Bright corners in OIII data with Redcat91 setup

andrea tasselli · Nov 14, 2025 at 06:41 PM

This typically happens when there is a difference in the calibration files (as per gain and offset) between flats and lights (or even between flats and master bias/dark). Frankly, with a 3nm filter you should have really scant background light all.

Hi Andrea,

Same set of darks/bias that were used to calibrate O also calibrated Ha. Not sure how it would affect O differently from Ha. Also gain is set at 100 for all frames and calibration frames as well

How did you do your flats? This looks to me like a flats problem.

andrea tasselli · Nov 14, 2025 at 06:52 PM

Weirder things have happened so do not underestimate the capacity of things to go belly up. I'd suggest you calibrate just the OIII frames first without dark or bias but with the flat and secondly without flat but with dark/bias and report back.

Will do…thanks

Have you tried tuning your flats and re running calibration on the o?

Odd that it is only one filter that does this. I can think of one possibility specifically for NB filters.

When light strikes a NB filter at an angle the wavelength band pass shifts. This is why faster lower F-ratio systems need special NB filters, the lower the F-ratio the greater the angle of the incident light cone on the filter, and the more the bandpass shifts. But also, the farther from the center of the FOV, the greater the angle of incidence away from perpendicular, and the more the shift. So in fast systems with a NB filter, the center of the image is brightest, the farther away from the center to the edges is darkest. All this depends then on how fast your system is, how narrow your filter band pass is (very narrow, definitely a stress), and of course how big your sensor is (full frame, another stress). The other thing is filters have an opacity to other wavelengths, this could change with angle too. And there are reflections from filters, so reflection/scattering might differ as well from center to edge. And finally this depends on the structure of the filter, its reflectivity, refractive index, thickness, and what band pass the filter is targetted to, and of course, exactly where the filter band width centre is positioned with respect to the NB line of interest. So behaviour can be very different with different filters.

All this being said, I would have though a correctly made flat would show the same behaviour as the light, and fix these types of problems. However, a very bright flat illumination can penetrate through the filter depending on it’s optical density and the angle, while the much fainter sky would not. Do note though, the more severe the vignetting, the more difficult it is to get a perfect flat.

Some things to try.

Don’t use any correction, and then see how bad the vignetting is with the different filters, is the OIII vignette much worse compared to Ha or SII without using any corrections for your lights? If so, the worse vignetting may be making your life much more difficult to get a good flat.

Make sure your OIII flat is not clipping or not too low at the bottom or too high at the top of the histogram, you want it to be in a nice linear region in the center of the range.

I don’t see how a bias or dark would affect this, as long as your bias and dark are good. I have had my bias and dark change on me suddenly, so make sure you do new fresh ones.

Personally, I always do biases, darks and flats in otherwise dark room, to prevent any stray light, doing flats in a bright room could result in a light leak somewhere in your system.

Rick

Daemon de Chaeney · Nov 15, 2025, 09:41 AM

Hi Rabeea,

Something you said makes me think you are using the same set of flats for all filters. If that were the case, that would probably be the problem, instead of having a set and master for each filter. Apologies if I misunderstood you though. The other thing to note is that 2” mounted filters are a bit narrow for the 6200 and usually unmounted 50mm rounds are recommended so you have less vignette and worse, no reflections off the filter mount which will definitely fall on the frame.

Cheers

Daemon

Hi Daemon,

Different flats are taken for each filter

Could it be a gradient from the moon one side and from LP on the other? Or moon on oneside→Meridian Flip→Moon on the other side?

John Hayes · Nov 15, 2025, 05:21 PM

CMOS cameras generally have a very linear response curve; however, some sensors are not linear with very short exposures. How much exposure time did you use to create your flats? In general, as long as your exposures are longer than around 2 seconds, you shouldn’t have a problem. However, if you used shorter exposures, you could be running into a problem with linearity.

Remember that if there is no vignetting, no stray light, the sensor is linear, the darks are matched, and you use the correct math to do the calibration, it will always work.

John

The IMX455 does appear to show this behaviour. My first flats with it were taken well under one second and despite being correctly lit and appearing more or less ok, were diabolically bad for calibration. I went to 3 second flats and have never had a calibration issue since.

AnakChan · Nov 16, 2025 at 04:15 AM

I have been living with the ASI6200MM with overcorrected flats for about the past 4-5 years. Most prevelant with the Oiii filter (similar to the OP). I did not notice this issue with ZWO’s 2” mounted (I think they’re 7nm), but switching to the Antlia Pro 3nm unmounted, I see them on the Oiii.

I’ve tried :-
- on different my different scopes of different focal lengths and different F-ratio (Takahashi µ250CRS, Pentax 125SDP, Takahashi FSQ85ED) - note same imaging train(*) from tilt adapter to camera, but different appropriate extension tubes per scope.
- different gains (namely 0 & 100) and offsets (20, 50 100), and different temperatures (0C, -5C. -10C) ; and yes matching darks/flats/biases and new, not reused….also tried flat darks instead of bias)
- different flat sources - proper astro LED panels, tracing panels, sky flats, T-Shirt method, etc.
- different ADU percentages (NINA flatwizard) from 20% up to 80% in 10% increments
- (*) masking any potential light leakages between the camera & EFW, EFW & OAG, OAG & tilt adapter, etc.
- moved my filter masking around to see if a particular filter mask was introducing the issue
- loaning my Antlia Oiii to my friend who used it with his QHY600 and did not see any issues

So far no luck. Thinking is the camera left, I posted to the ZWOAsrto BBS forum as I was happy to pay for an ovehaul/refresh of my 5 yr old ASI6200MM. But, they suggested increasing my Oiii lights sub exposures (I was using 5 min), and higher flats ADU. Trying this, did improve with 10 min Oiii light subs, but not with 15 min light subs. I probably need to experiment more but after 5 years of trying different things, and having tried many different settings & combinations, I’m running out of ideas.

P.S. my note for the OP here is “You’re not alone with this issue”.

Feels good to know im not struggling alone 😅

Do you think its a manufacturing defect on the 6200mm or is it an issue known for all users? Also I havent tried stacking without flats, is it by any chance better without flats?

Rabeea Alkuwari · Nov 16, 2025 at 06:19 AM

Feels good to know im not struggling alone 😅

Do you think its a manufacturing defect on the 6200mm or is it an issue known for all users? Also I havent tried stacking without flats, is it by any chance better without flats?

TBF, I really don’t know if it’s a manufacturing defect or not. But I ordered in late 2019, and received delivery in early 2020 - so I’m really one of the 1st batches. ZWO has revised the ASI6200MM since - specs seem the same but it seems to have a new external design/chassis at least. That’s kinda why I posted onto the ZWO’s BBS site. I’m happy to pay for an overhaul but seems they wanted me to try longer exposures and higher flats ADU.

I think I have tried stacking without flats but TBH, I need flats - primarily with the motes and dust spots, it’s so tedious to create synthetic dust spots to remove.

I’ll try try longer light Oiii subs again with greater ADUs, but that may not necessarily be practical for, say, bright targets where one may not want to have longer Oiii’s (in Bortle 6-7 skies) - balancing on sky fog with “long enough” subs to fix overcorrected flats would be a PITA.

I may try a different camera one day.