Suspected D10 shutter failure - General Help and Assistance on using CHDK stable releases - CHDK Forum
« previous next »
Pages: [1]   Go Down

Suspected D10 shutter failure

  • 8 Replies
  • 5257 Views
*

Offline reyalp

  • ******
  • 14140
Suspected D10 shutter failure
« on: 27 / February / 2015, 23:06:00 »
Advertisements
On my last timelapse, 76 of 5831 shots were all black, often with several in a row. If I crank the brightness up I can see some noise and hot pixels in those images, so I'm guessing the shutter failed to open.

I have suspected some of the flickering I've seen in earlier timelapses was hardware related.

Total shoot count in firm info is 84801 at the end of the timelapse, so I guess I shouldn't really complain.

Mecha condition is 0, which AFAIK is what it has always been.

I have a vague memory of other threads about similar problems, but a quick search didn't find much.

This isn't my first choice of camera to take apart :(
« Last Edit: 27 / February / 2015, 23:13:19 by reyalp »
Logged
Don't forget what the H stands for.
*

Offline lapser

  • *****
  • 1093
Re: Suspected D10 shutter failure
« Reply #1 on: 28 / February / 2015, 09:54:11 »
I had a shutter failure on the G1X after about 500,000 shots. It started out with black in the corner of the shots with fast shutter speeds, and progressed to totally black shots.

The flickering happens at shutter speeds faster than about 1/1000. I especially noticed it on the SX50. So I always set maximum speed at 1/1000 and use the ND filter for exposures brighter than that.

I think the shutter wears out faster at high speed also. I haven't had any more shutter problems since limiting the speed to 1/1000 on the G1X (1/800 on the SX50), and I've done a LOT of time lapses.

I don't think you'll be able to fix anything if the shutter is bad. I sent my G1X in to Canon repair and they replaced the entire lens assembly for $240. It's a flat fee to repair it regardless of the cause. The D10 should be a lot cheaper.

The SX50 portion of this time lapse starts at 2:00. Note the flickering while the sun is up, and how it stops after the sun goes down (and the shutter speed slows below 1/800).

https://www.youtube.com/watch?v=JVs5nD0LZgI

Logged
EOS-M3_120f / SX50_100b / SX260_101a / G1X_100g / D20_100b
https://www.youtube.com/user/DrLapser/videos
*

Offline reyalp

  • ******
  • 14140
Re: Suspected D10 shutter failure
« Reply #2 on: 01 / March / 2015, 18:09:23 »
Thanks for the comments, and good to see you are still lapsing :)

On my cameras, I haven't noticed the flickering being restricted to high shutters speeds. I also didn't see blacked out corners before the failed shots. Strangely, I've done several thousand images since without any more failures  ???

I guess if high shutter speeds increase wear that would explain why they limit below what the camera seems to be capable of.

edit:
On further observation high shutter speed definitely appears to be responsible for at least some flicker:
http://chdk.setepontos.com/index.php?topic=11081.msg120974#msg120974
« Last Edit: 12 / March / 2015, 16:20:22 by reyalp »
Logged
Don't forget what the H stands for.
*

Offline reyalp

  • ******
  • 14140
Re: Suspected D10 shutter failure
« Reply #3 on: 07 / March / 2015, 20:21:36 »
Oddly, I'm up to ~99500 shots on this camera without seeing the problem again.

On the run where the black shots occurred, the script did try to use very short exposures (~1/10,000th) but the problem continued even after the shutter speed got to normal ranges, up to 1 second. The other difference was that my debug drawing code was turned off, which could affect timing I guess ???
Logged
Don't forget what the H stands for.
*

Offline lapser

  • *****
  • 1093
Re: Suspected D10 shutter failure
« Reply #4 on: 07 / March / 2015, 21:55:38 »
https://www.youtube.com/watch?v=end2sR2KgF8

I found the test time lapse I posted that shows when the shutter on the G1X failed, starting at 1:26 in the above video.

Quote
The log shows everything going normal and then 2 completely dark pictures followed by the erratic exposure changes. So it looks like the mechanical shutter stuck closed for 2 shots, and then the next one forced it open. I suspect that damaged something that now keeps the shutter from opening fast enough.  It looks ok until the shutter speed gets faster than about 1/20 seconds, then gradually gets worse as shutter speed increases. The speed for the 2 dark frames was 1/724 second, followed by a 1/6 second exposure that forced the shutter back open.

I suspect that your D10 shutter stuck in the closed position and then freed itself. I think my G1X shutter was pretty worn out at 500,000 shots, and never went back to normal after sticking closed. Hopefully, you'll get another 400,000 shots on the D10!

