jerrykur wrote:
Yet private fixed wing AC owners in the US can and do fly until conditions warrant repair and our planes are not falling out of the sky.
Getting back to cameras there are plenty of reports of 1 series cameras having shutter failure at 20 K, and other cameras still going strong at 500K.
They are though. I know a couple private aircraft owners who both had put their engines 'on condition' only to have them fail in flight well after their recommend overhaul life. A scheduled overhaul is much cheaper than a connecting rod through a crankcase, as a shutter replacement is much cheaper than a sensor replacement. Point being, the gamble is yours, but it is a gamble nonetheless.
It's quite bizarre the way this has become "replace your shutter before the time Canon say is simply the MTBF or it will destroy your sensor", which is not the usual case at all.
helimat wrote:
When doing so can prevent damage to other parts of the camera, as brainiac suggested. He wasn't talking out of his ass.
Indeed. When you know full well you have exceeded the design specifications, it becomes a matter of common prudence at least to have it checked for wear, replaced if necessary.
Mechanical wear is mechanical wear, whether in an automobile engine or a shutter. The comparison of the shutter to a timing belt is apt.
MTBF means just that: there is no counter in your camera, counting down to destruction. One camera is going to croak at ten shots, another will go ten million. No way of knwoing which is which until you try them.
Actually, this is not a MTBF figure. According to Canon, it's a design test goal. Canon has never released any actual MTBF figures for their equipment.
miccullen wrote:
It's quite bizarre the way this has become "replace your shutter before the time Canon say is simply the MTBF or it will destroy your sensor", which is not the usual case at all.
Like I said, it's your call. I would chose the safe option.
^Yep. I had one go on a 1D2 and had to replace the sensor too. $1500AU. Shutter count was 65K. The replacement also went at 67K and the repair place wanted to also replace the whole sensor unit because of a scratch on the AA Filter. ****Edit**** Canon here won't guarantee an AA Filter replacement and so replace the filter and sensor as a unit. ****End Edit**** I declined that part of the repair and still can't see the scratch on any images. So, yes it does happen. If it goes again at around the same count, I'll be really pissed.
Keep an eye on your images for the tell-tale signs of shutter failure. there are lots of threads here on FM with pictures that show the signs. Don't keep shooting if you see those signs because replacing sensor or even AA filters will add a lot to the cost and down-time.
dasrocket wrote:
PM via shutter replacement is a little stretched no? Is there anyone besides Richard who had their shutter explode here?
I can think of one other case I can remember where a shutter failure caused sensor damage.
The other question to ask is how many actuations where on the 5D at the time of failure? If it was substantially below Canon estimated lifespan that raises in my mind a big question regarding the accuracy of the Canon estimate. The second post in the thread contains a link to some statistics about failures. While I do not think there is enough data to be statistically accurate it is interesting to note the number of shutters that failed around 10-20% of Canon's estimate and the number that exceed Canon's estimate.
Shutters don't have a MTBF (Mean Time Between Failures) because they are never repaired and put back into service - they're completely replaced. The more correct term is MTTF (Mean Time to Failure). They don't contain replaceable components (like car engine timimg belts) that have a much shorter service life than the complete assembly.
That said, any number with 'mean' in it is pretty much useless, because shutter failures don't follow a gaussian ("normal") distribution. That's why Canon never uses the terms MTBF or MTTF anywhere in its literature when talking about shutter life. Failure rates follow a 'bathtub' curve rather than a bell-shaped one. (Google 'bathtub curve' if you want a more complete explanation than I'm able to give.) The curve has three parts:
1. There is a (relatively) high failure rate for 'nearly new' assemblies. This part of the curve is often called 'infant mortality', and is where manufacturing defects show up.
2. Shutters that survive 'infant mortality' have a very low failure rate for a long time. There's a high probablity that they will keep functioning correctly until they hit the third ('wearout') part of the curve. They'll probably last much longer than any 'mean time to failure' number.
3. When you hit the 'wearout' portion of the curve, failure rates start to rise rapidly.
You don't really know where the wearout part of the curve begins until you have a large number of failures to analyze. If Canon's engineers did a good job of estimating the servce life for the shutters, it will be somewhere beyond the 200k number in their marketing brochures. If you want to know where the wearout part of the curve begins, you need a lot of data about shutters that failed _after_ they reached 100k or 200k, or more actuations. I don't know where you can get that data.
I think that Hammy's regular shutter replacement (after every 300k or so clicks) makes good sense. But I think that anyone who thinks a shutter is significantly more likely to fail at 220k than it is at 180k or 150k is making a wild guess without any good evidence to support it.
Edit: I'm talking about shutters on 1D (pro) cameras; 5D shutters don't have nearly the same life expectancy
You don't really know where the wearout part of the curve begins until you have a large number of failures to analyze. If Canon's engineers did a good job of estimating the servce life for the shutters, it will be somewhere beyond the 200k number in their marketing brochures. If you want to know where the wearout part of the curve begins, you need a lot of data about shutters that failed _after_ they reached 100k or 200k, or more actuations. I don't know where you can get that data.
