PDA

View Full Version : Pronounced vibrations, broken IBIS?


Tordan58
4th October 2013, 08:17 AM
Hello,

By looking at the pictures taken the past days, virtually 0% sharp pictures and from the way the camera vibrates and the noise/humming I have a feeling that the IBIS on my OM-D E-M5 has broken or is no longer operating the way it should.


Here are symptoms:

I use the camera with long focal length lenses (600mm+), with the setting "Half Way Rls With IS" on, IS mode 1.
Upon half pressing the shutter release button, first a gentle vibration is felt as the IBIS engages and for a short moment the image in the EVF stabilizes. This gentle vibration has always been present even when camera was new (as far as I remember).
However after a short while (a second or so) the vibrations intensify and a relatively low frequency humming can be felt and even heard. As this takes place the EVF image gets severely blurred.
After yet another second or so the vibrations will peak. Next the vibrations are gone and the image in the EVF is "unstabilized", as when using the camera without IS.
Furthermore: I also found out that tilting/pitching/yawing the camera 90 degrees while the shutter release button is half pressed causes a noise, reminding of the sound of metal on metal.


If above description is unclear I could try to capture the behavior on video.

Given the bad experience I have had with Olympus service, I wonder if anyone could check the behavior you have on your OM-D E-M5 and let me know before I reluctantly sent the camera to Olympus and claim the guarantee. The camera is almost new, it was exchanged by Olympus in May 2013 after the previous one broke (broken shutter). I have not mistreated or dropped the camera in any way.

/Tord

David Morison
4th October 2013, 08:51 AM
I find the IBIS works fine with 4/3 and M4/3 lenses but with a non-native lens (Canon 400, Pentax 500) it is difficult to get an image that is stabilised in the EVF. I have set the IBIS to the focal length of the lens in use and even tried focal lengths either side of the actual one. Interestingly when using the 2x DTC, even with a M4/3 lens, the sensor dances around a lot more than without the DTC, and makes more noise. I can also see the image going slightly in and out of focus after MF on non-native lenses. I'm interested in seeing whether this happens with the E-M1 to determine if it is generic or a fault with the camera, although it has happened since new.

David

Tordan58
4th October 2013, 09:09 AM
I find the IBIS works fine with 4/3 and M4/3 lenses but with a non-native lens (Canon 400, Pentax 500) it is difficult to get an image that is stabilised in the EVF.

...
David
David,

I found the IBIS was working fine with my scopes and rendered a stable image when FL was set at 600mm (always first time) 800mm (usually stable first time, sometimes had to re-engage IBIS), 1000mm more tricky (eventually after re-engaging the IBIS).

The thing is that it now seems broken.

bilbo
4th October 2013, 09:49 AM
Do you have this problem with native lenses?

OM USer
4th October 2013, 09:51 AM
One of the firmware updates did change the way the IBIS operated. Before the update the camera would hum when turned on, after the update it would only hum on the half press and the hum was a bit louder. If you have not got the this update that would be worth a try but I suspect you update with each firmware release.

I have no problems with IBIS on my 50mm OM Zuiko or a 500mm mirror. The latter gives very soft pictures lacking in contrast so I don't use it but the picture is stable.

Ian
4th October 2013, 10:14 AM
If you are using lenses manually (non-Olympus Four Thirds or Micro Four Thirds) of course (and I think you already know this) you need to set the correct focal length in the IS settings. Is there any chance there might be an incorrect setting value?

Ian

Tordan58
4th October 2013, 10:41 AM
If you are using lenses manually (non-Olympus Four Thirds or Micro Four Thirds) of course (and I think you already know this) you need to set the correct focal length in the IS settings. Is there any chance there might be an incorrect setting value?

Ian
Ian

I am familiar with the need to adjusting IS settings to fit the actual FL used, if no exact match use the closest value.

/Tord

brian1208
4th October 2013, 10:41 AM
3 and 4 sound unusual, certainly nothing I have experienced with mine using either m4/3rds lenses or my legacy lenses, up to and including my 500mm Tamron mirror

5 is to be expected as the IBIS is only capable of accommodating the relatively small movement associated with hand-holding. I saw the same effect with my canon bodies if I moved the camera too fast with IS on

Tordan58
4th October 2013, 10:43 AM
Do you have this problem with native lenses?
I haven't used and don't plan to use native (m4/3 lenses) with the body. I could try, I should have one somewhere.

