It is clear from the preceeding pages that diffraction is my preferred focus technique. A few other points may be worth mentioning.
1) If you are going to use either Hartmann or diffraction focus, you will need to use a fairly bright star for focus. When this is coupled with thermal focus shift it spells GOTO. Without GOTO it may be very hard to get from your focus star and set up to image your object within an acceptable amount of time and temperature change. Even with GOTO, it is usually best to use a bright star as close as you can to your object. I have found that the brightest star in the constellation is usually adequate.
2) Some systems may have so much mirror flop that going very far in the sky to a focus star may cause more problems than it solves. Those poor folks either need to fix the flop (preferred) or stick to a focus method that can use a less bright field star. This is a question of the lesser of two evils, flop or a less than ideal focus method.
3) The most important thing of all is to characterize the behavior of your system. If you know your depth of focus, your step size, and your thermal and mechanical behavior, you will be in a position to control your focus precisely.
4) In the final analysis it is what works for the person doing the imaging. These views are based on my testing and experience, but you will need to decide for yourself what works for you and with your equipment.
5) If you are serious about imaging you may want to look into an HFD based automated focus system. These are accurate, scriptable, and just plain easier. I would probably learn to focus the "old fashioned way" too, however. It will serve as a check and a backup for the automated methods.- Bill McLaughlin, January, 2002 -