1. Reminder: Please use our affiliate links for holiday shopping!

E-M1 purple bar issue

Discussion in 'Olympus Cameras' started by mpg01, Oct 16, 2014.

  1. mpg01

    mpg01 Mu-43 Regular

    154
    Oct 21, 2012
    Has anyone experienced this purple bar issue? This is not the uniform banding seen on the E-M5.

    I was doing some night sky test shots with the E-M1 with the 25mm f1.8 and noticed a very distinct purple bar along the top edge of the exposure. I did some experimenting and found that it seems to appear between 1" and 0.5" shutter speed and between ISO 640 to 1600, all at f1.8. I didn't do a comprehensive aperture bracket but still noticed it up to f2.2. While the exposure indicator was not flashing I suppose this range could be considered under exposed for a night sky with average light pollution. I also tested with the 45mm f1.8 and got the same results so the rules out a lens issue.

    At ISO 2000 and above, for the same speed and apperture, the band disappears. I gong back through my images with the same exposure for the last year I came across another night sky image with the same issue so it's not new.

    My concern is, could this be a curtain or some mechanical issue? I've attached an image to illustrate the issue.
    PA150010-1.


    If it's just exposure and others experience this then I can just make sure I don't shoot at this exposure. I have contacted Olympus support but I thought I'd pass it by the community as well.
     
  2. barry13

    barry13 Super Moderator; Photon Wrangler

    Mar 7, 2014
    Southern California
    Barry
    I haven't noticed it, did some shooting of the eclipse...

    Barry
     
  3. pdk42

    pdk42 One of the "Eh?" team

    Jan 11, 2013
    Leamington Spa, UK
    I'd be surprised if it's due to a shutter problem - the line is too diffuse and in any case it would be unusual at such a long shutter speed. The colour cast is hard to explain too if it were due to the shutter. The diffuseness tends to indicate an optical problem rather than digital electronics, although again, the purple colour is hard to explain. If I had to hazard a guess, I'd say it's some reflection off the edge of the frame surrounding the sensor - it might be worth an inspection with the lens off and see if there's any black paint missing or such like. Long shot I know...
     
  4. kwalsh

    kwalsh Mu-43 Top Veteran

    775
    Mar 3, 2012
    Baltimore, MD
    Weird. Purple usually indicates one of two things - some sort of unusual flare in the lens/mirror box or else something making all the RGB pixels on the sensor go up together. So for instance, if the edge of the sensor were getting hot we'd see increased dark current in all the pixels near the edge of about the same amount regardless of whether they were R, G or B. When the camera white balance is applied these equal weighted R G B values become purple.

    This really doesn't look like either of those things, but it is hard to tell whether this is a 100% crop or a downsized full frame. If we could see the edge at 100% it would be clearer if this is optical or sensor related. If it is repeatable the obvious thing to try is an exposure with the lens cap on - if you still get the purple then it must be the sensor.
     
  5. mpg01

    mpg01 Mu-43 Regular

    154
    Oct 21, 2012
    I truly appreciateall of the feedback. After many more test shots, trying different settings, I've discovered that this only appears with anti-shock (0sec) on. I'm able to pretty consistently reproduce that fringe @ the setting listed in the OP. Shooting without Anti-shock or setting it to anythinng other than (0sec) and the issue disappears. So that means under 1/320sec the first curtain is elecronic rather than mechanical. I'm not sure how that might acuse this but at least I have something solid to talk to Tech support about. It's also good to know I can work around it.
     
    • Like Like x 2
  6. kwalsh

    kwalsh Mu-43 Top Veteran

    775
    Mar 3, 2012
    Baltimore, MD
    Hey, great detective work! This should be something relatively easy for other E-M1 owners to try and reproduce if you can describe a reliable test case. It could in fact be something true of all E-M1s given it is a pretty specific case it which it happens and perhaps other people haven't noticed it yet.
     
  7. barry13

    barry13 Super Moderator; Photon Wrangler

    Mar 7, 2014
    Southern California
    Barry
    Hi,

    I'm not seeing it on mine, although I'm not sure I'm shooting the same lighting conditions; the sky is clear and dark right now, but I also tried some indoor shots with overhead lights on (above the frame and in the frame).

    Tried the 12-40mm and 45mm lenses, wide open, with Anti-Shock enabled, various shutter speeds from 0.5" to 2".

