[Rondec]

I guess one thing I would like to see is failure rate after the lenses have been fixed. The question is if there is something that just went wrong and once it is fixed, it will be good for a long time, or is it a part that is prone to fail, even after replacement.

[8540tomg]

I think the poll is a good idea as I have been seriously looking at the DA* 15~50 lens. I have been reluctant to pull the trigger as I've been reading about a number of problems with this lens of late. I spoke to the a sales person at the local Henry's store and they said they had no complaints with any of these lenses. Of course they were trying to sell me a lens.

I suppose it would be good to know how many of these lenses Pentax have produced in each category. If they cranked out 25,000 DA* 16~50 lenses and only a hundred or so had issues the odds of getting a bad one are slim. Are these sort of production figures available anywhere?

Tom G

[stevebrot]

I can't really answer your question. First, I really see 3 categories of "experience" with SDM:

1. Never Failed (pretty obvious)
2. Have experienced non catastrophic failure(s) (i.e. failed to work from time to time, but not consistently, and not very often)
3. Catastrophic failure (i.e. stopped working altogether and required maintenance)

Even then this is still not portraying reality as usage (and conditions) has to be taken into account. for e.g. Never Failed... well..."I bought it yesterday and it hasn't failed yet !" that does not that compares to, "I have taken over 5,000 images and it never failed " ? or else... "It failed on me after using it for 5 hours in sub-zero temperatures with no protection / warming what-so-ever"

In other words, no simple survey will ever portray an accurate picture of the reliability of SDM. On top of that, to get significant statistics, you'd need to establish a minimum, RANDOM, sample. i.e. one needs to randomly target SDM Lens Owners as of course, those who experience failure are more motivated to answer your survey than those who did not, hence skewing the result.

My thoughts anyway...

Thanks for your thoughts. I can think of at least a dozen more ways to improve this survey. Unfortunately, the polling feature is not very sophisticated. What I would have really loved to have included was some notion of the age of lens at failure, how long the user has owned their lens, how often it is used, and whether it has been repaired/replaced more than once.

Your case #2 sounds like the kind of issue we all get with the screw drive lenses. If light is poor or there is little contrast with the subject, the AF will hunt or simply give up. The is usually a body issue, though intermittent failure might be indicative of total failure in the future.

As for the poll validity...be my guest to due a valid random sampling. It should not be too difficult to get the contact information for all sales.

The main reason I did this is because I keep reading claims that the vast majority of SDM lenses are failing. If that is true, then I would expect that the majority of SDM owners on this site would have had a problem.

Steve

[stevebrot]

If you want some numbers, here are at least 50 failed lenses. There are actually more failed lenses as some of the posts in the list are for more than one lens.

https://www.pentaxforums.com/forums/831080-post170.html

Thank you
Russell

Thank you Russell for taking time to compile this list. How many non-failed lenses at dpReview?

BTW...Do you own a SDM lens and did you vote?

Steve

[stevebrot]

I guess one thing I would like to see is failure rate after the lenses have been fixed. The question is if there is something that just went wrong and once it is fixed, it will be good for a long time, or is it a part that is prone to fail, even after replacement.

That is one of my concerns as well. As noted above, the polling facility in the forum software is pretty primitive (sort of has to be). A wizard-like flow would be a lot better. From what I have read on the Web, there is anecdotal evidence that fixed lenses often fail again.

Steve

[stevebrot]

i don't vote, cause i ve got no more than 200 shots with DA* 300mm. Most where in a zoo at 5 degrees celsius.
No problems so far.

Well, at least we know that yours was not DOA in the box when you got it!

Steve

[deadwolfbones]

I don't have any SDM lenses, but my sole HSM lens had its motor fail about two weeks after I got it.

[builttospill]

I also didn't vote, because my DA 17-70 is at Pentax's shop now for repair. However, my SDM didn't fail, it just lost the ability to lock focus on it's own. So I can't say "bad" or "good".

[VAV]

I can't really answer your question. First, I really see 3 categories of "experience" with SDM:

1. Never Failed (pretty obvious)
2. Have experienced non catastrophic failure(s) (i.e. failed to work from time to time, but not consistently, and not very often)
3. Catastrophic failure (i.e. stopped working altogether and required maintenance)

Even then this is still not portraying reality as usage (and conditions) has to be taken into account. for e.g. Never Failed... well..."I bought it yesterday and it hasn't failed yet !" that does not that compares to, "I have taken over 5,000 images and it never failed " ? or else... "It failed on me after using it for 5 hours in sub-zero temperatures with no protection / warming what-so-ever"

In other words, no simple survey will ever portray an accurate picture of the reliability of SDM. On top of that, to get significant statistics, you'd need to establish a minimum, RANDOM, sample. i.e. one needs to randomly target SDM Lens Owners as of course, those who experience failure are more motivated to answer your survey than those who did not, hence skewing the result.

