Ad
Ad
Ad
Pages: [1]   Bottom of Page
Print
Author Topic: Canon 1DS MKIII and 24-70 Lens--dropped :(  (Read 3444 times)
dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« on: September 16, 2012, 06:58:39 PM »
ReplyReply

It finally happened. I dropped my 1DS MKIII and 24-70 about 2.5 feet onto a wood floor, not concrete, luckily.

I didn't notice anything wrong with it, but today while shooting in San Francisco, I noticed the manual focus ring was really stiff on the 24-70.

Sure enough, the ring that holds the hood is bent making it hard to turn the manual ring. Auto focus is still fine.

Now I'm wondering what all I haven't noticed inside the lens and body that are tweaked from the fall. I'm concerned the alignment of mechanical parts in the lens and the mirror in the body might be compromised.

If I send it in for repair, what should I request, other than the bent hood ring on the lens, in order to make perfectly sure nothing is wrong with either the lens nor the camera when I get them both back?

What preliminary tests can I do to see if there is any obvious problems?

As always, thanks much.
Logged

Craig Lamson
Sr. Member
****
Offline Offline

Posts: 764



WWW
« Reply #1 on: September 16, 2012, 07:22:45 PM »
ReplyReply

It finally happened. I dropped my 1DS MKIII and 24-70 about 2.5 feet onto a wood floor, not concrete, luckily.

I didn't notice anything wrong with it, but today while shooting in San Francisco, I noticed the manual focus ring was really stiff on the 24-70.

Sure enough, the ring that holds the hood is bent making it hard to turn the manual ring. Auto focus is still fine.

Now I'm wondering what all I haven't noticed inside the lens and body that are tweaked from the fall. I'm concerned the alignment of mechanical parts in the lens and the mirror in the body might be compromised.

If I send it in for repair, what should I request, other than the bent hood ring on the lens, in order to make perfectly sure nothing is wrong with either the lens nor the camera when I get them both back?

What preliminary tests can I do to see if there is any obvious problems?

As always, thanks much.

I feel your pain, tripped over a tether cord and pulled over a tripod with a 1DsII and 24-70.  Smashed to bits, unrepairable.  Insurance replaced both, the camera with a MKIII.  I hope you have coverage.

Tell repair what happened, let them look to see what they find.
Logged

Craig Lamson Photo
www.craiglamson.com
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3553


« Reply #2 on: September 16, 2012, 08:26:12 PM »
ReplyReply

What preliminary tests can I do to see if there is any obvious problems?

Hi,

I feel sorry for your unfortune. It revived my memory of an EOS-1 series DSLR + 200mm lens crashing on a stone tile floor from 5 feet, together with the coat hanger rack I hung it on ... The only effect I noticed was a dented lens cap.

If you want to verify for decentered lens elements, besides the apparently obvious manual focus ring deformation, then you should shoot a resolution test target that allows to detect/quantify such aberrations. My target allows to determine both analytically and visually how much the lens was 'impacted'.

Cheers,
Bart
Logged
dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #3 on: September 16, 2012, 09:05:26 PM »
ReplyReply

Thanks Bart. It wasn't the manual focus ring itself. It's the ring that the hood screws onto that got not bent, really, but moved or pulled out a little. So it's crooked and rubs the manual focus ring making it tight. I think the brunt of the force was absorbed by the lens and hood, for that's what broke the fall. I have a RSS bracket on the camera and i "think" it hit the L bracket not the body.

I can't find any dents or paint chips anywhere on the body. And also, it hit a wooden floor on the second story, which is also wood under it, not concrete. So it didn't hit a completely solid object. I stopped the floor and it did give a little. Still, these are very technical pieces of equipment.

Also I'm not noticing any binding when the lens element pulls in and out from 24-70 either.
Logged

Paul2660
Sr. Member
****
Offline Offline

Posts: 1678


WWW
« Reply #4 on: September 16, 2012, 09:28:50 PM »
ReplyReply

The ring in question is also know as the "name ring" since the Canon logo is on it.  On early 24-70's this ring (mine included) had a problem of breaking loose and needing to be returned to Canon.

My first time was under warranty, 2nd time 2 years later was not.  Cost to repair was about 150.00. 

Paul
Logged

Paul Caldwell
Little Rock, Arkansas U.S.
Photography > http://photosofarkansas.com
Blog> http://paulcaldwellphotography.com
dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #5 on: September 16, 2012, 09:30:52 PM »
ReplyReply

I just read your test page information. Can i use a high quality laser printer for this? I don't even own an ink jet. I have a color laser though. Also, can i use just some bright white heavy, quality paper? I'm assuming that the test pattern was developed only for either Canon or Epson ink jets?