    Barry
     
  8. jaeae

    jaeae Mu-43 Rookie

    10
    Apr 14, 2014
    Hmm.

    Definitely looks like "amp glow" to me. Google it, many cameras have (or had) it. I was in an impression that it was mainly CCD issue, but I might be wrong.

    I once had the E-400 slr, and it had quite bad purple glow..

    The long-exposure dark frame substraction might just clean that glow away, which could explain the occurrence mpg01 described.

    - at short exp.times there's not enough glow to be visible (and not much "hot pixels")
    - long exposures (over 1 sec? ) use second exposure as do dark frame substraction in order to remove "hot pixels" --> also no purple glow
    - between those exp.times, dark frame is not used --> glow visible
    - as ISO sensitivity is set higher, second dark exposure is taken sooner (shorter exp.times) --> no glow above iso 1600
     
  9. mpg01

    mpg01 Mu-43 Regular

    154
    Oct 21, 2012
    Barry, thanks for taking the time to test. I'm glad you're not seeing the same effect. I think amp glow is pretty spot on. I'll add that to my correspondence with Olympus tech and see what they say. Thanks again.
     
  10. barry13

    barry13 Super Moderator; Photon Wrangler

    Mar 7, 2014
    Southern California
    Barry
    fwiw, I have dark frame subtraction disabled.

    Barry
     
  11. fortwodriver

    fortwodriver Mu-43 Top Veteran

    959
    Nov 15, 2013
    Toronto, Ontario, Canada
    Frank
    On a long exposure, the battery draw can cause the bottom of the camera to warm up. It may just be warming up enough to make the sensor glow. Keep in mind that whatever appears at the top of the frame, is occurring at the bottom of the sensor.

    If you turn on DFS does it go away?
     
  12. eteless

    eteless Mu-43 All-Pro

    Jun 20, 2014
    Normally without electronic first curtain the entire sensor is powered up before the shutter travels, with electronic first curtain the mechanical shutter is replaced by the electronic wipe across the sensor. I can think of reasons why it would happen with the EFCS disabled, but not with it enabled.

    If the shutter were leaking light before it traveled the bottom area where the shutter opening is parked would have a light leak like what you show... but with EFCS it should be immune to that, so I have no clue.
     
  13. JimUSNY

    JimUSNY Mu-43 Regular

    122
    Nov 5, 2013
    Mid Hudson Valley NY
    Jim
    interesting wonder why they would even have a 0 second anti shock setting? wonder how it even works on a mirrorless system, just turn the sensor and off on after desired time with no shutter?
     
  14. eteless

    eteless Mu-43 All-Pro

    Jun 20, 2014
    While they can do that you get weird rolling shutter artifacts as it's a cmos sensor(they can in theory design it do to global shutter, however everything comes at a price). The read speed (thus shutter travel time) is limited by the bit depth you read off the sensor, thus to get the highest bit depth they use a mechanical shutter to block light.
     
  15. nicksoti

    nicksoti New to Mu-43

    9
    Aug 16, 2015
    Hi everyone,

    sorry for bad english.

    I came across this thread as i have same problem with purple bar.

    Mine does such at about iso 1600 and up, 1/2 sec and slower in dark scenes. Sometimes whole picture's highlights get that purple color besides top bar.

    Also sometimes in lower iso only visible after lifting darks in orf file 2 stops.

    Should I sent it for repair? What was the final verdict after mpg01 contacted Olympus service?

    I am really worried as i use camera for paid work.

    Thank you in advance,

    Nick, Greece.
     
  16. mpg01

    mpg01 Mu-43 Regular

    154
    Oct 21, 2012
    I never did hear back from Olympus but the solution for me is to not use the 0 second anti shock setting in my nighttime shooting. Never a problem once I do that. I think eteless summed up the issue and I don't think this would be something that is repairable. It just seems to be the outer fringes of the cameras capabilities.
     
  17. nicksoti

    nicksoti New to Mu-43

    9
    Aug 16, 2015
    Thank you very much for reply.

    Actually i asked 2 members of biggest greek forum dpgr.gr to do some tests.

    1st copy showed no bar at all.
    2nd copy showed the bar, bud at less extend than mine. Also the file showed less thermal - electronic noise than mine, like it was at half stop iso less.

    Maybe it is a sample variation. However, i consider this as a problem. half second exposure should not show any purple bar or extra noise at all, even with 1st electronic curtain.