M81 weird flats

[AJE]

Morning all,

Having major problems with flats.

I've taken nearly 6 hours of data in 5 and 10 minute subs. But the flats are letting me down

They make the image far worse that without them, there is some nice data there, but without decent flats I can only get so far...

Anyone fancy having a go?

This link is a RAR file containing all the Lights, Bias and several different sets of flats. It's just over 500 Mb >>HERE<<

I would really appreciate any help, suggestions, instructions...

Thanks a lot

Ant

[RobH]

Hi Ant....not downloaded your file as my net connection is snail like!

What camera are you using and have you ever had trouble before?

Rob

[AJE]

Hi Rob,

The camera is an ED80, H9 and SX Filterwheel. UVIR (clear) filter used.

Yes I've had a love/hate relationship with flats - the odd thing is that it seems to depend on the object. I generally don't have problems with Nebulea, but Galaxies cause me no end of grief - which is almost always flats related.

The night in question was very dewey and I'm not convinced that my dew control is working that well. Maybe I'll give it another go.

I have had this issue before, so I have taken to taking several sets of flats each session, each at a different ADU level (generally 15k, 20k and 20k) and seeing which works best as it seems to differ from night to night.

Snail like - tell me about it - the file took nearly 5 hours to upload, that's what you get for living in the country

Cheers

Ant

[cgarry]

Hi Ant,

What is the issue you see when you use flats?

Chris

[AJE]

SteveL has just sent me a message that it's the lights that are the problem

Maybe I'm having more of a problem with DEW than I thought. Fairly evident in this movie - I have to keep going out and drying the lens... I did have a dew strap on the camera, but I think it stopped working

kXVDkhdTyY4

Chris, from what Steve says it seems that as my FLATS were taken at the end of the session, the objective was at it's worse and they are over correcting. One thing that he did was to look at the lights and compare the first and last, there was a gradual build up of dew over the night...

Shame really!

Cheers

ant

[lukebl]

Can you post a jpeg showing the problem. I think the prospect of downloading a 500MB file might put people off!

[cgarry]

Chris, from what Steve says it seems that as my FLATS were taken at the end of the session, the objective was at it's worse and they are over correcting. One thing that he did was to look at the lights and compare the first and last, there was a gradual build up of dew over the night...

Shame really!

Thanks for that Ant. The reason for my interest is I have had huge problems with over correcting flats. I actually blamed the gradual build up of dew myself, but it turned out not to be the cause in my case.

Until I figure out what is wrong with my flats I do have a work around that allows the my dodgy flats to work. This involves adding an offset to every pixel of the master flat to prevent the over correcting.

Cheers,

Chris

[RobH]

This involves adding an offset to every pixel of the master flat to prevent the over correcting.

Cheers,

Chris

How do you do that Chris?....I'm having big problems with overcorrection on my H18 flats although I've been shooting at around 25000 ADU, but when I looked up the full well depth of the kodak 8300 chip the other day, I was shocked to find that it's only 25500, so I suspect that my flats have been much too bright, and should really be shot at around 8000ADU

Ant.....with an H9 you should be shooting flats at around 20-25000 ADU.

Re. Dew. Use a dewheater and also make a long dewshield from camping mat material....I often just use a long dewshield only on my 80mm scope (1 ft long) and even on really dewy nights, this tends to keep it clear for most of the night.

Rob.

[cgarry]

Hi Ant,

I have to agree with SteveL completely, your lights do show very obvious signs of dew building up. By the end they actually have what looks to be negative vignetting, where the outside of the image is brighter than the centre.

Just for fun, here is your first and last exposures with the same histogram stretch.

First sub:

Last sub:

One issue I have had in the past when comparing exposures like this is the tools do an auto-stretch which can hide the differences between the images. You need to be sure the same stretch has been applied to both images.

Look on the bright side, you are now able to free up over 500M of hard drive space...

Cheers,

Chris

[AJE]

Look on the bright side, you are now able to free up over 500M of hard drive space...

Cheers,

Chris

Chris, Mmmm hardly a bright side

In an attempt to try and rescue the imaging session I took a load more flats the next day. So it's actually 1.1Gb of HDD space

Rob, camping mat ordered. I also have to look at the dew strap as I'm pretty certain that isn't working

Cheers

Ant

[cgarry]

How do you do that Chris?....I'm having big problems with overcorrection on my H18 flats although I've been shooting at around 25000 ADU, but when I looked up the full well depth of the kodak 8300 chip the other day, I was shocked to find that it's only 25500, so I suspect that my flats have been much too bright, and should really be shot at around 8000ADU

Okay these are the steps I take to rescue my flats. I use Astroart 5, but I guess the idea would work with other tools.

* Do a normal stack with all the lights, darks, flats and bias frame as usual. For me this results in an over corrected image and master calibration frames being generated.

* Take the generated master flat and use the statistics function to get the maximum ADU value in it (MAX_ADU).

* Calculate the maximum offset that can be applied to the master flat without clipping with: MAX_OFFSET = 65535 - MAX_ADU

* Add this offset to your master flat (Arithmetic->Add Offset in AA5) to generate a new mater flat, save this new master flat.

* Do a new stack with all the lights, darks and the new master flat. This usually results in an under corrected image.

* Using an iterative process try different master flats by adding different offsets to the original master flat. Offsets can be between 0 and MAX_ADU, with offsets nearer to 0 causing over correction and offsets nearer to MAX_ADU causing under correction.

This is all a bit of a faff, but it does seem to work for me. I would much rather figure out what is causing the over correction though!

Cheers,

Chris