I looked at a comparison of the guy with the lens that he replaced, and i could not see any difference in the two ens he tested--lol. What should I be looking for?
Logged

dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #6 on: September 16, 2012, 09:35:40 PM »
ReplyReply

Interesting. So that's why it broke loos on impact while  nothing else did. Do you know if there was a recall for that? I recently sent my 5D original back in for the mirror replacement, although I had no problems with it coming loose. I didn't even know about the recall, then just happened on it on the net. canon repaired it free of charge even after 5 years. (I haven't shot with the 5D since 2008. It's just a back up now.)

Back to the name ring breaking loose, mine is still tight, uh, really tight now since it's wedged in crooked. lol Did the ring problem cause any quality problems with the lens, to your knowledge?
Logged

BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3553


« Reply #7 on: September 17, 2012, 05:49:03 AM »
ReplyReply

I just read your test page information. Can i use a high quality laser printer for this? I don't even own an ink jet. I have a color laser though. Also, can i use just some bright white heavy, quality paper?

Hi,

A laser printer won't give the best continuous tone results for smooth gray tones, but you can try. Alternatively you could have it printed by a photolab or print service on photochemical paper, just make sure that they don't resize or sharpen the image (no image 'enhancements'). That will change the output size, because they usually print at a lower PPI, so you may have to specify a paper size that's at least 260mm on it's short side.

Quote
I'm assuming that the test pattern was developed only for either Canon or Epson ink jets?

I've designed the target to be printed at the highest possible output resolution, to allow shooting it from shorter distances (25x focal length). Basically the only difference between the Epson/Canon printer versions is the file size, which would result in a target size of 130mm square (one could fit 2 of them on a single A4 size paper) when printed at the highest (native) output resolution. That should also avoid that the printer driver resamples the image. However, one can print at a lower PPI setting, which will increase the output size, but then the lower resolution target needs to be shot from further away (50x focal length or more) to keep challenging the lens resolution.

Quote
I looked at a comparison of the guy with the lens that he replaced, and i could not see any difference in the two ens he tested--lol. What should I be looking for?

One would typically look for signs of lens element decentering, which would show as an asymmetrical blur in the center of the target. When everything is okay, then the center will show a nice circular blur, not elliptical.

Cheers,
Bart
Logged
Paul2660
Sr. Member
****
Offline Offline

Posts: 1678


WWW
« Reply #8 on: September 17, 2012, 12:19:21 PM »
ReplyReply

I don't think there was any recall, Canon handled each case as it came in.  If you go back far enough on this site or DPreview and search name ring 24-70 you will find quite a bit written on it.
It tended to break loose and rotate freely which made use of the hood impossible.  There were actually websites dedicated to improvised fixed.  I had one in place for about 2 years then it broke and
I had to break down and pay Canon.  I will say that the 2nd time the fix worked. 

Paul


Logged

Paul Caldwell
Little Rock, Arkansas U.S.
Photography > http://photosofarkansas.com
Blog> http://paulcaldwellphotography.com
francois
Sr. Member
****
Offline Offline

Posts: 6782


« Reply #9 on: September 17, 2012, 12:25:32 PM »
ReplyReply

I don't think there was any recall, Canon handled each case as it came in.  If you go back far enough on this site or DPreview and search name ring 24-70 you will find quite a bit written on it.
It tended to break loose and rotate freely which made use of the hood impossible.  There were actually websites dedicated to improvised fixed.  I had one in place for about 2 years then it broke and
I had to break down and pay Canon.  I will say that the 2nd time the fix worked. 

Paul




Same here... Mine broke loose after three years and I had to pay around $300 to have it repaired. Initially, Canon service (in Switzerland) told me that they would consider the repair to be free!  Angry

Logged

Francois
dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #10 on: September 17, 2012, 09:23:10 PM »
ReplyReply

Thanks Bart. I'll have it printed with a chemical printer. Also, thanks for the explanation on what to look for. I'm wondering if you wold be interested in looking at the raw file once I shoot it? I can upload to my server.

This could save me a lot of time and money, since if everything looks ok, I can simply return the lens to get the name ring reset.

