Question about flats

[Geoff45 (Geoff)]

There are a number of ways of taking flats, but as far as I can see, they all offer the opportunity for the flat not the be a uniform grey. Twilight flats for eample might show a slight red cast and DIY lightboxes could have any sort of colour cast, depending on the light source used.

Two questions:
Is this a problem?
If it is, does it help to convert all flats to grey level before calibration?
Geoff

[Terry B]

Mostly the colour of the flat is unimportant unless it is taken through different coloured filters then flats for each filter should be made.
Flat fielding should be done before debayering just after dark subtraction. In essence they are black and white images at this point.

[Geoff45 (Geoff)]

Originally Posted by Terry B

Mostly the colour of the flat is unimportant unless it is taken through different coloured filters then flats for each filter should be made.
Flat fielding should be done before debayering just after dark subtraction. In essence they are black and white images at this point.

Yes, that makes sense.
Geoff

[Tandum (Robin)]

IMHO, any flat is better than no flat and I note DSS makes a master flat in grey scale anyway.

[Geoff45 (Geoff)]

Originally Posted by Tandum

IMHO, any flat is better than no flat and I note DSS makes a master flat in grey scale anyway.

Yes, I have investigated a bit further and is seems that many of the astro software programs (eg ImagesPlus) automatically do this in their calibration routine.
Geoff

[leon]

That right Geoff, the actual different shades of blue or whatever have no effect on the final outcome when used in Image Plus for example, it is automatically done in the calibration process.

However if if you like you can convert the flat to gray scale manually by using the Gamma and Luminance in Image Plus.

Leon

[Hagar (Doug)]

Hi Geoff, Having just done a lot of research into flat framing I can say a lot but in most cases the software used to, firstly calibrate the flat images and compile the images into a master flat, taes care of the colour cast in the original images and inevitably converts them to a greyscale image.
Programs such as Nebulosity actually take care of this and run a smothing filter over the final image.
I use a QHY8 and now an Electro Luminecent sheet to create flat frame masters and have found the need to also include Bias frames in the process to get the very best results with this camera. I do all calibration and master flat creation in ImagesPlus and find the process a very simple one.
Like all processing this is all black magic to me but the end result of even ordinary images is greatly enhanced with the use of flats and bias files. Experimentation seems to be the way forward so give it a go and I think you will be pleasantly surprised by the final results.

[multiweb (Marc)]

My EL sheet generates a blue/white light. I shoot 8s flats with the QHY8 (20 of them) and aim for approx. 18000 ADU on the red pixels. I then combine them in a master (median/sigma reject) in CCD Stack and subtract a matching master dark (8s). Seems to work for now. Still testing.
I tried to debayer first then treat each channel separately during calibration. There was no significant improvement and it was more work. So calibration should be done with bayer matrix still in the FITS. Good Flat fielding proved to be the most complicated thing I came across in astro photo so far. I really struggled to get it right and still am.

[Hagar (Doug)]

Originally Posted by multiweb

My EL sheet generates a blue/white light. I shoot 8s flats with the QHY8 (20 of them) and aim for approx. 18000 ADU on the red pixels. I then combine them in a master (median/sigma reject) in CCD Stack and subtract a matching master dark (8s). Seems to work for now. Still testing.
I tried to debayer first then treat each channel separately during calibration. There was no significant improvement and it was more work. So calibration should be done with bayer matrix still in the FITS. Good Flat fielding proved to be the most complicated thing I came across in astro photo so far. I really struggled to get it right and still am.

Hi Mark, Something that has come out of a lot of research is that full well for the QHY8 is about 25000 e. I have been experimenting with 18000, 25000 and 9000e max and with taking a series of Bias frames at 0.002sec.I then put them all together in IP3 and have found the best results with a max of 9000e. I am still struggling with flats and my QHY8 but have found that without a bias or dark to help calibrate I might as well not bother with them. I use Nebulosity 2 for capture with Gain at 1% and offset set to 65.

[multiweb (Marc)]

Originally Posted by Hagar

Hi Mark, Something that has come out of a lot of research is that full well for the QHY8 is about 25000 e. I have been experimenting with 18000, 25000 and 9000e max and with taking a series of Bias frames at 0.002sec.I then put them all together in IP3 and have found the best results with a max of 9000e. I am still struggling with flats and my QHY8 but have found that without a bias or dark to help calibrate I might as well not bother with them. I use Nebulosity 2 for capture with Gain at 1% and offset set to 65.

Thanks for all the info. I'll keep these levels in mind and try out. Back to the drawing board