Something's gone wrong and I need help!

[BrendanC]

Hi all,

My past three shoots have yielded weird results - unaccountable vignetting and poor detail.

In my efforts to understand what's going on, I tried restacking the last decent image I did, about a month ago. It came out differently. I don't know why.

This is how it came out originally, about a month ago, just very quickly stretched in StarTools, no colour processing etc:

This is how it came out just now, using the same files in DSS:

See what I mean? It's almost as if the flats aren't being applied, or something. They're using the exact same files - darks, dark flat, flats, no bias.

This leads me to believe that something, somewhere, has changed. My thoughts are:

* Something's different in DSS, possibly to do with the flats. I've tried stacking on other machines and get the same results however. I'm going to give this another go to make sure.

* Something's wrong with my darks. I extended my darks library while the weather was bad, so I'm wondering whether some of my darks have 'gone bad' somehow, and now everything's over-compensating. I've got an offsite backup of the original library so I'll be getting hold of that too and seeing if that gets me anywhere.

Either way, I really, really, really need someone's help before I spent hours and hours trying to figure this one out. I've got lots of plans for the next few clear nights but without ironing this problem out, I'm in real trouble.

Does anyone have any suggestion about something that might have changed in DSS to do this? Or, is my hypothesis about the changed darks library a possibility?

Thanks, Brendan

[PeterCPC]

I don't use Startools or DSS but, as the only thing that you have changed is the darks, I would try a reprocess using the old darks to see how that turns out.

[BrendanC]

I think you're right. It's almost as if there's an overcorrection happening. The flats are definitely fine, I just took them in exactly the same way I always do.

Any more opinions on this one?

Thanks again for the help.

[ampleamp]

With clear nights still forecast, you can still take your subs while you figure out what has gone wrong; then you will have lots of processing during the next lot of clouds 😁  

This does look like an over-correction and to me looks like an overcorrection of the flats, at least with the above that’s where I would start, temperature differences??. When I struggled once with calibration I used the trial version of APP; changing software made me rethink what I was doing. What I liked about APP was you just throw everything into the same folder, let it sort everything out while you enjoy the sunshine before you pick up the result. At least you will know if you can get a result with your existing data before you spend ages taking more. If you use the old calibration files first and then compare with a new stack recalibrated with the new calibration files you can start to tie things down with existing files, swoping different old/new groups in and out etc. I often create a raw uncalibrated stack, ie a baseline right out of the camera. Sure there will be vignetting etc but assuming it is better than above, you have removed the camera from the list of possibilities. If things were working, why did you update the calibration files? Unfortunately I dint think there are shortcuts here. 

[BrendanC]

Thank you, this seems to me to be bang on the money.

So, I've identified two things (I've been up since 6am working on this - yes, it worries me that much!)

• There were some rogue 180s files in my 120s darks library. I've removed them, and that's fixed this problem. HOWEVER, it hasn't fixed other images that I've taken over the past two nights.
• I've stacked those images with no calibration at all, and the result is much, much better. So that tells me it's not acquisition or the camera. Now, given that the darks should be ok - and that I get the same result if I use the older darks - I'm starting to think that this is a separate problem with the flats. So, I'm going to retake the flats (the camera is still in the same orientation) and see what happens there.

The reason I extended the darks library was that I didn't have many subs at certain temperatures, so I just added some. But, as per my bullet above, I accidentally moved some of the new ones to the wrong folder. Doh!

So yes, a thorny problem which I believe is possibly two separate problems. Main thing is, as you say, the actual data is/are fine and if/when I nail this, I've got lots to be doing. 

[ampleamp]

Glad to hear you have made progress 👍

[BrendanC]

Yep. After much scratching of head and getting a bit sweary, I think I know what to do to fix this all now.

If there are any other issues going on, then after fixing these two, I'm sure they'll rear their ugly heads next!

[BrendanC]

Hi all,

Sorted. It was the darks getting mixed up, and re-shooting the flats did the rest.

Thanks for everyone's help!