ZWO ASI585MC Pro: strange lines in the B channel. - Forums

Hi.

I have recently bought an ASI585MC and, after some test, I have seen that there are strange lines in the B channel that appears in the final image even after calibration ( after Bias, Darks and Flats with WBPP ). I have contacted to ZWO's support department and they say it's something you can expect, as Sony doesn't warrant the uniformity of the sensor. They say I should use Flats to remove it, as I already have, with no success.

Anybody out there with the same issue? Seems that the ZWO's support department is there just to say "NO".

Thank you

Are the diagonal bands visible in either bias, dark or flat? If so they can be corrected. If not then the bands are due to something else, most likely in the optics ahead of the sensor. Besides, if you bought it recently you might still be able to return and get a full refund.

andrea tasselli:
Are the diagonal bands visible in either bias, dark or flat? If so they can be corrected. If not then the bands are due to something else, most likely in the optics ahead of the sensor. Besides, if you bought it recently you might still be able to return and get a full refund.

Are visible in the Flats, but when I apply them to the process in WBPP, the Flats doesn't remove the lines ( they appear as a blue shadow ). It's nothing related to the the optics, as the Flat image ( only B channel, the R and G are not affected ) included here was made just with the camera and the flat panel ( no telescope involved in this case ). Of course, the Flats used in the image process were made with the telescope. And yes, that's suppose to be the case, but they say that this is something related to Sony, not to them, so they haven't offered the option for an exchange or refund.

So, with the telescope attached, are the bands visbile in the master flat? Can you show them? And not the debayered ones, the original raw master flat (assuming the bias/dark isn't affected). If so, then a properly applied and created master flat would remove them (otherwise what would be the purpose of the flat?). And by "properly" I mean avoiding WBPP entirely.

What flat panel did you use?
It might be an issue with the refresh rate of the LEDs if the panel uses PWM to operate the LEDs.

Are the lights effected as well? If not there must be something wrong with the flats (LEDs in Flatpanel?).

if yes, it could be the sensor. Try other camera settings, and see if the lines still appear in your images.

andrea tasselli:
So, with the telescope attached, are the bands visbile in the master flat? Can you show them? And not the debayered ones, the original raw master flat (assuming the bias/dark isn't affected). If so, then a properly applied and created master flat would remove them (otherwise what would be the purpose of the flat?). And by "properly" I mean avoiding WBPP entirely.

I haven't used any other method than WBPP for a long time, so I have to remember the older process to generate a Master Flat. Never have had any issue with it, so what for?. Anyway, in raw flat images without debayering ( or darks, bias ), nothing can be seeing ( only "flat" grey images ). Only when I apply the debayering, the lines appears in the blue channel. This happens both if I make the flats using the telescope or shooting directly to the flat panel. So it's nothing related to the optic train. I will generate a Master Flat using ImageCalibration and post it later.

Edited: MasterFlat using ImageCalibration+ImageIntegration: looks exactly the same than the one generated by WBPP. Flat grey without debayering; Lines visible after debayering.

Jean-David Gadina:
What flat panel did you use?
It might be an issue with the refresh rate of the LEDs if the panel uses PWM to operate the LEDs.

Thanks for the tip. I will try with another flat panel to check if it could be related to that.

Armin Lukas:
Are the lights effected as well? If not there must be something wrong with the flats (LEDs in Flatpanel?).

if yes, it could be the sensor. Try other camera settings, and see if the lines still appear in your images.

Yes, the "Lights" are affected too. I have tried different gains to check if it could be affected by, but the lines remain there. According to ZWO's support, is something you can expect from Sony sensors, but I have at least 4 cameras with Sony sensors and haven't seen this before. I am trying to know if others with this new camera are experiencing the same issue.

Edited: MasterFlat using ImageCalibration+ImageIntegration: looks exactly the same than the one generated by WBPP. Flat grey without debayering; Lines visible after debayering.

You have to extract the B channel from the master raw flat. This will contain only the B pixels and not the blank ones and will show whether you have the issue as well or is related to something else. I guess is called CFASplit or something similar.

And then, obviously, if it is there alright then the question is in what way the calibration doesn't address this issue. So you'll have to revise your calibration steps one by one. Same gain, offset in all those steps and check the linearity of the sensor across the board (well, gain values) as there might be jumps with this lower class of sensors.

andrea tasselli:
Edited: MasterFlat using ImageCalibration+ImageIntegration: looks exactly the same than the one generated by WBPP. Flat grey without debayering; Lines visible after debayering.

You have to extract the B channel from the master raw flat. This will contain only the B pixels and not the blank ones and will show whether you have the issue as well or is related to something else. I guess is called CFASplit or something similar.

Thank you for the clue.

I have made a CFASplit and this is what the B channel shows: the lines.

I have already tried with different Gains ( the ASIAir doesn't allow to select a different offset ) and the result is just the same.

I have been chatting with a couple of Uranus-C cameras owners ( same sensor IMX585 ) and they don't have this issue. So Sony is not the problem, but ZWO. They only allow me to return the camera if the technical support considers that this is due to the making process, but their answer have been: "Sony don't guarantee uniformity of channel", so it's easier to blame others than to take action and exchange the device. I will try to deal with this somehow ( may be creating a synthetic Flat ) until I sell the camera and buy it from a different vendor ( as I see that PlayerOne is more serious than ZWO, I will buy one from them and move away from the ASIAir ). I have had to deal with ZWO's technical support arbitrariness in the past ( the fan vibration or the oil leak in the 2600 ) and I'm done with their behavior.

