Taking Flats with a DSLR

[PhilB]

 

[MarkS]

Mark, I'm interested in your dislike of light-boxes. I was considering making one of these, largely because in their image processing book, Richard Berry and James Burnell state "....light-box flats are by far the best for amateur astronomers..." However, your method clearly works very well. Could you put a little more flesh on the bones please?

Phil,

I've never read any books on astrophotography or processing - I prefer to develop my own theory as I go along

As far as flats go, I have discovered by experience that taking an image of the uniform zenith of the sky shortly after sunset or before dawn produces a much better flat than the 2 other methods I have tried:
a) Putting the telescope right against a TV screen (or PC monitor) displaying a uniform white screen
b) Photographing a TV screen displaying uniform white from the other side of the room.

Unless you build a sophisticated light box then the illumination won't even be uniform - so it probably performs less well than a TV screen in close up. So from worst to best I would rate the techniques as follows:

Worst (1) DIY Lightbox
       (2) TV screen close up
       (3) TV screen from a distance
Best  (4) Uniform sky

However I will admit though that a lightbox is very useful as a portable device in the field (literally!) especially when you have to make multiple flats for the multiple filters used.

So why do (1) and (2) produce the worst flats?  I think the answer is that the illumination is far too close to the scope's optics and they allow lots of "stray" light to enter the optics at all sorts of weird angles and reach the CCD in ways that would never happen in a real imaging session.

How often do I take flats for my DSLR?  Only when I notice that a particle of dust has appeared or disappeared in the frame.  In practice this only happens every few months.

Mark

[Mac]

How about ELS.

http://www.w-co.co.uk/el.php

http://uk.farnell.com/pacel/a4-blue/el-sheet-a4-blue/dp/142116

[MarkS]

How about ELS.

Thanks!  A DIY lightbox made from an ELS is as good as a TV screen close up.

One other thing I should have said is this:  if you use a small chip CCD then the method of creating a flat is not very critical.  But for DSLRs or and larger format CCDs then it becomes more critical.

Mark

[Mike]

Why would ELS not be as good as a sky?

[The Thing]

I now have an els panel and it works very well, better than the sky/diffuser combination I was using before.

[PhilB]

Thanks for you interesting reply, Mark. I shall experiment with this.

[MarkS]

Why would ELS not be as good as a sky?

As I said earlier, if the ELS is very close to the scope, I think it  allows lots of "stray" light to enter the optics at all sorts of weird angles and reach the CCD in ways that would never happen in a real imaging session.

Another way of looking at it is that using the sky, an "in-focus" image of the sky arrives at the CCD and so it has the same light distribution across the CCD as the "in-focus" sky background during a real imaging session. But using an ELS, an extremely out of focus image of the ELS arrives at the CCD so it almost certainly has a different distribution of light intensity.

To prove it mathematically would involve performing an integration of light flux at various pixel positions on the CCD.  I'll give it a go sometime.

[mickw]

To prove it mathematically would involve performing an integration of light flux at various pixel positions on the CCD

See what you've started now 

[Rocket Pooch]

Mark,

Underrstood, but it still better to use flats than not if you have dust bunnies.  

Also "How often do I take flats for my DSLR?  Only when I notice that a particle of dust has appeared or disappeared in the frame.  In practice this only happens every few months.", using your thought this means your own flats are faulty if you only take them periodically.  This is because you will never match the focus and alignment of the camera every time out in the field.  So your theory is sound but you yourself do not adhere to it, ha got ya!

Myself, again I don't take flats from here or France, no need if the signal to noise ratio is good enough, however if I was to do photometry I would do.

I'm with you Phil, thats a great book and the example in it are excellent.

Chris

[Carole]

the bright patch is not central in the CCD.  This almost certainly indicates that your camera is not sitting centrally in the light path.

I've checked the screws and the position of the camera and flattener and they are all pushed in as far as they can go, so I can't see anything not sitting centrally.  This makes me wonder if it is the telescope itself, could this be due to needing collimating as I notice my stars do not have spikes.

Carole
 

[RobertM]

If you want spikes then stick a cross of string across the front of your OTA, but warning, some people may take offence !!!

But I would say it looks likely your collimation is out.  Before you consider any nuclear option, you'll have to double check that the camera is sitting square in the focuser and theres is no sag in the focuser itself.  A tell tale sign of a camera not sitting square is out of focus stars on one side of the image.

Collimation will only need to be adjusted if it noticably affects image quality and that may not be the case here.

Robert

[mickw]

Quote
the bright patch is not central in the CCD.  This almost certainly indicates that your camera is not sitting centrally in the light path.

I suspect it might be the light source not sitting centrally on the camera

I've checked the screws and the position of the camera and flattener and they are all pushed in as far as they can go, so I can't see anything not sitting centrally.  This makes me wonder if it is the telescope itself, could this be due to needing collimating as I notice my stars do not have spikes.

Is the focuser fully extended or wound in ? - Thinking the tension may be too loose
How do you collimate a refractor ?

[PhilB]

You move  the plane of the optics so that they are perpendicular to the normal, the same as you do with a reflector. Trouble is, not all refractors have adjustable cells. But, don't panic!! Having seen your light-box, Carole, I think it likely that Micks option one is accurate. The light source is not on the optical access.

[Carole]

Phil,
I didn't use the light box for the flat in question as I hadn't made it at that stage I put a white sheet of paper over the end of the scope (in the mask that I showed you which only holds the paper in place and does not go in front of the scope) and then pointed it at the daytime sky, so no chance of the light path not being directed at the scope.  

If you want spikes then stick a cross of string across the front of your OTA

I don't particularly want spikes I just wondered whether that was another sign of the collimation being out as other people get spikes on the bright stars automatically and I don't.  

Is the focuser fully extended or wound in ? - Thinking the tension may be too loose

The focuser is currently sitting at my "focus" position, and I have no idea how to tell whether the focusser has sag.  Just wondering whether locking the draw tube (or not) might alter the light path fractionally.  

What I will do is a Star test next time I am out and see what that shows both with draw tube locked and not locked and see if there is any difference.  Though what I will do it it's out when it's locked I don't know as I need to lock it to keep focus from slipping as I have experienced in the past.

Carole