I cut the bad part out and salvaged the rest of the time lapse, after the shutter failure but at slower shutter speeds. The cut happens at 1:24 in the final video:

https://www.youtube.com/watch?v=3JCYDoE_eg8
Logged
EOS-M3_120f / SX50_100b / SX260_101a / G1X_100g / D20_100b
https://www.youtube.com/user/DrLapser/videos
*

Offline reyalp

  • ******
  • 14140
Re: Suspected D10 shutter failure
« Reply #5 on: 07 / March / 2015, 22:48:30 »
Quote from: lapser on 07 / March / 2015, 21:55:38
Quote
The log shows everything going normal and then 2 completely dark pictures followed by the erratic exposure changes. So it looks like the mechanical shutter stuck closed for 2 shots, and then the next one forced it open. I suspect that damaged something that now keeps the shutter from opening fast enough.  It looks ok until the shutter speed gets faster than about 1/20 seconds, then gradually gets worse as shutter speed increases. The speed for the 2 dark frames was 1/724 second, followed by a 1/6 second exposure that forced the shutter back open.
That's interesting. The failure mode I saw is quite different.
Quote
I suspect that your D10 shutter stuck in the closed position and then freed itself.
On the run that failed, there were ~76 closed shutter shots distributed pretty evenly over the ~5800 total. In the attached plot, the single points away from the lines are completely black images. Shots in between appear normal, and I subsequently did over 10k shots in a couple of runs without any failures.

Today I did some more testing with more similar settings to the failed run and got 1 failure in 200. ???
Logged
Don't forget what the H stands for.
*

Offline lapser

  • *****
  • 1093
Re: Suspected D10 shutter failure
« Reply #6 on: 08 / March / 2015, 00:25:30 »
I've got over 180,000 shots on my D20 without any problems. I've never let the shutter speed get faster than 1/1000.

My refurbished G1X with the new lens assembly has over 500,000 on it, which is about the same number I had on the original G1X when the shutter failed. I took a lot of shots at 1/4000 with the original, but I've stayed slower than 1/1000 with the refurbished one. Hopefully it will last longer that way.
Logged
EOS-M3_120f / SX50_100b / SX260_101a / G1X_100g / D20_100b
https://www.youtube.com/user/DrLapser/videos
*

Offline c_joerg

  • *****
  • 1252
Re: Suspected D10 shutter failure
« Reply #7 on: 12 / March / 2015, 03:59:27 »
Hello,

nice time - lapse with the G1X. I hope I get it also with my G1X....
Which script you are using?

Quote
G1X shutter was pretty worn out at 500,000 shots

I hope that will by also on my G1X.


Independent to your shutter problems:

Fast shutter speeds are also a common cause of flicker in time -lapse. The reason behind flicker at higher shutter speeds is due to inaccuracy of the shutter mechanism timing at higher shutter speeds. May be the error of shutter is +-0.1ms.  On Tv=1/1000s you have an error of 10% and on Tv=1/100s it’s 1%.

I made a lot of test to investigate of flicker in time – lapse in the following way:
1) I made 150 pictures from an unchanging subject under tungsten light with same settings.
2) I analyzed lightness of every picture (delta, standard deviation)
3) I compared different camera settings.

On the results I could really see different between long and short shutter speeds, or open and closed aperture.
Unfortunately I lost the data with the shutter speed. The problem with closed aperture is now fixed with rawopint.lua because the aperture did not open and closed for measure.


The second think is the ‘180 degree shutter rule’ which is basically also valid for time – lapse. I know Tv=4s in a 8s Interval is not realistic.  But you have to try getting close to it.

https://luispower2013.wordpress.com/2013/03/12/the-180-degree-rule/
http://en.wikipedia.org/wiki/Rotary_disc_shutter
 

That is the reason, why I always use the ND filter and now also close up aperture to get longer shutter times. Next summer, I try to use an external ND 1000 on my G1X…..

Logged
M100 100a, M3 121a, G9x II (1.00c), 2*G1x (101a,100e), S110 (103a), SX50 (100c), SX230 (101a), S45,
Flickr https://www.flickr.com/photos/136329431@N06/albums
YouTube https://www.youtube.com/channel/UCrTH0tHy9OYTVDzWIvXEMlw/videos?shelf_id=0&view=0&sort=dd
*

Offline reyalp

  • ******
  • 14140
Re: Suspected D10 shutter failure
« Reply #8 on: 12 / March / 2015, 16:30:40 »
Just for completeness, as I mentioned in the other thread I'm still not certain if the "shutter failures" I saw were actually a hardware failure, or a bug. At the moment, I'm leaning toward bug, although what mechanism would cause the specific behavior is puzzling.
Logged
Don't forget what the H stands for.
Pages: [1]   Go Up
« previous next »
 

Related Topics


SimplePortal © 2008-2014, SimplePortal