RE: Focus indication

[Tom Droege <tdroege@veriomail.com>]

OK, I was wrong to call this a focus project.  It is much more than 
that.  Picking a few stars and getting a mean psf tells you a lot.

Lett's say Rob figures out some way to do this.  I know it can be done 
because Mike G. does it in Image Scientist, and I assume IRAF does it and 
many other similar programs.

So let's suppose we get numbers for the two cameras and derive the NS and 
EW components.  These would then appear under the pictures in Mondown.  Now 
give me limits to set for the four numbers so that an alarm goes off if the 
values move beyond my set limits.  Note that one of the answers that such a 
program should give is "I can't find a psf".

The alarm goes off.  By looking at the numbers I can tell:

1) Clouds have moved in.
2) The V lens is fogged.
3) The I lens is fogged
4) The RA Drive has hung.
5) A shutter is stuck.
6) There is an electronic problem
7) Something has moved the mount
8) A light has been left on somewhere

And many other things.  Presently I bring up an image and look at it.  I 
can't do this automatically so often long periods of running are lost 
because something goes wrong that I could fix - but I don't know anything 
needs fixing.  OK, I could not sort out everything in the above list.  I 
might just know that something is wrong.   So a bad psf is a "Not Good" 
indication.  But that is what is useful.

Tom Droege



At 07:01 PM 8/6/01 -0700, you wrote:
>Chris,
>   We use a similar technique with the PMM scanner since
>it is taking an image of a photographic plate.  Since photographic
>plates contain exposed 'grains', the optimal focus is the
>one with the highest spatial frequency content.
>   However, there are a few reasons why this technique may
>not work or at least work optimally for astronomical imaging.
>First, even if the technique works, you have no indication of
>which way to move focus, which is the advantage of the focus
>plate technique.  Second, multiple frames must be taken, each
>with its 40 second readout.  Finally, most of the pixels just
>contain noise, so your technique won't necessarily optimize focus
>on the (relatively) few real stars.  This is different
>from the daylight photo situation.  Focus with the Mark IV is done
>once in a blue moon since Tom has shown there is little
>temperature dependence and once the systems are working you
>are unlikely to knock something.
>Arne