My flats are over calibrating?

[pipnina]

At first, with my orion image, I noticed it was suffering from vignetting. Since I intend to make a better version this winter I figured I should try and fix that by learning to make flat files. I attempted it at first with my camera focussed at infinity, at the same f/ratio and focal length as the original but at a much lower ISO (100 instead of 1600) However, when I put them through DSS they have made the edges of the image brighter than the middle!

Not sure what I did wrong, anyone know what could be wrong?

DSS output: http://i.imgur.com/bwrDDY8.jpg

Help appreciated!

    ~pip

[StargeezerTim]

Hi... I always thought flats should be the same ISO as the light frames. Worth a try.

[lensman57]

At first, with my orion image, I noticed it was suffering from vignetting. Since I intend to make a better version this winter I figured I should try and fix that by learning to make flat files. I attempted it at first with my camera focussed at infinity, at the same f/ratio and focal length as the original but at a much lower ISO (100 instead of 1600) However, when I put them through DSS they have made the edges of the image brighter than the middle!

Not sure what I did wrong, anyone know what could be wrong?

DSS output: http://i.imgur.com/bwrDDY8.jpg

Help appreciated!

    ~pip

Flats should be taken at the end of the imaging session without disturbing the orientation of the scope/lens and at the same focus position. Do not alter the ISO this is important . I doubt very much that your present set of flats are a match for you old image of Orion Constellation.

A.G

[carastro]

I always used to use 100iso for flats and AV and they worked just fine, but then I never did widefield imaging at that time.

I had read up somewhere that you DON'T need the same ISO for flats as for the imaging subs and that you should use 100iso and that is what I used to use throughout my entire time imaging with a DSLR and never had any problems with overcompensation and my flats worked just fine.  

Carole 

[StargeezerTim]

Flats should be taken at the end of the imaging session without disturbing the orientation of the scope/lens and at the same focus position. Do not alter the ISO this is important . I doubt very much that your present set of flats are a match for you old image of Orion Constellation.

A.G

I always used to use 100iso for flats and AV and they worked just fine, but then I never did widefield imaging at that time.

I had read up somewhere that you DON'T need the same ISO for flats as for the imaging subs and that you should use 100iso and that is what I used to use throughout my entire time imaging with a DSLR and never had any problems with overcompensation and my flats worked just fine.  

Carole 

Ha ha.....  I suppose its worth trying both ways...suck it and see and see what works best...

[brrttpaul]

interested in this post as I read somewhere also to reduce ISO for flats,

[lensman57]

Ha ha.....  I suppose its worth trying both ways...suck it and see and see what works best...

You must take the Flats at the same ISO as your Light and Bias frames. The  first step in processing the stack is the subtraction of Master Bias from the individual Flat frames, these are then stacked to make a Master Flat. If these are not matched in ISO the resulting Master Flat will not be a match to the Lights. ISO setting in a DSLR is nothing but a gain value and by altering the gain the level of the noise is altered. It is best to take all the calibration frames at the same ISO setting so the pedestal noise values match. 

A.G

[michael8554]

I too have read that flats should be taken at low ISO, so as not to add more noise to the final image - hmmm.

Also possibly to get avoid over-exposing with a very sensitive camera at high ISO?

Michael

[pipnina]

You must take the Flats at the same ISO as your Light and Bias frames. The  first step in processing the stack is the subtraction of Master Bias from the individual Flat frames, these are then stacked to make a Master Flat. If these are not matched in ISO the resulting Master Flat will not be a match to the Lights. ISO setting in a DSLR is nothing but a gain value and by altering the gain the level of the noise is altered. It is best to take all the calibration frames at the same ISO setting so the pedestal noise values match. 

A.G

I did think, perhaps since ISO 100 has 12+stops of dynamic range and 1600ISO has more like 8 stops, the flats would be on different scales. But surely if it's less contrasty, it would be under-calibrating?

It might be that I had my histogram higher than the middle? I wanted to reduce the noise in the flats as much as possible...

[glowingturnip]

I think lensman has it right - if your flats aren't at the same ISO as your bias, then you would need to be taking an extra set of flat-darks at the same ISO as your flats and use those to subtract the noise from the flats, instead of bias-subtracting the flats.

If you've done that though, then I don't see any other reason why your calibrated flats would need to be at the same ISO as anything else - the flats are calibrating the optical train, whereas darks and bias are calibrating the imaging system.

[ollypenrice]

Flats over correct if they are not dark calibrated. The only dark you need for this is a bias which will do the job fine. DSS needs to know that it is to use this copy of the bias as a dark for flats, though. I know nothing about the gain settings but I'd have thought that the bias used as a dark would have to be at the same setting as the flats. I can't see why the images and flats would need the same setting but maybe they do.

Olly