Also, I didn't see anything about testing for mirror alignment. I'm assuming that to test for that, I'd need to use the Leica disto meter to square the lens and camera to the subject file?
« Last Edit: September 17, 2012, 09:28:06 PM by dwdallam » Logged

BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3553


« Reply #11 on: September 18, 2012, 04:43:03 AM »
ReplyReply

Thanks Bart. I'll have it printed with a chemical printer. Also, thanks for the explanation on what to look for. I'm wondering if you wold be interested in looking at the raw file once I shoot it? I can upload to my server.

Sure, no problem. Just send me a PM with the URL and PW, or post a crop here.

Quote
This could save me a lot of time and money, since if everything looks ok, I can simply return the lens to get the name ring reset.

Also, I didn't see anything about testing for mirror alignment. I'm assuming that to test for that, I'd need to use the Leica disto meter to square the lens and camera to the subject file?

Do you mean alignment of the target, or the camera mirror? In case of the target alignment, yes try to square it reasonably well, but there is no need to get it perfect. When you focus (preferably with Live View and/or contrast detect AF) on the center of the target, it will be fine when you use an aperture of f/5.6. That would give enough DOF to compensate for slight misalignment. Besides, when you let them make a photochemical print (remember, no resizing or image alterations), you'll probably need to shoot from a larger distance (due to the larger size of the target), which makes the angular deviation from perpendicular even smaller.

If you want to be really sure of the lens, it is useful to also shoot 4 individually focused shots at the corners. However, we won't know if any deviation found there wasn't in the lens to begin with. Most lenses have slight decentering and different corner performance, which usually goes unnoticed unless in critical reproduction work.

As for any required camera mirror adjustment, that should become obvious if the viewfinder gives a markedly different focus suggestion than Live View does. It would also show as a difference between regular Phase Detect AF and Live View.

Cheers,
Bart
Logged
dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #12 on: September 18, 2012, 02:24:44 PM »
ReplyReply

Small problem here. Whe3n I open the Canon file in PS, it says the file size is
2071 px
3071px

3.412"
3.412"

Resolution: 899.998

Also I need to save it as a jpg (maximum no loss) for printing. I can also use a tiff.

The Noritsu printer I think only goes to 320 ppi/dpi I think.

So should I use 320ppi and uncheck "resample image" ?

That would give a 9.597 square image for printing.
« Last Edit: September 18, 2012, 02:27:05 PM by dwdallam » Logged

joneil
Full Member
***
Offline Offline

Posts: 136


« Reply #13 on: September 18, 2012, 02:58:58 PM »
ReplyReply

  I sure know where you are coming from.  I am waiting for my Nikkor 24-70 to come back from repair from Nikon.   Teh Nikkor and Canon 24-70 are very similar lenses, and cost near the same price.  In my case I dropped the lens alone, not the lens and body.  Two feet to pavement.  It appeared to have no damage, but afterwards while it would auto focus fine, all images I shot were all soft at the edges.     tried it on two camera bodies, same thing.

  So back to Nikon, and the repair cost is about 1/3rd the price of a new lens, but what can you do?  7 -10 "businenss days, if the parts are in stock" before I gt my lens back.  Grrrr.

   so two bits of advice.  Not wanting to scare you, but with a lens that good, you maybe should look real seriously at sending it back.  now in my case, the apparent loss of image quality was visibile right away. but why take chances with such a good lens.

   Another thought is your next purchase, if any, look at some of those Zeiss manual focus lenses.   I use them myself.    The reason i say that is in the past I have dropped manual focus Nikkor lenses, and  they still work fine and are very sharp.  Not that I want to put the issue to the test with my Zeiss, but one thing all those lens tests and lens reviews never seem to cover - at least none I have seen, but do correct me if I am wrong - is with all these new auto-focus lenses, especially zooms, and double especially if they are image stabilized / vibration reduced is, you ever drop one, more often than not you are toast.    That's from direct experience.  The Nikkor 24-70 is one of their best "professional" lenses on the market today, and a two foot fall did it in.  My older Nikkors, bigger falls (one, once down a cliff), and they still work fine)
 
   
  so something to think about.  Good luck, and get that lens looked after asap if you can.

good luck



Logged
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3553


« Reply #14 on: September 18, 2012, 07:16:51 PM »
ReplyReply

Small problem here. Whe3n I open the Canon file in PS, it says the file size is
2071 px
3071px

3.412"
3.412"

Resolution: 899.998

3071 pixels square is correct. The PPI resolution then depends on the output size (3071 / PPI = size in inches).

Quote
Also I need to save it as a jpg (maximum no loss) for printing. I can also use a tiff.

Highest quality JPEG should be no problem, unless there is a file size upload limitation.

Quote
The Noritsu printer I think only goes to 320 ppi/dpi I think.

