[unkipunki]

On the plus side there hasn't been a new incident report for 5 days. Have all the K3 owners been too busy to take photos this weekend, or have they been particularly reliable due to it being May?

[audiobomber]

Well, it depends. Originally I also thought, that the reaction is a bit over the top, but there are times, when the photo opportunity happens just once and it can't be re-done. An example: weddings. "You may kiss the bride" and K-3 goes into mirror-y nirvana. Given the fact, that the usual duration of that kiss is about 0.0008 seconds on average (speaking from experience), the photog may as well nail himself next to JC, if THAT happens. In a situation like this even the backup camera won't help, this IS the only split second available for that particular shot. I agree that this is a highly specialized application, very likely that most of K-3 owners will never encounter situations like these, but for some even the theoretical knowledge, that there's an issue lurking in the shadows, may be unnerving.

Let's keep it real. There's a subset of K-3 owners with the mirror flap problem. There's a small subset of that subset who are pros. From there, how likely is it that the camera happens to misfire at the crucial time?

---------- Post added 2014-05-05 at 22:13 ----------

I lost a great shot a couple weekends ago when it happened to me for the first time (with fully charged batteries in grip and body, sir audiobomber).

Then I would say you have a different problem than I do.

[Rupert]

I don't have a K3, but probably should. I'm still well pleased with my K5 original, but the extra MP of the K3 are enticing.

Not fully understanding all this flapping...and it probably would not be a problem for me, since any wedding I might shoot would be a squirrel wedding ...and for a few peanuts, they would pose the wedding kiss as many times as I wanted.

However, I am wondering how this flapping action, if frequent enough, would affect the mechanics of the camera? Could it harm the long life of the mirror mechanism? This would be a concern for me, since I tend to keep a camera a long time.

Regards!

[karro]

audiobomber - agree, the probability is small, minuscule, even. whether this poses a real problem for the owner, is for the owner to decide. should this hold someone back from getting the K3? no, no and NO! it is the best camera Pentax was able to create so far.

Rupert - reading the threads, I've got the feeling that for those, who've experienced problem more than once, it "kinda/sorta" tends to happen each 5-6k shutter activations or so. IF this is true, I'm close
As for the mechanics - the camera is designed to withstand series of 8.3fps. When it happened, I didn't spend time analyzing or recording the sound, it was more like "wait, what?.. power off! WTF?? Battery out!", but during the process it sounded like about that speed, maybe a bit faster - which is understandable, because neither aperture nor shutter were activated.
Let me put it this way: if this can break the camera, I'd say, that series of 8.3fps are much more likely to break it.

[Rupert]

Rupert - reading the threads, I've got the feeling that for those, who've experienced problem more than once, it "kinda/sorta" tends to happen each 5-6k shutter activations or so. IF this is true, I'm close
As for the mechanics - the camera is designed to withstand series of 8.3fps. When it happened, I didn't spend time analyzing or recording the sound, it was more like "wait, what?.. power off! WTF?? Battery out!", but during the process it sounded like about that speed, maybe a bit faster - which is understandable, because neither aperture nor shutter were activated.
Let me put it this way: if this can break the camera, I'd say, that series of 8.3fps are much more likely to break it.

That makes sense. I am wondering what would happen if you just let this flapping run its course...would it stop eventually, or would it continue until the battery was drained? It would take some daring to experiment, but it might shed some light on the situation? Sounds like the camera is getting a message to shoot at a high frame rate...when it is not....which would be a software problem?

Regards!

[Circuits]

On the plus side there hasn't been a new incident report for 5 days. Have all the K3 owners been too busy to take photos this weekend, or have they been particularly reliable due to it being May?

Well it just happened to me for the first time this morning.

