Taking Flats with a DSLR

[RobertM]

As a slight aside, most small professional observatories get away with dome flats rather than some light source shining directly down the tube and this seems to work fairly well.  Evenly illuminating a flat surface is also easier though portability may be an issue 

For the lens I've found shining a diffuse light on to a white sheet on the side of the observatory (aka dome flats) works very well.  It works specially well if the light source is far from the sheet to blurr any imperfections.  If there are no dust bunnies then synthetic flats can do a perfect job especially if there are no strange gradients and that's what I tend to use for narrow band.

Robert

[mickw]

when you are imaging a nebula or other very remote object the photons are travelling parallel to the optical tube

But other photons (not from the target) are approaching at all angles.  Admittedly they will not have a direct path to the imaging device but they are still there.

Taking flats against a dawn sky etc. also has the danger of introducing unwanted artefacts such as gradients across the FOV due to moisture in the air etc.

I agree that using a lighbox you are imaging an out of focus lightbox.
Similarly using the sky, you are imaging an in focus variable/uncertainty.

I would have expected that a lightbox only sufficient intensity to illuminate any debris would be a better option, at least it's controllable and consistent (hopefully)
I know the maths say different.

[Carole]

Phil's quote

I think I might be starting to see the light

Pun 

BTW I am not going to say HIJACK ALERT!!! as this has invoked an interesting discussion.

Carole

[Rocket Pooch]

Mark,

What you have actually done is move the focus point of the scope by moving the light source, this is normally caused when the light is not evenly illuminated.

Also a FR with one or not hase nothing to do with the quality of the flat, the fat should include the whole image train including, scope, FR, filters, camera, all at the same settings as per the origional image.  What you have done is prove if you focus and move the focus you get odd results.

Hence my earlier comment about your camera flats, they are technically flawed.

I'm with Phil on this, all your need is an evenly illuminated flat, without the image train changing anf bingo it works, and I can prove this using maths from the flats I have used with some om older images where alfer applying my flats the image is somewhere in the region of 99.998% flat, thats good enough for NASA its good enough for me.

AND YOUR NUTS thinking about maths when cycling in london, god someone could have stepped off a bus in front of you.

Chris


P.S. Taxi for one!

[MarkS]

Mark, I think I might be starting to see the light. However, it's the first image in your experiment, where there are in-focus stripes, that worries me. When taking a flat the telescope would be focused on infinity and nothing placed at close range would be in-focus. This being so, I not sure that moving the image plane through what is a relatively short distance would have any noticeable effect on the out of focus image.

Yes, the scope is focused at infinity but, yes, it still produces an image of the light box only a foot or so behind the usual imaging plane.  Strange, but true.  The experiment demonstrates a fact that is otherwise quite unintuitive.  The experiment also demonstrates the very noticeable effect of moving the that image plane.

[Carole]

god someone could have stepped off a bus in front of you.

Chris could you please DROP THIS I am getting fed up with you keep having a dig every so often and it's not funny.  I was badly injured by an idiot cyclist, it's not something to poke fun at and keep bringing up every so often.  I find your remarks offensive and have so far bitten my tongue rather than have a public argument but since you persist in doing this you leave me no choice. 

Carole

[MarkS]

At the end of the day the important thing  is to produce a flat that is "good enough".
I readily admit that in most cases a light box is good enough for most practical purposes - for instance, it works very well for my ED80 (a Refractor) and for my camera lenses.

However in my experience it did not work well for the C11 (an SCT) nor for the Tak (a Newtonian), so sky flats were required.  The experiment demonstrates why this is the case.

Mark

[Carole]

Mark, I have done another set of flats which seem to have worked out OK using a mask over the aperture.  The mask has a piece of white typing paper on the front and I can point this at the sky whether or not it is cloudy and raise and lower the ISO according to how bright it is.

I think this will work better for me than a light box and as I don't have to change filters during an imaging session as I am using a DSLR I think I am going to stick to this method, I can always do them the next day if necessary now I can leave things set up.

I still have the slightly off centre problem, and I will give this a try without having the drawtube extended just to see whether focus sag is a problem. 

Carole

[Rocket Pooch]

Thats really odd if its happening on the SCT and Tak, my Vixen was ok, hmm more test needed me thinks.

Chris

[Rocket Pooch]

god someone could have stepped off a bus in front of you.

I find your remarks offensive and have so far bitten my tongue rather than have a public argument but since you persist in doing this you leave me no choice. 

Carole

Hi,

Your right, that subject is probably a little long in the tooth, I'll park it.  Also to be honest you don't need to bite your tongue either I'm a complete insensitive git and I know it.

If you ever fancy a public argument I'm game, maybe not tonight, I don't think the membership would be too impressed.  Maybe next DSC.

Chris

[Carole]

If you ever fancy a public argument I'm game, maybe not tonight, I don't think the membership would be too impressed.  Maybe next DSC. 

No thanks I prefer a quiet life.

Carole

[Mac]

Fight Fight Fight.

[mickw]

Behave, all of you..............

Otherwise there'll be some slapped legs   

[Ian]

Mark,

I've had a good think and I'm not sure either

1. you're right and I don't get it
2. you're wrong

My issue is understanding the physics of the difference between a focussed image of a diffuse light source, and an out of focus image of an diffuse light source.

I shall think about it more when I'm in the pool tomorrow. Do you fancy posting your maths, see if I can follow it?

[MarkS]

Ian,

The maths only exist in my head at the moment.  I'll spew it out onto paper as explanatory diagrams at the weekend.  Here's another attempt to explain the same thing.

In summary:
The focal reducer is the heart of this problem: the focal reducer screws up your flats.  If you are not using a focal reducer there is no problem.

In more detail:
With your scope and focal reducer combination focused to infinity it will produce focused stars on your CCD imager.  But at the same time (don't touch the scope focuser!) it will also produce an image of any object you stick immediately in front of your scope.  But the focal plane of this image is a foot or so behind your CCD imager  (This is what my experiment demonstrated). You can calculate exactly where it appears by using the lens formula and the focal length of the reducer. So what appears on your CCD imager is an out of focus image of this object.

Now if you move this object slightly further from the scope (don't touch the scope focuser!) - the focal plane of this image moves and what appears on your CCD imager is a differently de-focused image.

So if the object in question is an illuminated sheet, we have proved that moving this sheet changes the shape of the blur you record as a flat at your CCD.  So one or both are not the "true" flat you require to correct your optical system.

Mark