While I do approve of moving out of ZWO ecosystem I still think you should be able to calibrate that effect away.

andrea tasselli:
While I do approve of moving out of ZWO ecosystem I still think you should be able to calibrate that effect away.

I have tried different approach today and I always get the same result: blue/yellow line crossing from bottom left to top right. I have tried with ImageCalibration, applying different PixelMath operations, following a tutorial to make a synthetic flat, trying to make new flats... Nothing works here. I can't see any real alternative, because I haven't seen something like this out there ( although I will keep searching ). The only way I get those lines out is painting them with CloneStamp or with Photoshop. The obvious problem is when I remove something that really exist in the image... A complete mess!!

Well, the usual way with hard-to-crack cases such as this in AB is to make the source images and calibration files available to the wider audience here and see whether some-else's approach might yield the hoped-for results. Just a suggestion here.

andrea tasselli:
Well, the usual way with hard-to-crack cases such as this in AB is to make the source images and calibration files available to the wider audience here and see whether some-else's approach might yield the hoped-for results. Just a suggestion here.

Ok, I have upload to my Google Drive a .zip file with the MasterBias, MasterDark, MasterFlat and 10 Lights in .fit format ( almost 200 MB ). May somebody knows how to deal with this:

https://drive.google.com/file/d/1Ef_Iew7tY5m44hqNgzJIPTb4ADZUeHgm/view?usp=sharing

If I'm not grossly mistaken you're taking these images without any filters. And with a refractor. This is very much no-no. In fact NO-NO. The image won't color calibrate properly and neither would the flats (which, given the example, are rather too flat). Get a IRCut pronto and repeat again. There is still a faint trace left but I guess the main issue is the massive IR leak.

andrea tasselli:
If I'm not grossly mistaken you're taking these images without any filters. And with a refractor. This is very much no-no. In fact NO-NO. The image won't color calibrate properly and neither would the flats (which, given the example, are rather too flat). Get a IRCut pronto and repeat again. There is still a faint trace left but I guess the main issue is the massive IR leak.

Yes, you are right. So I have added tonight the IRCut filter. It's windy, so don't think I'm going to get good results, but at least confirm if the problem in the channel B can be solved that way. I made the Flats with the Automatic option in the ASIAir. It works perfectly both with the Hyperstar and the Focal Reducer in my C14, but seems that I have to try manually with this camera and scope. Thank you.

andrea tasselli:
If I'm not grossly mistaken you're taking these images without any filters. And with a refractor. This is very much no-no. In fact NO-NO. The image won't color calibrate properly and neither would the flats (which, given the example, are rather too flat). Get a IRCut pronto and repeat again. There is still a faint trace left but I guess the main issue is the massive IR leak.

Yes, you are right. So I have added tonight the IRCut filter. It's windy, so don't think I'm going to get good results, but at least confirm if the problem in the channel B can be solved that way. I made the Flats with the Automatic option in the ASIAir. It works perfectly both with the Hyperstar and the Focal Reducer in my C14, but seems that I have to try manually with this camera and scope. Thank you.

Finally, you were right!! It was the filter. I use to work in mono or with a Canon DSLR, only with an MC for planetary, and didn't know I have to use the IRCut filter for DSO with this camera. Now it looks as expected.

Thank you very much for your support and patience

Looks like I have something similiar on my asi585mc pro. Im using sv bony uv/ir cut filter and those lines are present with and without filter on all of my flats,master flats and theres a faint blue tint on my final images too. Regardless of camera orientation, lines are always going from top right corner into the center.

📷 flat.jpg

AdamSWA · May 21, 2026, 06:55 AM
Looks like I have something similiar on my asi585mc pro. Im using sv bony uv/ir cut filter and those lines are present with and without filter on all of my flats,master flats and theres a faint blue tint on my final images too. Regardless of camera orientation, lines are always going from top right corner into the center.
📷 flat.jpg

Your histogram is far to the right. Take new flats, get it to left of center.

If using Nina flat wizard target about 30k to start.

Hopefully that fixed it.

TiffsAndAstro · May 21, 2026, 12:14 PM
AdamSWA · May 21, 2026, 06:55 AM
Looks like I have something similiar on my asi585mc pro. Im using sv bony uv/ir cut filter and those lines are present with and without filter on all of my flats,master flats and theres a faint blue tint on my final images too. Regardless of camera orientation, lines are always going from top right corner into the center.
📷 flat.jpg
Your histogram is far to the right. Take new flats, get it to left of center.
If using Nina flat wizard target about 30k to start.
Hopefully that fixed it.

Single flat histogram is around 30k. I dont know why, but after stacking with naztronomy script in siril, histogram on master flat goes way to the right. All my flats have this blue line, even before stacking into master flat.

Take a flat without optics (or even an image if you know how to do it) with a pinhole lens and check that the bar is still there. If you can’t calibrate it away and it doesn’t depend on the optics then it’s a defect of the colour dye.