Re: ATM CCDs, star size & seeing

[Al Kelly <akelly@ghg.net>]

>Stuart Field wrote:

>Here's something that's been bothering me - why is it that bright stars appear
>larger in CCD images? ....<snip>....

>Now it might be that seeing, integrated over a few minutes, is smearing things
>out.  But then I would ask why big scopes (i.e. 16 - 24") give better deep-sky
>pix than small ones.  Let's consider the images due to a 6" and a 24" f/6
>scope.  Both have the same size diffraction spot, about 2.44 lambda = 1.3 um
>diameter which would seem *much* smaller than a typically CCD pixel (e.g. 8 um
>for "Cookbook 245")....<snip>...

Primarily, you have answered your own question with the reference to seeing
effects, but you have been given some bad data:  the pixel size with my
CB245 used in its default 378x242 binning mode is 18 x 19.7 microns.
Certainly this is much larger than the diffraction-limited resolution of the
scopes I use it with, but the fact is that I take exposures up to a minute
or so in duration, so the limiting resolution factors are seeing, pixel
size, and focal length, not aperture.  Also I stack numerous of these
relatively short exposures to achieve the high SNR necessary for "deep"
imaging, so seeing effects from my imaging sites assure star images usually
from 2 to 4 arcseconds in size, depending on the night's seeing.

Larger scopes tend to have longer focal lengths, so this may be skewing your
perception of the situation.  With my CB245 and C-8 at f5.6 (1143mm focal
length), my system resolution is about 3.5 arcseconds per pixel, so I could
not split a 2" double on the best of nights; however, with my CB245 and 32"
f4 Newt (3175mm focal length), my system resolution is about 1.2" per pixel,
splitting that same double on a very good night.  

Al K.
--------------------------------------------------------------------
Al Kelly .... e-mail: akelly@ghg.net 
         .... web site: http://www.ghg.net/akelly/