A test for RASA + OSC in severe light pollution areas

6 replies73 views
YingtianZZZ avatar

As I recently published in this picture: https://app.astrobin.com/i/mpt4sg

I’ve been working on making my RASA11 + 2400MC Pro setup working for a while. However, I was getting very strange artifects even with skylight flats, night sky flats, or flat panel. I always get something like this: 📷 image.pngimage.pngLast night I revisited the setup, and I thought of one of my friend told me OSC needs more effort to use: you have to split RGB, integrate them separately and then recombine them. I wasn’t using OSC at that time. But somehow I took his advice. I retake skylight flat at dawn, then I debayer all images - Flats, Darks, Lights - to RGB separate channels. Then I got a very flat background - at least compared to previous one. As top image shows, even with aggressive stretch, artifects are not overwhelming targets across the full frame image. bb63dd9adc5e577977b45971a402c2d8.pngSo for anyone who is also suffering strange artifects with OSC in high light pollution area, I would suuggest a try of this way, just first debayer to separate RGB channel and integrate, maybe the result will be much better

Helpful Engaging Supportive
andrea tasselli avatar

This is very much a NO-NO thing to do. You don’t debayer calibration frames or the true calibration just isn’t happening at all.

YingtianZZZ avatar

andrea tasselli · Apr 29, 2026, 09:17 AM

This is very much a NO-NO thing to do. You don’t debayer calibration frames or the true calibration just isn’t happening at all.

I see no reason for this. Debayer is math, why not?

andrea tasselli avatar

Everything is math, ultimately. But no, de-bayering is interpolation using best-fit functions so your original calibration data goes out of the window (signally noise profiles). Besides, you calibrate the raw data so it is a case of like with like. If you want to split the CFA array in its components then it is a different thing.

YingtianZZZ avatar

andrea tasselli · Apr 29, 2026, 09:38 AM

Everything is math, ultimately. But no, de-bayering is interpolation using best-fit functions so your original calibration data goes out of the window (signally noise profiles). Besides, you calibrate the raw data so it is a case of like with like. If you want to split the CFA array in its components then it is a different thing.

I think you are right. I’ll check the workflow, then follow this: 1. raw RGB image do dark+flat calibrate, 2. Debayer them, 3. Integrate 3 channels separately. Is this correct?

andrea tasselli avatar

If you calibrate (this needs include the cosmetic correction phase if you have one) the raw frames with the un-debayered master calibration frames then that would be all right.

Rostokko avatar

Is splitting integration across 3 channels actually going to lead to different (better) results? If so, why?

Related discussions
Need help with Flat calibration
Hi guys, I’m in a big problem for two months and couldn’t find a solution. In my remote setup of Skywatcher-250 Quattro, I’m consisting getting dust donuts in low height areas, maybe light polluted sky area. I know that easiest way is clean the filte...
Flat calibration issues directly relate to author's artifact problems with flats.
Feb 17, 2026
ZWO ASI 2600MC producing doughnut artifact in center of images
Hello everyone, I have had a ZWO ASI 2600MC for about a year now, and I have noticed that all the images I have taken with it since then have a strange artifact in them. Basically, it looks like a doughnut in the center of the image. This is particul...
Doughnut artifact in OSC camera similar to author's strange artifacts with 2400MC Pro.
Nov 27, 2025