So should I use 320ppi and uncheck "resample image" ?

Correct, uncheck "resample image" first, then set the PPI that the printer will use. The output size will then be calculated automatically.

There are different Noritsu printers (Photochemical and Inkjet), with different native resolutions (300-640-720-1440). Check which resolution machine they will be using.

Cheers,
Bart
Logged
dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #15 on: September 18, 2012, 08:42:05 PM »
ReplyReply

Bart the Noritsu is a chemical printer.

I changed the resolution to 310, which made the image 9.9" sq. using the same settings as above.

I had it printed on 12x12" paper.

After I got the print back, it was 12"x12" so I'm thinking the operator did not read my specifications as to NOT adjust anything. It should have had 1" border all around the 12"x12" paper.

Should I have it reprinted since somehow the resolution is all changed up now, thanks to Costco (which I never use anymore because of this time wasting crap).
Logged

BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3553


« Reply #16 on: September 19, 2012, 08:20:42 AM »
ReplyReply

Bart the Noritsu is a chemical printer.

Then ít's either a 300 PPI or a 640 PPI printer when they use current models.

Quote
I changed the resolution to 310, which made the image 9.9" sq. using the same settings as above.

I had it printed on 12x12" paper.

If they had followed your instructions of no resampling, the PPI would be ignored and it would have printed pixel for pixel at the native printer PPI. A 12x12 inch output paper would have been large enough for a 300 PPI printer, so that's okay.

Quote
After I got the print back, it was 12"x12" so I'm thinking the operator did not read my specifications as to NOT adjust anything. It should have had 1" border all around the 12"x12" paper.

Sadly that happens all the time, good operators are a rare commodity, now the image got interpolated up to a 17% larger size (and it was already 2x larger than designed).

Quote
Should I have it reprinted since somehow the resolution is all changed up now, thanks to Costco (which I never use anymore because of this time wasting crap).

Formally they didn't execute the order as requested (it became a 256 PPI image instead of 300) so that would warrant a reprint, but on the other hand the low resolution version will still be usable. The only real issue is that the image now has a significantly lower resolution than intended (256 PPI versus 600 PPI). That means you need to shoot it from a larger distance to get a low enough magnification factor to challenge the lens. That distance, something like 60x focal length, may be somewhat inconvenient indoors.

I received your link to the 4 sample shots of the Costco version, but they were shot @ approx. 2.15m (or a 1:31.5 magnification factor), which is too close for critical evaluation. For the 66mm focal length you used, the low resolution chart would need almost twice that shooting distance (50-60x focal length). The 2 smaller chart shots (from your laserprinter?) were small enough to challenge the system resolution, and they show no obvious signs of lens decentering. So it looks like the optics are not affected.

Cheers,
Bart
Logged
dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #17 on: September 19, 2012, 04:01:16 PM »
ReplyReply

Well that's good news. I needed to send in the lens anyway to have the name ring repaired--I probably should have just kept it and used it like it was to save money. But I just hardly can't stand using something like that.

As for the camera itself, I was really paranoid about that and sent it in too. There just isn't anyway to know if, for instance, the mounting ring is a little out of round now allowing a light leak, or the mirror alignment got knocked off a little, or a million other things.

I just hope the technicians do the right thing and find any problems. I left instructions to calibrate the lens, which probably doesn't need it, i.e., before the fall it was just superb in sharpness. Incidentally, my 70-200IS L had to be sent in and calibrated after I first received it. When I got it back, there was an incredible increase in resolution and sharpness. I also asked them to test the camera for mirror alignment problems and viewfinder alignment too.

While it's in for repair, I think I'll use the laser printed version and test the 70-200 with your test pattern on my 5D. I've seen some strange stuff with it recently. I think you replied in an earlier post about my concerns. It seems soft, but when I take a picture of something close up, it's very, very sharp. It has me stumped.

Last, and most importantly, thanks very much for looking at those images and your help.
« Last Edit: September 19, 2012, 04:03:13 PM by dwdallam » Logged

dwdallam
Sr. Member
****
Offline Offline

Posts: 2044



WWW
« Reply #18 on: October 21, 2012, 12:24:21 AM »
ReplyReply

Bart,

I finally got around to testing my 70-200 after the camera came back. If you have time, here is the link to the files:
www.dwdallam.com/shared/BartTestFiles.rar

These were taken at various zoom and f-stops on a tripod, and a release, but not using mirror lockup.

Thanks much.
Logged

Pages: [1]   Top of Page
Print
Jump to:  

Ad
Ad
Ad