Tordan58
4th October 2013, 10:46 AM
3 and 4 sound unusual, certainly nothing I have experienced with mine using either m4/3rds lenses or my legacy lenses, up to and including my 500mm Tamron mirror

5 is to be expected as the IBIS is only capable of accommodating the relatively small movement associated with hand-holding. I saw the same effect with my canon bodies if I moved the camera too fast with IS on
Thanks Brian for sharing your experience as reference.
I will try to record 3 and 4 on video to help getting an idea how bad it looks and sounds.

bredman
4th October 2013, 02:10 PM
It sounds faulty to me. Simple.

Ralph Harwood
4th October 2013, 11:46 PM
Ian

I am familiar with the need to adjusting IS settings to fit the actual FL used, if no exact match use the closest value.

/Tord

Hi there Tord!

Just a quick question - do you use an adapter with a chip in it (I seem to remember you have in the past)? The reason I ask is because recently when checking the EXIF data on some of my photos taken with a mirror lens and the adapter, the EXIF data showed the lens length as 65535mm (presumeably the highest number the chip could contain as a default). On my e-30 the manual states that if the camera gets a focal length from the lens this will override the set value you enter via the IS menu.

I was wondering if you were using this adapter whether the issue is caused because the sensor is trying to massively overcompensate for movements due to an incorrect focal length - 60 to 100 times more movement than is necessary! This could possibly cause the increased vibration as the slightest movement would cause a massive reaction.

Cheers,

Ralph

Tordan58
5th October 2013, 05:42 AM
Hi there Tord!

Just a quick question - do you use an adapter with a chip in it (I seem to remember you have in the past)? ...
Cheers,

Ralph
Hi Ralph,

No, the adapters I use have no AF confirm chip. You are right, I did use a chipped adapter but gave up since some of the adapters will cause the camera to malfunction. And the EVF+magnification are a better support than the chips.


Thanks
Tord

Tordan58
9th October 2013, 07:24 AM
During last session the camera failed in rendering any stabilized image on the viewfinder. Camera is vibrating and humming, however shake blur is clearly visible on pictures unless using short exposure times, something like than 1/ 2*FL

Camera has been sent back to Olympus for them to look at.

Tordan58
23rd October 2013, 08:50 PM
I just got the camera back from Olympus. As I suspected, the report from the workshop states that the IBIS was broken and has been replaced.

brian1208
23rd October 2013, 09:07 PM
A good outcome to a frustrating problem Tordan - enjoy the return of stability :D

Tordan58
25th October 2013, 05:02 PM
Guess what?

In preparation of the coming weekend I re-programmed the camera settings (the workshop had reset them) and went out for a short test to check the behavior and check that the settings were as I want them to be.

First pictures went OK, it was a delight using the stabilized EVF. When taking the 6:th or 7:th picture I find out the vibrations are back and IBIS is dis-engaging.

:mad:

This will be the third E-M5 failure that I face within six months period. Shutter, IBIS and IBIS again.

:mad:

Am I the only one here ... ?

:mad:

Falk
25th October 2013, 05:25 PM
Oh dear - that stinks. I feel sorry for you Tord! You probably should complain/claim an EM-1 as compansation for such cruel threatment!?

Tordan58
25th October 2013, 05:36 PM
I just sent a letter of complaint, not accepting the way they treat loyal customers, demanding a compensation and asking support to escalate to management.

jamsa
25th October 2013, 07:59 PM
Yet another reason to question the sense of sticking with M4/3rds and the EM1....oh dear! Keep us posted thanks

Tordan58
26th October 2013, 03:06 PM
I have the behavior on a short movie that clearly demonstrates the issue and how it materializes.

Is there a way to share the movie on this forum?

/Tord

jamsa
26th October 2013, 03:26 PM
I have the behavior on a short movie that clearly demonstrates the issue and how it materializes.

Is there a way to share the movie on this forum?

/Tord

You can put in on you tube with a link on here or use vimeo the same way.
I have never found video files work embedded on forums....

Tordan58
26th October 2013, 03:48 PM
http://www.youtube.com/watch?v=oQS2Sl8OibI&feature=youtu.be

Tordan58
30th October 2013, 03:04 PM
Oh dear - that stinks. I feel sorry for you Tord! You probably should complain/claim an EM-1 as compansation for such cruel threatment!?
Olympus have admitted the IBIS is broken. The consumer law stipulates up to three repairs are acceptable, beyond this then the situation needs to be settled between consumer and reseller. Not sure if it needs to be same issue repaired three times, but I think it is the case.

Tordan58
15th November 2013, 01:54 PM
Here is the latest and greatest on this issue.

Today I was called on the phone by Olympus Support :)
The camera has been checked at the workshop. They started by resetting the camera to factory defaults and were unable to reproduce the issue after that. (This reset does things beyond what the menu "reset settings" does).