[audiobomber]

There are two differences (major in my mind) between the K3 and the K5 series and these are: 1) the K3 uses two separate motors to operate the mirror and the shutter and 2) they added a damper to the mirror to reduce vibrations from the mirror going up. The K5 series uses only one motor and it is unclear to me if they had any type of damping on the mirror. Thus comparing the 2 cameras is apples and oranges. Using two separate motors I would think that a sensor of some sort is required to determine that the mirror is in it's fully up position before the shutter is allowed to operate. Otherwise you could be taking pictures with the mirror partially up. So with a low battery, the mirror motor might not have enough power to raise mirror (overcoming the additional effects of the damper as well), causing the mirror to return to the down position without triggering the shutter since the camera didn't sense the mirror fully up. Since the shutter wasn't triggered the camera could then cause the mirror to go up again, repeating the same scenario over and over.

This seems to be what happened to my K-3 (twice). The mirror was flipping so rapidly that I doubt it was reaching the stop. Unfortunately I did not look in the OVF to confirm, I was busy trying to stop it. The AA batteries in the grip were almost dead. The flop stopped after about 5 seconds, probably because the batteries were too exhausted to continue.

---

Last edited by audiobomber; 05-06-2014 at 08:59 PM.

[Uluru]

Not fully understanding all this flapping...and it probably would not be a problem for me, since any wedding I might shoot would be a squirrel wedding ...and for a few peanuts, they would pose the wedding kiss as many times as I wanted.

No one really understands it, and that is the "problem inside the problem". Being natural problem solvers, people immediately jump into pre-conclusions, using established paths and shortcuts we are accustomed to do in order to get to the bottom of problems quickly. Including the OP, of course, who wrote all about it all over the internet.

That is our user's and human's fault, or, a very difficult trait if you prefer. It should be tolerated and educated. Never escalated and turned into a massive paranoia and witch-hunt.

Any comparison with previous K-cameras and previous experience is hardly possible because the new K-3 has a redesigned shutter and a mirror-box. It has a new software too, new chassis, different energy requirements, different fps rate, different internal air dynamics, different internal thermal dynamics, you name it.

There are so many unknowns that what we see as a "problem" (flipping mirror) may in fact be a remedy for a certain situation and a much bigger problem camera operates in and tries to remedy itself. It may as well be a kind of alarm. But we shall never know that if we pre-conclude a "problem".

Instead of pre-concluding anything — and especially lack of QC on Ricoh Imaging's side — it is best to assemble evidence from reputable sources, making sure people who report it also confirm their statements with photos of their cameras, and photos of the bottom plate with serial numbers clearly visible.

After a certain number of proven cases has been collected, a document can be created and sent to Ricoh Imaging Japan, asking their understanding and offering user's co-operation in solving the mystery behind the issue.

Half the solution of any problem is in good attitude.

[Rupert]

It should be tolerated and educated.

But we shall never know that if we pre-conclude a "problem".

Uluru.......That is one of the finest "spin jobs" I recall ever reading. It compares to runaway car acceleration as being seen as a possible benefit, where......for some magical reason, the car senses a carjacker in the vicinity and decides on its own to get you out of there at 90 mph.......against you will!

It is not the purchasers obligation to make a diagnosis and repair of an obvious defect, it is the responsibility of the manufacturer, in this case Ricoh. I can't imagine anyone with this problem seeing it a some hidden benefit and writing Ricoh to thank them......can you?

I haven't seen any Pentax bashing in this thread, and don't see a need for it, but covering for an obvious problem with acceptance is not going to get a solution from Ricoh. Some guy at Ricoh/Japan may read your post and say "Look see! Shooter boys like flapping mirror".

Regards!

[DRabbit]

I'm not exactly sure that I'll ever understand people's need to explain a problem away. I'd certainly be on board with you if I was the only one that ran into the issue, or it was limited to merely three users, but based on the level of activity in this particular K3 forum itself, I'd call the mirror-flapping issue anything but rare. Do you realize we've collected more reports of this issue than the forum has collected serial numbers of owners itself, and double the amount of positive reviews that exist on Amazon for the camera?

