Patchy cloud, poor transparency and light edges to subs.

[Mike73]

I've had quite a few clear nights recently but transparency has been fairly average at best with some patches of clouds occasionally drifting through.

Each morning I check my subs deleting the ones which have this obvious edge brightness (that brightness varies in strength) then add all the what appear to be good subs to APP.

I calibrate all my subs with 40 x darks, flats and dark flats then check in APP that my calibration files have worked which as you can see in the first screen shot they do. Now the problem starts when I integrate all my lights and calibration files into a final image and I get light edges as in the second screen shot. It does look like a problem with my flats but as each single calibrated sub looks fine I dont think flats are the problem (although could be wrong I usually am!?).

Each sub is 300s and at the end of each night I probably have around 50% that are keepers.

Does this look like a problem due to cloud or am I barking up the wrong tree altogether?

[ollypenrice]

This cannot reasonably be a problem coming from the sky since it is aligned perfectly with your chip, nicely parallel with the two short sides. That could happen - but wouldn't!

I think it may also be present in the single calibrated sub you present here but, in a single sub, it is much less obvious. You could check by measuring the background brightness along a horizontal line starting at the left hand edge and working inwards. Is the background in the single sub consistently brighter near the edge?

My guess is over-correction in the flats, a well known problem without, unfortunately, a well-known universal remedy. Often it comes from not using flat darks but you say you've used them. Do your flats show a change in brightness along the two short sides? Maybe post up a stretched flat here?

I wonder what would happen if you made your flat darks slightly longer (or possibly shorter) than the flats exposure...

Olly

[tomato]

Has this problem just appeared on this dataset or do you have others that you can compare it to from the the same location and set up? It would be better to find and eliminate the cause of the introduced artefact, but being as you have the data in APP, have you tried running the light pollution correction tool on the image? It does a great job on removing gradients, and I’m guessing the algorithm doesn’t differentiate between sources of the gradient, so it’s worth a go.

[Mike73]

@ollypenrice @tomato thanks for your input. 🙂

 

Olly I completely agree that it makes no sense that the problem stems from the sky but its the only variable light source that could effect my subs and since this problem isn't consistent and I'm only seeing around 50% of subs like this it was the only thing I could think of.

You can see it quite clearly again in the single sub from last night but until now I haven't looked at my master dark which very closely matches the light areas you see in the single uncalibrated sub. 🤔

All the master calibration files are recent, made in the last week. My master darks were taken in my darkroom (I still shoot and print from film).

 

Yes I did try the Light pollution Correction tool in APP but didn't really know what I was doing so just got a crazy result 😂. Theres also Gradient Xterminator in PS as a last option as well but I'd much rather get to the source of the problem first.

 

[Mike73]

..and this is 55 x 300s calibrated and heavily stretched from last night.

 

[tomato]

I cropped the edges on your 55 x 300s in APP then ran the LP correction tool locating sample boxes in the lighter regions on each side and some strategically located boxes in the areas with less nebulosity.  I am not a widefield  nebula imager so I'm not sure if the resulting image is closer to what would be expected or if APP has over corrected.

 

[Mike73]

3 hours ago, tomato said:

I cropped the edges on your 55 x 300s in APP then ran the LP correction tool locating sample boxes in the lighter regions on each side and some strategically located boxes in the areas with less nebulosity.  I am not a widefield  nebula imager so I'm not sure if the resulting image is closer to what would be expected or if APP has over corrected.

After your previous comment I thought what the hell I'II try the trial of Gradient Xterminator and was really surprised how well it did. I'm not finished with trying to get to the bottom of the strange subs but at least using software I can move on. I've read quite a few people mention how good APP's LP Correction tool is, I just haven't got that far into processing with it yet (only downfall to APP is lack of guides IMO).

But thanks for bringing it to my attention anyway, I'II dig into it tomorrow. 👍

[ollypenrice]

You say you did your master darks in your darkroom. Our ASI camera has a metal chip cover. You should certainly use this for taking darks, but perhaps you did. Keeping light out of sensitive cameras by any other means is difficult to impossible in my experience.

The effect you're getting would be a doddle to fix using Pixinsisght's Dynamic Background Extraction.

Olly

Edit: Perfect repair by Tomato to my eye.

Edited September 4, 2021 by ollypenrice

[david_taurus83]

It looks a lot like the ASI1600 amp glow. Have you tried to calibrate with another software like DSS?

Edit; Also, in your first post, it's still present in the first screenshot. It's evident on the right hand side. Very faint but it's there. I'm guessing this is a calibration issue with darks.

Edited September 4, 2021 by david_taurus83