by using 2CMS-1 which is a high gain mode on broadband images?

First, as a matter of good practice, you should adjust the flat illumination so that you can use 2 second or longer exposures. I personally try to use 3s or longer. Second, unless you are using very long exposures for your flats, you don’t need to use flat darks. I personally never use flat darks with my IMX455 cameras and I’ve never seen a problem even with up to 500s exposures.

Have you tried adjusting the USB speed in the driver? That may be the source of your problem. How long is your cable and have you cleaned the connectors? From what you are describing, it sounds like it might be due to some electrical issue. I run the same camera in the original 2CMS mode and I’ve never seen this sort of problem so something funny is going on here. Again, it shouldn’t be the mode—unless something is wrong with the camera.

John

Hi John,

Thanks a lot.
Since I was not sure of what the problem was, I tried several combinations of debugging. Narrowed down to two potential problems -

1. My pixinsight WBPP settings
2. Multi night images

This images was captured over few nights. Though I take care to preprocess each nights data separately, I have inadvertently preprocessed multiple nights in a single stack instead of stacking them separately.

Then, I reset the WBPP pixinsight settings and integrated them again.
And the problem is now gone!

Thanks again for your help!

📷 Problem_fixed.jpg

Good stuff Deepan! Sometimes those settings can be critical and I’ve run into similar problems every once in a while. All it takes is a little bump to the settings to get it back into the right solution space. For me, it is often a problem related to the automatically selected master for either the local normalization or the alignment that causes trouble. Either way, I’m glad to see that you got if fixed.

• John