I would suspect Canon has the data, and that they claim as high a figure as their data can support.
Yeah, you're probably right, but it may not be not quite that simple.
The engineers do the best design they can with the budget they're given, and then try to validate the design by testing pre-production samples. After the camera goes into production, they keep testing and watch the data from their service centers for signs of unexpected failures. But, they don't know how well the 'real world' failure rates match their test data until after the camera's been in production for years - long enough for a lot of shutters to wear out.
Canon keeps raising the advertised shutter life for each new generation of 1D cameras. I expect that one reason for that is that they keep incrementally improving the design and the components they use, but there are enough 'very high mileage' older cameras around to make it look like the earlier published estimates were real conservative.
Aside from some manufacturing problems (like the bunch of ealy 1Ds2's that failed prematurely), they seem to have figured out how to make darn reliable shutters. I'm not losing any sleep over my (currently) 'low mileage' 1Ds2 suddenly crapping out when it hits 200k actuations.
Jun Zhou wrote:
hmm, my friend spent $750 on replacing the shutter with CPS.
Probably talked him into a mirror box replacement as well. They tried that with me and I told them to go take a flying leap. Canon is really bad about trying to sell you on a mirror box replacement when you replace the shutter. The problem with that is the MTBF on a mirror box is 1,000,000 frames. I ended up paying just under $300.00.
My shutter was replaced around 260K and it was still going strong. I replaced it because I had a service voucher with Canon that was about to expire. Hammy has had several shutters approach 500K before they failed. And BTW, in 7 years of shooting for Hammy I have never seen a shutter "explode". I have seen several fail but never explode and damage other components. Not that it can't happen I have just never seen it.
I wouldn't worry about it. Most of the time it will warn you with either white or black lines on your shots.
Okay, at the risk of being labelled a whiner, I’m going to post some images taken by my 1DmkII shortly before the shutter expired the first time… I’m posting this, not because I’m wanting to again bemoan the fact that it happened to my camera, but because there are people on this thread who in all honesty haven’t heard of it happening. I just want to show you what happened with my camera and tell you that the shutter failed and jammed the mirror mechanism on my particular camera AND also took out the sensor (although I believe it was more a marked filter that Canon then replaced as a unit). I’d like people to be aware of what to look for in their images, and to be aware that simply continuing to shoot MAY result in something like what happened to my camera.
Please, for all of you reading this, please note that I am choosing my words carefully. Don’t read what I’m saying and jump all over me because of something you’re reading that I HAVEN’T written…..
The images are in chronological order and taken over two days but only over about 200 actuations on the shutter. The last of them (which look like the camera strap is in front of the lens) were taken in burst at 8.5 FPS whilst I was tooling around with the camera. Hopefully the EXIF data is intact, but I’ll check that after it is posted.
So, on image 1, I couldn’t see any indications of impending shutter failure. On images 2 thru 6 you will see at the very top of each image a lighter straight line. It’s on a slight diagonal, but it is straight-edged. If you look closely, you may also notice a lighter area about halfway down each frame. It is not straight-edged, but more curved. It could easily be mistaken for light or flare (except if you’ve seen it before on one of your images…lol). The remainder of the images show that something is definitely amiss – although by the time I’d seen these, the camera had ceased to function.
The point is, that happened over just a couple of hundred frames and that it happened at the wrost possible moment – when shooting at high speed burst. That caused other collateral damage to the camera. It can happen. It has happened to me and by the sounds of it to Brainiac as well. It didn’t “explode”. It did shatter.
Just keep an eye on your images and if you see the tell-tale signs, then don’t just keep shooting. Send it in for repair. It’s probably not a common occurrence, but that is no consolation if it happens to YOU. I hope this post has been of some use to you and my apologies to the OP if this has hijacked your thread.
<----------- 3 1dsmkII shutters for me, 3 separate cameras. 1 while testing to purchase, one purchased and returned, one purchased, used a week, then shutter replaced.
jeremy_clay wrote:
<----------- 3 1dsmkII shutters for me, 3 separate cameras. 1 while testing to purchase, one purchased and returned, one purchased, used a week, then shutter replaced.
Fail.
Wow... That is crazy. With that kind of luck, I would suggest staying inside during electrical storms.
Highly unlikely it will get paid. A product wearing out isn't an insured event under most policies. Its a warranty issue with the manufacturer and if you are out of warranty you are out of luck.
Jman13 wrote:
If that happens, I'll file an insurance claim.
What Kirk said is of course correct (and he would know!) Of course Canon has this data, and more likely they have tested it themselves, just like every other product manufacturer. Not only that, but I'm sure the engineering isn't all that different shutter to shutter across the lines so the experience from other cameras is also relevant. The warranty isn't based on a guess - its analyzed and part of the cost estimates the company has figured in to warranty repairs going forward.
RDKirk wrote:
I would suspect Canon has the data, and that they claim as high a figure as their data can support.
