dan hilts said:
In Ed Moreno's review of the Celestron 80ED he states "the collimation error is WELL within the 3 arc-minute limit required for high-contrast viewing."
This limit is something I have never heard of before. Can someone fill me in? Where does the limit come from? Any references to literature I can obtain?
If your telescope isn't collimated, it's akin to it having a broad error in the optical surface--except, of course, that you can fix the error in this case simply by collimating it. The extent of the error increases with the amount of miscollimation.
The reason you haven't heard of this limit before, most likely, is that it depends on the focal ratio; the faster the telescope, the better the collimation must be. So it's not always 3 arcminutes; for a slower telescope, it can be much higher. You can read a bit about this in Daniel Rickey's article in the June 2006 issue of Sky and Telescope
But in my opinion, it's not a very interesting rule of thumb, since as far as I know, there's no simple way to measure the angular extent of your miscollimation. I suspect it's there in the review mostly to impress upon us in a quasi-quantitative way the accuracy of the manufacturing's optical and mechanical alignment. That's the sort of thing I could do without in my telescope review. Anyway, I would read that sentence basically as, "The telescope was well collimated."