A representative from Olympus Japan who happened to be on-site at the workshop in Portugal also had a look at the camera and could not find any issue.

They asked me if I could send them the lens to test... :D and I explained that would be a cumbersome and an expensive thing to do. (Apparently they have no legacy lens at all at the WS, and the longest they have is 300mm). I explained that they can reproduce the issue with any lens, any make as long as it is long enough, but they don't have any.

They also agreed that the functionality is supposed to work, since it is supported up to 1000mm.

So I asked them to send the camera back to me.

I can envision one of three scenarios after receiving the camera (hopefully by next weekend?)

[A]Camera IBIS still dysfunctional
[B]Camera IBIS functional, but for a short period of time only (one day)
[C]Camera IBIS functional, and will remain functional for long time

Anyone dare to bet?

pdk42
15th November 2013, 11:39 PM
Here is the latest and greatest on this issue.

Today I was called on the phone by Olympus Support :)
The camera has been checked at the workshop. They started by resetting the camera to factory defaults and were unable to reproduce the issue after that. (This reset does things beyond what the menu "reset settings" does).

A representative from Olympus Japan who happened to be on-site at the workshop in Portugal also had a look at the camera and could not find any issue.

They asked me if I could send them the lens to test... :D and I explained that would be a cumbersome and an expensive thing to do. (Apparently they have no legacy lens at all at the WS, and the longest they have is 300mm). I explained that they can reproduce the issue with any lens, any make as long as it is long enough, but they don't have any.

They also agreed that the functionality is supposed to work, since it is supported up to 1000mm.

So I asked them to send the camera back to me.

I can envision one of three scenarios after receiving the camera (hopefully by next weekend?)

[A]Camera IBIS still dysfunctional
[B]Camera IBIS functional, but for a short period of time only (one day)
[C]Camera IBIS functional, and will remain functional for long time

Anyone dare to bet?

I think the odds are something like:

A) evens
B) 5-3
C) 10-1

Place your bets!

Tordan58
2nd December 2013, 09:24 AM
I got the camera back the other week. To recap, Olympus could not find any issues with the IBIS. A factory reset remedied the issue.

I have been using the camera at three occasions, a few hundred pictures taken at 500mm FL. So far, so good.

Still, there must be something fishy in the IBIS design or software or the camera would not have behaved the way it did.

Case permanently closed, hopefully.

Ian
2nd December 2013, 12:12 PM
I got the camera back the other week. To recap, Olympus could not find any issues with the IBIS. A factory reset remedied the issue.

I have been using the camera at three occasions, a few hundred pictures taken at 500mm FL. So far, so good.

Still, there must be something fishy in the IBIS design or software or the camera would not have behaved the way it did.

Case permanently closed, hopefully.

Fingers crossed that all's well that ends well!

Ian

pdk42
2nd December 2013, 10:08 PM
I sort of live in dread that the IBIS will fail one day. It's a fantastic feature, but it's an electro-mechanical system and as such is guaranteed to have a higher failure rate than electronic components. We can add the shutter to the list too.

It would be good if Oly published an MTBF for the IBIS - would make interesting reading.

Tordan58
2nd December 2013, 10:59 PM
I had an interesting discussion with Olympus Sweden support today - what measures to take in case the IBIS exhibits the behavior that caused me to send the camera to repair twice.

I was told that the workshop have identified issues related to using the EM-5 in combination with long focal lengths (1000mm). I am close to that (and actually even above at times when combining my scope with EC14/EC20 for more extreme conditions). A firmware update that remedies the issue might be released, but this is not yet confirmed.

I also was told that you can come quite close to resetting to factory settings you remove the battery for at least 24h (I did not know about this, the workshop have electronics/SW tools to force a reset).

My thinking: since the behavior was identical at both occasions, it could be that the IBIS was not defect the first time, and that the workshop exchanged it to be on the safe side. And the issue might have to do with SW rather than electro-mechanics (hence the potential firmware update).

Should it happen again, I will do before being out of camera for the two weeks it takes to get it fixed. ANd keep you posted of progress and status if you are interested.

My contribution to statistics regarding IBIS MTBF

4 months/ > 5000 pictures taken from 600mm to 1000 mm
1 hour / 10 pictures at 800mm
> 1 week / > 200 pictures at 500mm (I haven't dared to test with the long scope yet. Maybe the coming weekend, weather pending).

PeterBirder
2nd December 2013, 11:56 PM
It would be good if Oly published an MTBF for the IBIS - would make interesting reading.