Ultimately, it's an issue that effects a wide range of serial numbers. We've likely exhausted the members here with the issue, but that doesn't mean it should be swept under the rug, thought of as a "feature" or ignored by Pentax.

In response to the notion of shooting weddings, Karro brought up the point of missing the shot. This is part of the reason it was an issue for me. What I'm shooting professionally cannot be recreated. If a shot or footage (from time-lapse) is lost, it's gone forever. When a client is relying on your to get that shot, no backup camera is going to help once the moment is gone. If it only happens, on average, every 3000 shutter actuations, you can say all you want that it's "uncommon" and shouldn't matter, but if your paying client ends up being the one to suffer the lost photos, good luck explaining it to them.

But, I also brought up weddings as an example of a quiet atmosphere where the mirror-flapping issue could actually do more than just cause you to lose the shot. I'd hate to think that in the middle of a quiet church, the camera would suddenly become a disruption itself, ruining the moment for the couple, their guests and even the videographer. Not only might you lose the photo, but you could easily become the focus of some pretty angry clients. Weddings, Christenings, Vow Renewals, Corporate Keynote Speeches... any paying gig in front of an audience in a quiet venue would cause me a lot of concern... and could be reason enough not to rely on the K3. Oh, and back to that once every 3000 number again... if you're a busy wedding photographer, it's not unheard of that you'd be shooting 1000 photos a week. Would it be acceptable for you to run into this issue once a month or every two months? That could be a lot of partially-botched weddings.

Yes, maybe professionals shooting weddings are a subset of a subset of K3 owners... but if a camera company is trying to grow its brand and wants a good reputation, having an issue like this crop up with enough reported cases isn't good for them - and that's bad for everyone, including the casual shooter. Look at Fuji as an example with it's light-leak issue. I'm sure XT1 owners who shoot astrophotography are a subset of a subset of Fuji owners, but Fuji still tracked the issue down quickly, addressed it, and repaired the cameras of anyone who had the problem. I tried to research the issue as well to see how many "reports" through forums they had, and I seem to only be able to count up about 50ish.

As stated earlier, we've had at least two reports of people who had the issue that, at least in theory, seemed to lead to other problems. One user had the camera cease to function at all after a mirror-flapping incident. Another had material spewn onto the sensor and into the viewfinder after a mirror-flapping occurrence. Call both anecdotal or coincidental if you want, but given the violent nature of the runaway mirror itself, it's not hard to imagine that it might cause other problems, even if it only happens to you once.

For me, it comes down to reliability. As a professional photographer, if one particularly camera is less reliable than another, I have no choice but to pick the best one for the job. I chose Pentax initially because part of what I do is commercial time-lapse. As a camera with a built-in intervalometer you would assume Pentax wants to be known as a camera maker who produces a body that can accomplish the job. But with the K3, that isn't the case. I'm happy to report that the K5-II and K5-IIs seem to operate flawlessly, and as intended... but that just isn't so with the K3. It's not an over-reaction to state so, as it's clearly a problem.

Last, but not least, inside of this thread we haven't even addressed other reliability issues with the K3. Along the way I had the camera lock up a lot more than any other camera I've ever shot with (by far). I'm talking about lockups without mirror-flapping, just silent, camera-will-not-respond, freezes that required the camera to be turned on and off several times, or the battery to be popped. You can see one of those lockup right in my video in the original post at the beginning of this thread. I also had lockups in the middle of shooting time-lapse, where the camera just stopped shooting, but it actually looked like it still was shooting (in the sense that if you hit the menu button, it would show you the status screen as if all was fine - it wasn't unresponsive, just not shooting anymore). These odd silent lockups occurred across all firmware versions, and happened with both K3s I owned. So, there's an even bigger issue of overall reliability that could be discussed here. Again, maybe not one that occurs ever time you turn the camera on, but one that certainly caused a lot of concern for me, especially in comparison to other bodies that seem to perform much more dependably.

