Alignment tests?

After my 17-55 took it's rather rough ride back from Canon loose in the box the focus does seem to be doing a better job, however I am concerned that something less detectable might still be wrong, such as some other elements of the lens shifting. Are there any tests I should do with the lens to make sure that everything else is ok with the optical alignment?

I could send it back and probably get Canon to pay to get it there (shipping was not cheap the first time around) since they didn't pack it properly and have it checked out, but I'd rather not go through the hassle if there are a few things I can just do at home to verify that it is working correctly.

The facetious answer would be, take pictures with it and see how they look. Other than doing the standard kinds of lens focus tests (focus charts, batteries, or whatever) throughout the zoom range, shooting flat field subjects with high detail to make sure it's not focusing asymmetrically, testing from f/2.8 through f/22 or so to make sure the aperture control is working properly, and slow shutter speed testing to make sure the IS is working OK, nothing obvious springs to mind other than take pictures with it and see how they look.

stsva wrote in post #14007923
The facetious answer would be, take pictures with it and see how they look. Other than doing the standard kinds of lens focus tests (focus charts, batteries, or whatever) throughout the zoom range, shooting flat field subjects with high detail to make sure it's not focusing asymmetrically, testing from f/2.8 through f/22 or so to make sure the aperture control is working properly, and slow shutter speed testing to make sure the IS is working OK, nothing obvious springs to mind other than take pictures with it and see how they look.

Kinda what I figured. The focus I have tested quite a bit since that is what it went in for.

I got one shot outside that I am really not thrilled with, but I don't think it's the lens so much as me for that one, kind of a shot into the sun wide open and it got some wicked CA going on with some backlit trees, but I just snapped the shot walking.

Keyan wrote in post #14008010
Kinda what I figured. The focus I have tested quite a bit since that is what it went in for.

I got one shot outside that I am really not thrilled with, but I don't think it's the lens so much as me for that one, kind of a shot into the sun wide open and it got some wicked CA going on with some backlit trees, but I just snapped the shot walking.

I don't think there are very many lenses, no matter how good, that will handle that kind of shot well.

Trees and grass are good subjects, but not in looking for CA on backlit ones.

I have an alignment issue with my 17-40 and it was quite obvious.

I would first shoot a wall. Focus on the center using LV (with tripod) and simply zoom in to the left side. Then move the focus ring and see if you can 'substantially' improve the sharpness. Then check the left side and the top and bottom. Do this at both ends of the zoom and in the middle.

You could also check the corners, but that can be more confusing because a lot of lenses have some issues in the corners and I wouldn't do it - you'll just end up convincing yourself that something is wrong, even if it isn't.

Unless you can 'substantially' improve the sharpness as seen in x10 LV, I would not bother sending it back. There is always going to be some tolerances and IMO (and experience) going overboard is easy to do.

One other check I would do is to shoot some trees at f2.8-f7 and check that one side is not obviously softer than the other. This should be very, very obvious. If in doubt, it's probably fine.

Yeah, I think it is and I am pixel peeping being paranoid because of the situation.

It is silly sharp now and the focus hits reliably.

Yeah I knew that shot was not a good test at all Wide open into the sun was not exactly a great plan on my part.

They should give you some time - at least a couple of weeks - to check it out. A little paranoia in not mad, but you need to know when to stop. you only just got the lens back and I'd be checking it, even if it looked perfect. I had to send by 17-40 back in after the first adjustment.

Rather than just looking to see if something is sharp, or not sharp, you definitely need to compare one side of the image with the other. Only is one side, or top vs. bottom are substantially different would I see a problem.

With my 17-40 for instance I did the wall test and it was obvious in LV that if I focused more towards infinity the right side became significantly sharper and if I focused more towards zero, the right side became sharper.

Also I could easily see the issue in landscape shots, even at f8 where trees on one side were noticeably less sharp than trees on the other side or the frame.