Is this bad flats or something else?

[simmo39]

Hi all. Not sure where Im supposed to post this one so If im in the wrong place I sorry. Now to the problem: As it gallaxy season ( not my fav time of year ) I thought I would give the Leo trip ago using my SW 72 ed. I know is not the scope for this target but I was just playing and trying to get my head around RGB imaging. I have had some good results with my set up in narrow band but RGB is giving me real problems. Im not sure wheather its my flats or internal reflections or just to long subs but im getting greenish swiirls in my images on all 3 filters ( less so on the green ). I can remove some of it with masks and ABE / DBE in pixinsight but there is still a stubborn section that is always there. Here is the rgb image after combination and a stretch;

Im happy with the targits but the bacground is driving me up the wall. Any ideas on whats wrong or how I can get rid of the haze?

Thanks i advance.

Simmo

[vlaiv]

This looks like high altitude clouds rather than anything else.

You can check if it is by doing a trick - split subs into two groups - stack each group as it was whole session (you can do this with only one channel if you did LRGB imaging) - compare two resulting stacks for this pattern. If pattern is the same - it is camera related, if it's not the same - it varies with time and it is probably external - most likely high altitude clouds.

You can even inspect how many of the frames are impacted - calibrate them and bin them like crazy (so you get very small size - something like 320x240 or so) - this will boost SNR of each sub so you'll be able to see all of the features in each sub. Do same stretch on each and make animated gif out of it - it will show passing clouds if this was in fact what caused issue in the first place.

[simmo39]

1 hour ago, vlaiv said:

This looks like high altitude clouds rather than anything else.

You can check if it is by doing a trick - split subs into two groups - stack each group as it was whole session (you can do this with only one channel if you did LRGB imaging) - compare two resulting stacks for this pattern. If pattern is the same - it is camera related, if it's not the same - it varies with time and it is probably external - most likely high altitude clouds.

You can even inspect how many of the frames are impacted - calibrate them and bin them like crazy (so you get very small size - something like 320x240 or so) - this will boost SNR of each sub so you'll be able to see all of the features in each sub. Do same stretch on each and make animated gif out of it - it will show passing clouds if this was in fact what caused issue in the first place.

Hi Vlaiv, I have done something like you sugest. The rgb subs where taken on seperate nights and the patern is the same for all 3 sets of subs. The only improvment i got was on the third night doing green when I increased the ADU of the flats in APT from 20000 to 25000. The patturn was still there but less prononced. Thats why im wondering if its a flats problem. Im using a gain of 76 on my ASI 1600mm for the subs whitch is new for me as I have been using 200 for narrow band would that have any bearing?

[vlaiv]

I've seen on numerous occasions that people have issues with APT and automatic flats. Could be that it is issue with flats.

How about doing flats "manually"? Avoid any automatic procedure for flat creation and just shoot regular "light" subs with flat panel and stack those yourself to see if it will help?

[simmo39]

1 minute ago, vlaiv said:

I've seen on numerous occasions that people have issues with APT and automatic flats. Could be that it is issue with flats.

How about doing flats "manually"? Avoid any automatic procedure for flat creation and just shoot regular "light" subs with flat panel and stack those yourself to see if it will help?

WEll ill give that a try tonite and see what I get. The only thing Im not to sure of is the actule sub length to us for the flats. I have read all the thearoy stuff on flats and I have to say a lot of it is over my head but will give it a try.

[Anthonyexmouth]

5 minutes ago, simmo39 said:

WEll ill give that a try tonite and see what I get. The only thing Im not to sure of is the actule sub length to us for the flats. I have read all the thearoy stuff on flats and I have to say a lot of it is over my head but will give it a try.

I'll be watching for the answer to this myself, been using APT for my flats too and maybe it's time to try a new method. 

[vlaiv]

1 hour ago, simmo39 said:

WEll ill give that a try tonite and see what I get. The only thing Im not to sure of is the actule sub length to us for the flats. I have read all the thearoy stuff on flats and I have to say a lot of it is over my head but will give it a try.

You need to "guess" it - by looking at histogram. It needs to be to the right but it must not clip in any of the colors (there will be three peaks and all should be visible and not clipped to the right). Once you find exposure length that works - shoot flats with that exposure length and remember it. Then shoot flat darks with same exposure length.

[simmo39]

18 hours ago, vlaiv said:

You need to "guess" it - by looking at histogram. It needs to be to the right but it must not clip in any of the colors (there will be three peaks and all should be visible and not clipped to the right). Once you find exposure length that works - shoot flats with that exposure length and remember it. Then shoot flat darks with same exposure length.

Hi vlaiv. Well i tried last night with my best guess and it did come out a little better, Still not 100% right but I think Im heading in the right direction. It didnt help that my framing was way out with my other subs but at least I feel im getting somewhere. I still wonder aslo about internal reflections but not sure how to best sort that.

If you look close you can still see the swirls but not half as bad as what it was.

[simmo39]

19 hours ago, Anthonyexmouth said:

I'll be watching for the answer to this myself, been using APT for my flats too and maybe it's time to try a new method. 

Hi Anthony, not sure if i have solved the problem but looks better. It was all guess work Im afraid so no cunning plan to tell.

[vlaiv]

2 minutes ago, simmo39 said:

Still not 100% right but I think Im heading in the right direction.

Could you post following:

1. light sub

2. dark sub (or master dark)

3. flat sub

4. flat dark sub

(or bias if that is what you are using instead of flat dark)

[simmo39]

1 minute ago, vlaiv said:

Could you post following:

1. light sub

2. dark sub (or master dark)

3. flat sub

4. flat dark sub

(or bias if that is what you are using instead of flat dark)

Yep will do that in a bit.

[simmo39]

17 minutes ago, vlaiv said:

Could you post following:

1. light sub

2. dark sub (or master dark)

3. flat sub

4. flat dark sub

(or bias if that is what you are using instead of flat dark)

Flat dark sub

Master dark

Single flat sub

Single light sub

Hope they are ok, had to play around a little with them to load.

 

[vlaiv]

1 minute ago, simmo39 said:

Hope they are ok, had to play around a little with them to load.

Ah, sorry, I should have been more specific. I wanted to suggest you to upload original .fits straight from camera so I can check actual pixel values and do some statistics with the data.

Scaled down jpegs are of no use for that purpose.

[simmo39]

1 hour ago, vlaiv said:

Ah, sorry, I should have been more specific. I wanted to suggest you to upload original .fits straight from camera so I can check actual pixel values and do some statistics with the data.

Scaled down jpegs are of no use for that purpose.

yep, I tried to upload the fits but i keep getting an error, i will keep trying. I think the net is heavy with traffic.

[simmo39]

Sorry vlaiv, cant seem to get the fits to upload, PNG and jpeg works fine. 

[vlaiv]

16 minutes ago, simmo39 said:

Sorry vlaiv, cant seem to get the fits to upload, PNG and jpeg works fine. 

Maybe dropbox or some other file sharing service like we transfer? We transfer does not require login / account, won't keep files for too long, but long enough for interested parties to download. I think it's couple of days now that files are being kept for download.