Ultimately, at least in theory, it seems Pentax is trying to address the issue behind the scenes by stating that Japan is working on it. Since they haven't been more transparent or public about it, all I can do is assume it's true. I do hope that's the case - for all of us who want Pentax to continue to make great cameras into the future, and for the subset of the subset of us who really want to use these cameras professionally.

[jpzk]

@ DangRabbit:

Excellent post No. 206 !!

Cheers!

JP

[agkchang]

Ditto post #207 and #207.

---------- Post added 05-07-2014 at 03:12 PM ----------

Meant to say post #206 and #207.

[peterh337]

My K3 did it almost straight out of the box. I don't think I shot more than a dozen pics with it.

But it happened with a non Pentax battery - Hahnel HL-PL90 1560mAh. That battery wasn't cheap and as far as I can tell it wasn't a fake like so much stuff on say Ebay.

It hasn't happened again with the Pentax battery but I was not able to reproduce it no matter how hard I tried, even when running the battery right down.

So it is either battery related, or the whole battery charge status business is a redherring and something else is causing it.

As an electronics engineer, I am pretty sure it is caused by a dip on the internal supply rail (caused by the heavy current drawn by the motor(s) operating the shutter) which crashes the processor. There won't be a simple firmware fix for that (if it is that) other than to somehow reduce the current drawn by the motor. Nokia had exactly this issue on the 808 phone and they seem to have 99%-solved it by slowing down the transmit rate... nobody will notice

[fcrespo]

thanks peterh337, interesting analysis.

I think I am the only victim that suffered a total K3 failure after the mirror flap event. I remember vividly that the battery was very low, with only one tick in the K3 LCD battery monitor.

This is consistenty with peterh337 hypothesis, hinting that one should avoid getting close to the battery threshold.

By the way, I sent the K3 for repair on 5th March and I am still waiting (8th May today!). They tell me (www.photo-univers.fr) it will be shipped back next week.

[peterh337]

It is certain that Pentax can control the current drawn by the motor(s), with firmware.

The motors will all be driven via D-A converters, with ramp-up and ramp-down profiles done to minimise resulting camera movement during the shooting. The DACs will either be receiving data packets from the firmware, or they will be implemented (this is a cheap and dirty way of getting a large number of DAC channels) by pulse width modulation of general-purpose I/O pins on the CPU (i.e. entirely in firmware) and lowpass filtering the waveform. Either way, the firmware has total control over how much current the motor is going to be drawing at any instant.

So if the motor current is the culprit, a quick and dirty firmware fix is clearly possible. It would possibly involve some "corporate subterfuge" in that the maximum achievable camera frame rate might go down a bit, but who will notice?

It's like dropping the Nokia phone's max uplink speed from say 10mbits/sec to say 2mbits/sec, in order to reduce the power drawn during 3G/HSPA transmission so the internal supply rail doesn't drop too much. You are not going to advertise what you did and nobody is going to be measuring it - especially as data rates are never the same twice and half the time the mobile signal is cr-p anyway.

Given typical corporate cynicism, one would not expect Pentax to officially fix this. That would be all over the forums and would open the floodgates to (a) everybody with a K3 who is too "IT-dumb" to upgrade the firmware returning their camera under the warranty and (b) all stock in the distribution channel being returned to Pentax. Commercially, they cannot do that until all affected K3s are out of warranty (i.e. they might do the "Sony infolithium battery job" on it). After all, they have not even admitted they are aware of a single person reporting this issue. All they will admit to is that they have read this forum. The best we can expect is a fix sneaked out under the guise of a firmware upgrade to fix something innocuous, or maybe one which "improves general software stability" (time to go for a University of Upper Warlingham MBA in corporate bull---t ). That is how serious cockups are fixed these days. You cover your back first, and sneak out a fix quietly.

If they are really clever, they might do a fix which slows down the frame rate only when the battery is getting a bit low.