Narrowband flats problems.....

[Spaced Out]

Hi All

I hope someone can help a novice and tell me where I am going wrong please.

I used flats and other calibration frames for the first time recently and got good results with a couple of LRGB images.

My latest effort is my first narrowband image to use any calibration frames and it’s all gone wrong ! I used DSS, added in the flat frames I took for each filter but the final stacked images all show vignetting and when the images are stretched this becomes really ugly all around the edges of the image.

I didn’t take any dark flats this time (I was tired at the end of the session and forgot !) and I hadn’t got round to creating the darks yet. Might this be the issue or should I be doing something different for NB filters ? Maybe different DSS settings ?

Below is an output image from the combined NB filters + flats (showing some ugly edges), and also one of the flats (Ha).

The camera is a ZWO ASI 1600, I used APT with a target ADU of 25000 for the flats.

Thanks in advance for any help you can offer.

[geordie85]

Your flats are over correcting. I've always just used my luminance flats and they've always worked fine. 

I've never taken separate flats for other filters.

[davew]

Flat frames need calibrating before use so it could be a good idea to shoot some flat darks. Just use the same settings to take the darks as were used taking the flats. I've heard the minimum time for taking bias frames with your camera is around 0.2 seconds so anything longer should be good to go.

Dave.

Edited November 5, 2020 by davew

[david_taurus83]

Is that what the flat looks like or has it been stretched? It should be an even grey image unstretched. I never had an issue with narrowband flats and I exposed to around 21k ADU with the ASI1600.

[Spaced Out]

1 hour ago, david_taurus83 said:

Is that what the flat looks like or has it been stretched? It should be an even grey image unstretched. I never had an issue with narrowband flats and I exposed to around 21k ADU with the ASI1600.

Thanks. That was a single flat frame with the Ha filter unstretched, below is the master flat.

The couple of times previously I have used flat frames with a L filter they looked similar and it worked really well. 🤔

Edited November 5, 2020 by Spaced Out

[Spaced Out]

5 hours ago, geordie85 said:

Your flats are over correcting. I've always just used my luminance flats and they've always worked fine. 

I've never taken separate flats for other filters.

Thanks. Any ideas why they would be overcorrecting ? Do you use L flats for NB too then ?

[Spaced Out]

1 hour ago, davew said:

Flat frames need calibrating before use so it could be a good idea to shoot some flat darks. Just use the same settings to take the darks as were used taking the flats. I've heard the minimum time for taking bias frames with your camera is around 0.2 seconds so anything longer should be good to go.

Dave.

Thanks, I've got new flats and dark flats now from a different night. I'll maybe try using them instead, I've not seen any obvious dust on the flats and I just want to try and correct that horrible vignetting if I can.

[geordie85]

Yeah I do.

I've no idea why they're over correcting. Maybe try aiming for a higher and lower adu, see if that makes any difference.

[MartinB]

Your flats should be fine given your ADU levels.  I'm not sure why your flats have over corrected, it is a common problem but the diagnosis can be tricky.  The first thing to try is to do a full calibration rather than just using flats.  Get some darks using the same exposure as your lights.  Do get some flat darks using the same exposure as your flats.  Make sure they are clearly specified as dark frames.  I am not familiar with DSS but it is very likely there is an "optimize darks" option.  Make sure this isn't selected as this is a notorious way to mess up calibration.  Because your exposures will all be matched you don't need any optimisation.    Adam Block suggests that CMOS flats shouldn't be too short and recommends exposures greater than 7 seconds.  I've not had problems with shorter exposures but may be worth considering if you continue to have problems.  

Will be interested to hear how you get on.

[rickwayne]

Bias frames are important for the math to work out correctly for flats. The 1600 is known to have squirrelly performance at very short exposure times so dark flats  are the way to go with that camera. Fortunately you can take the dark flats any time with a cooled camera, no need to take them at the end of the night. Just be sure to use the same exposure as your "light" flats.

As for shooting flats on just one channel, I suppose if it works, that's hard to argue with. Assuming your filters have all the dust bunnies in the same spots, that is. I don't enjoy that happy circumstance!

[vlaiv]

3 minutes ago, rickwayne said:

As for shooting flats on just one channel, I suppose if it works, that's hard to argue with. Assuming your filters have all the dust bunnies in the same spots, that is. I don't enjoy that happy circumstance!

That only works for vignetting with parfocal filters. People that do that probably keep their filters and other bits in optical train very clean (sealed filter wheels and screwed in optical train that does not get disassembled often).

[rickwayne]

And of course one astronomer's "parfocal" is another's "what the heck is this blurry stuff trying to pass itself off as an image?"

Just in case vlaiv's point is insufficiently clear, when you change focus, the vignetting changes too, invalidating flats taken with other filters.

Sadly, no matter how much I clean them and am careful with them, my filters always have griblies, and parfocal? It is to laugh. YMMV, of course.

[Spaced Out]

Hi everyone

Thanks for all of your replies, really helpful.

I added another nights subs and full calibration frames and it has improved things a lot ! 👍

I suspect my lack of calibration frames, particularly dark flats may have been the issue ?