Sorry to hear about your problems, Ian.
Just to keep a sense of perspective on this though, in all of the thousands of postings I've read on the various internet forums since the 40D was released (including DPReview, POTN, Photo.net, Fred Miranda and more - pretty much all the biggest photography sites)
the only two I've read that directly quote a mirror as the source of a 40D failure are yours and Paul's.
Just Google "40D + mirror": plenty of discusson about mirror lock-up, but not much indication
if any of mirrors blowing up.
The point being, a "known" fault does not mean a
common fault (I know you aren't saying this, Ian, but others are - without much basis for doing so, in my humble opinion).
Karl and Rob - no offence lads, but you're seeing "issues" that just aren't there. This
isn't a common problem. Yes, there are some bodies out there with problems, but I would say that generally, as far as the 40D is concerned, the available evidence indicates that Canon have
very much "got their act together".
Hell (
Paul to note! ), I had two
30D shutters fail on me - one camera at only 5,000 actuations, the other barely out of the shop.
It's another "known fault", but that doesn't mean that the 30D is inherently prone to shutter failures just because it happened twice to me. The first was fixed under warranty, the second replaced by the shop, and both are perfect now.
Things break, and sh*t happens.
This experience didn't put me off subsequently buying a 40D, either.
Of course, the 40D's mirror assembly is a faster, more sophisticated mechanism than was in the 20D/30D, with motors instead of springs to move the mirror - but it's (allegedly) a more heavy-duty item to cope with the higher fps rate.
Maybe the additional sophistication increases the theoretical risk of failure, but it still appears to be a
truly rare problem.