I have to disagree.
I have had considerable experience in my days as a radar systems engineer in running a project in which we had to demonstrate a specified reliability level for an MoD/NATO project in order to secure acceptance and payment. Having needed to determine reliability figures for all replaceable sub-assemblies we then had to spend many months in meetings with civil servants to persuade them that these figures didn't mean that they didn't need to buy any spares! MTBF's relate to system reliability and take account of repair time and a host of other factors. Component Failure Rates are the measure of the reliability of an individual item and are predictions, again based on a number of assumptions and are only meaningful in the context of a significant statistical sample. An individual component or a number of those components may still fail in a short period of time but the overall mean or average failure rate of the total sample still be achieved.

To quote a failure rate for a unique component like the IBIS would be meaningless or at best misleading.

Regards.*chr

ps. I believe Olympus do quote some type of "expected number of shutter operations" for many cameras but there is much more historical/empirical data available for these components than for the IBIS mechanism which is a recent development.

pdk42
4th December 2013, 09:58 PM
Hmmm... Well, they do publish an MTBF for shutter life so I don't see the difference. I do understand that the M in MTBF is "mean" and that an MTBF of 1 million hours doesn't mean it can't fail after 10 mins. I would think by now that Oly would have enough data to give some sort of meaningful MTBF, and perhaps even a standard deviation.

I think it would genuinely interesting - esp if the MTBF of the IBIS is less than the shutter.

PeterBirder
4th December 2013, 10:56 PM
Hmmm... Well, they do publish an MTBF for shutter life so I don't see the difference. I do understand that the M in MTBF is "mean" and that an MTBF of 1 million hours doesn't mean it can't fail after 10 mins. I would think by now that Oly would have enough data to give some sort of meaningful MTBF, and perhaps even a standard deviation.

I think it would genuinely interesting - esp if the MTBF of the IBIS is less than the shutter.

Sorry but that is not correct.
They do not publish MTBF's for shutters.They publish "expected number of shutter operations". This is related to the shutter wearing out which is not a measure of reliability.

For an explanation of MTBF see this. Mean time between failures - Wikipedia, the free encyclopedia

The figure you are looking for is the Failure Rate. see here. Failure rate - Wikipedia, the free encyclopedia

A key statement in this article is:

" Failure rate data can be obtained in several ways. The most common means are:

Historical data about the device or system under consideration
Many organizations maintain internal databases of failure information on the devices or systems that they produce, which can be used to calculate failure rates for those devices or systems. For new devices or systems, the historical data for similar devices or systems can serve as a useful estimate."

The IBIS of the OMD E-M1 is an entirely new design. It is "similar" to that of the E-M5. However the E-M5 has only been in service for less than two years and therefore the data sample available is unlikely to be large enough to give a meaningful statistic. I'm sure that Olympus will have made reliability assessments in their design process but these will not be sufficiently accurate for them to release them to users.

As always these things come down to the old adage "lies, damn lies and statistics". If you don't have enough samples of the data you are using to generate a statistic the statistic will have a very large margin of uncertainty.

Regards. *chr

bilbo
5th December 2013, 07:49 AM
As always these things come down to the old adage "lies, damn lies and statistics".Actually, just statistics.

Wee man
5th December 2013, 11:12 AM
surely you mean just lies?

bilbo
5th December 2013, 11:53 AM
surely you mean just lies?No. Olympus haven't released any MTBF figures so how could they lie?

pdk42
5th December 2013, 11:25 PM
Well, I'm clearly not a statistician but I would think that the MTBF and failure rate have an underlying relationship. A high failure rate will reduce the MTBF surely?

In any case, I'm not really trying to argue about the precise form of statistical measure, merely trying to get some feel as to how long most of us can expect the IBIS to last. I'm sure that Oly have some data on this - derived either from field failure stats or internal test results.

Either way, it would be good to know what that is. Information on shutter life is made available and given that electro-mechanical parts are those which will, in all likelihood, fail first, then I think we should know given that the IBIS is the other main component of this type!

Tordan58
2nd January 2014, 12:46 PM
Keeping you updated: the behavior is back.

After engaging and stabilizing the image, the IBIS disengages after a second or two (some humming can be heard and some vibrations can be felt, not as bad as previous but still).

I will try a reset, taking out the battery for 24 hours and see if that helps.

pdk42
2nd January 2014, 01:40 PM
Maybe it's that Swedish air?

Tordan58
2nd January 2014, 02:19 PM
Hi Paul, what do you mean?

I think the IBIS has shortcomings/SW defects that cause it to (with a certain probability, and under certain circumstances) run out of bounds when used with FL longer than around 500. It has happened three times now, operating conditions have all been well within specifications (I'd say 20C, 10C and latest time 5C), non-condensating humidity.