My thoughts anyway...

There seems to be another category of failure - the SDM mechanism works fine as long as the lens is being used but SDM problems arise if the lens is unused for some time.

[philbaum]

There seems to be another category of failure - the SDM mechanism works fine as long as the lens is being used but SDM problems arise if the lens is unused for some time.

I've had my DA300 for over a year, will put it away for a month or more at a time, then use it. Use my DA 50-135 fairly frequently and don't store it for much of the time.

Had problems with neither, even in cold weather, down to 19 deg F recently.

I really feel that the problems are not occurring with the majority of lenses, but don't know that for a fact. The reports are worrying, i'll admit that. Glad i bought them before reading about the reports. Seems like the furor over the problem is dying down a bit, if i'm not mistaken. Could it be that Pentax has fixed the cause, i certainly hope so. Great lenses optically and i wanted the 50-135 so that i could take pictures in a theatre without the screw noise. has worked great for that.

[Russell-Evans]

How many non-failed lenses at dpReview?

If the SDM failure rate was 10%, 50 failed lenses would mean there are 500 SDM lenses at DPR.

If the SDM failure rate was 1%, 50 failed lenses would mean there are 5000 SDM lenses at DPR.

If the SDM failue rate was .1%, 50 failed lenses would mean there are 50,000 SDM lens at DPR.

I don't think there more than 5000 SDM lenses at DPR. I would bet there are more than 500 SDM lenses at DPR, so I think the failure will have to be between 1-10%. The best case of a 1 in 100 chance for a $1000 lens doesn't really excite me. I personally would hope for at least 1 in 1000, but I don't actually think 50,000 SDM lenses have even been produced.

I don't know if the serial numbers are sequential, but my 50-135mm has a serial number in the 6000s. Anyone have a higher serial number on a 50-135mm?

Maybe it would be good to have people post the range of the serial number they have per each SDM lens? Maybe this maybe would give some idea of how many are produced in each lens type, 16-50, 50-135, 17-70, 60-250, 55, 200, and 300.

Thank you
Russell

[Russell-Evans]

I also didn't vote, because my DA 17-70 is at Pentax's shop now for repair. However, my SDM didn't fail, it just lost the ability to lock focus on it's own. So I can't say "bad" or "good".

I've seen three posts of this type of failure. When your lens comes back, would you mind replying to this with what was documented as the failure, please?

Thank you
Russell

[stevebrot]

If the SDM failure rate was 10%, 50 failed lenses would mean there are 500 SDM lenses at DPR.

If the SDM failure rate was 1%, 50 failed lenses would mean there are 5000 SDM lenses at DPR.

If the SDM failue rate was .1%, 50 failed lenses would mean there are 50,000 SDM lens at DPR.

I don't think there more than 5000 SDM lenses at DPR. I would bet there are more than 500 SDM lenses at DPR, so I think the failure will have to be between 1-10%. The best case of a 1 in 100 chance for a $1000 lens doesn't really excite me. I personally would hope for at least 1 in 1000, but I don't actually think 50,000 SDM lenses have even been produced.

I don't know if the serial numbers are sequential, but my 50-135mm has a serial number in the 6000s. Anyone have a higher serial number on a 50-135mm?

Maybe it would be good to have people post the range of the serial number they have per each SDM lens? Maybe this maybe would give some idea of how many are produced in each lens type, 16-50, 50-135, 17-70, 60-250, 55, 200, and 300.

Thank you
Russell

1-10% failure rate within warranty would not surprise me at all nor would a higher rate outside warranty. The repair incidence for dSLR bodies across brands is 3-5% (consumerreports.org) with Canon and Nikon leading the pack for failures. I would take that as an indicator of the QA target for that class of goods, lenses included.

Steve

[stevebrot]

I've seen three posts of this type of failure. When your lens comes back, would you mind replying to this with what was documented as the failure, please?

Thank you
Russell

Add your vote to the total as well.

Steve

[LeDave]

Wow the results from the 16-50 is surprising. I used to have the 50-135 but sold it, it has never failed on me and hopefully it doesn't fail on that person.