It seems that most CA Hinge failures are due to breaking rather than pulling out.

But both problems are usually linked to the same cause.

To illustrate the problem, first imagine this:

You have cut the slots, and put the elevator and hinges in place, but you have not glued anything yet. Then you push the elevator against the stab to remove any gap.

Now... with one finger, push the elevator down just past where it would normally deflect giving "Full Down" - Then, do the same in the UP direction.

Now look at the gap

Is there a gap? 99% of the time there will be, yet if you raise and lower the elevator again it won't grow any larger.

The reason for this is that even though you (Or the Manufacturer) has sanded a bevel on the LE of the elevator, it's not perfect and the elevator is pushing itself away from the stab.

To over emphasize this point, look at the picture below. You can see how the imperfect bevel pushes away from the stab. Granted, the lack of bevel in the picture is to over state the point, but it still happens on a smaller scale (Note: There's little or no point in trying to make your bevels perfect for too many reasons to list here, just accept the fact that you must work with this principle)

Ok, so let's say you want your model to be perfect. You cut the smallest hole possible in your cowl - Only to discover that your engine overheats cause it doesn't get proper cooling.

And you keep your hinge gaps as small as possible - Only to discover that your hinges are either breaking or pulling out

Now let's refer to the above scenario:

Every time the Surface flexes, it tries to push away from its mating surface. So:

A) If the hinges are not glued properly, they will pull out.

B) If the hinges ARE glued properly, they will break

This is why it is imperative to flex the control surface MORE THAN IT WILL MOVE IN FLIGHT - and then NOT close the gap before gluing