Mamiya M645 inaccurate shutter time?

Discussion in 'Medium Format' started by damien_kunik, Aug 2, 2013.

  1. Hi,
    I recently bought a Mamiya M645 (the original 1/500th one) which is my first foray into medium format. The camera itself looks like it has been very well cared for with zero sign of abuse. I only replaced the gooey light seals, which a job I am accustomed to on 35mm cameras.
    Back home, I tested the shutter time in the most simple way, using a stop watch. This means that I only tested "testable" shutter times from 1s to 8s. I have no more precise instruments around.
    Now, there seems to be an inaccuracy in those slow shutter times and while this question has been around the web here and there, I cannot seem to find a definitive answer.
    When I time a 1s exposure, the shutter seems to be open for a bit less than 2 sec. When I test 2s, shutter is open for about 3 sec. 4s is about 5-6 sec. 8s is about 11-12sec. Nothing super precise in my test indeed, but going from 8s to 11s is obvious enough even with my primitive method.
    Test was first done with a fresh 6v 4LR44 alkaline battery, then repeated with a 6v (labelled 6.2v here in Japan) 4SR44P silver oxide battery. The issue is the same in both case.
    While I never plan to use such long exposure times, I am a bit afraid that the problem might be the same with shorter ones. I'll shoot a test roll anyway, and color neg or b&w film will probably compensate for the noted inaccuracy, but I'm pretty bugged by the issue.
    Do you have any idea about such a problem? Should I care? Or is it normal reciprocity failure compensation built in on purpose for slow shutter speeds?
  2. From what i understand being 30% off is the norm for most mechanical shutters. What your describing isn't too bad.
  3. Well, not sure if this matters as you don't seem too worried about the inaccuracy, but the M645 shutter is electronically controlled, not mechanical.
  4. Short answer, the shutter needs servicing.
    Standard shutter speed tolerances for any shutter is 20% for 1/100 and slower, 30% for speeds 1/125 and faster.
    Some shutter manufacturers such as Compur set their tolerance at 15% for 1/100 and slower, 20% for speeds 1/125 and faster.
    Mamiya possibly uses the tighter tolerance. It does not matter if it is a mechanical leaf shutter or an electronically controlled focal plane shutter.
    Electronically controlled focal plane shutters have 2 curtains that work together to form a slit for fast speeds and the open-closed window for slow speeds. What is most important initially is that the speeds are consistent. If a given speed is consistently 35% slow then one can compensate and get good exposures. If a shutter is 1/2 stop or more off it should be serviced. The 20%/30% tolerance is 1/3 stop.
    You should be able to adapt the shutter speed tester discussed in this thread
    to your camera. The last post are the most up to date for currently available components.
    A shutter that is a stop or more off may quit working and not be repairable. If your camera has sat on a shelf or in a camera bag/case and has seen little use then the most likely cause is the lubricants have dried out.
  5. Okay, thank you very much for your answer and for the link. This tester is a clever piece of hardware. I will definitely build one.
    As I said before, I will shoot a test roll and see if there are some noticeable inaccuracies on the color neg, but I don't think it will show. Dried out lubricant might well be the issue here and the camera has probably sat for a long time unused. I noticed that the more I test it, the more the shutter turns out accurate (although not 100% yet, I would say). So maybe the stiff grease was slightly slowing the shutter down and my repeated test shots helped a bit.
    That being said, my camera is quite definitely in the 20-30% tolerance bracket now. Servicing will probably be in order anyway, at least to have the camera checked. But I really really want to shoot with this week, so I only hope that I will not make things worse.
  6. If you already own a iOS device (ipad, ipod touch, iphone) you could consider Lukas Fritz' little
    APP w. resistor:
  7. I've got the Lukas Fritz device on order. Alternatively, I've previously used anything that works as a video camera (phone, ipad, computer, DSLR, video camera!) etc. I trained the video camera device on the shutter, and recorded the shutter firing at different speeds. Afterwards review the video footage in slow motion - or even frame by frame, counting the number of frames that pass with the shutter open. With knowledge of the video frame rate per second its basic math to figure out reasonably accurately the duration of each shutter opening down to at least 1/8 second. Faster than that and the fractions of a second begin to take less than 2 frames of standard video, and are not reliably visible.
  8. Yes, if it is affecting your exposures, and you are worried about it, send it to a qualified repair person who one, has the equipment to accurately test it, and knows how to fix it if it requires it. I don't think you can get very reliable measurements with a stop watch. Maybe its good enough to make you wonder if you have a problem, but just. Have you tried running film through it and testing the exposures?
  9. Using a stop watch is nothing scientific indeed. It was just an easy -but telling- check.
    I shot 2 rolls today and will have them developed by a trusted pro lab tomorrow. Then I'll see. I obviously don't plan to fix the Mamiya myself if anything seems off. I save my unlicensed surgical skills for old Russian mechanical rangefinders. That being said, the above informations about shutter time testers are pretty interesting.
    On a side note, I'm in love with that camera. While cheap by today's standards, it was not an impulse buy and I was pretty sure that I would be pleased, but it is so much more than that... I'll have a hard time going back to my 35mm SLR now if CLA is necessary. It appears to me that I am definitely a waist-level finder enthusiast.
  10. The M645 uses an electronically timed shutter. The fact that the long speeds are way off indicates that it's a problem with the electronic timing circuit (cheap to repair), and nothing to do with the mechanical parts of the shutter (expensive). Otherwise the symptom would be inaccurate fast speeds or a varying density exposure across the frame.
    Primitive electronic shutters like this used a simple CR (capacitor-resistor) timing circuit. Most likely suspect is the capacitor. If the camera has lain unused with a dead battery, then the capacitor may have become "leaky" - not physically, but leaking electrical current instead of holding it.
    Damien, cross your fingers, touch wood and maybe say a prayer to whatever deity you believe in; then simply power up the shutter and fire it for a few long exposures. There's a slim chance that simply applying voltage to the capacitor will gradually heal up the leakage. This will be indicated by exposure times getting shorter, eventually getting close to or down to their marked speed. If that doesn't happen (may take a few days of intermittent use) then the camera needs to go off to a repairer for a replacement cappy or adjustment. But good luck anyway!
  11. I did this
  12. Primitive electronic shutters like this used a simple CR (capacitor-resistor) timing circuit. Most likely suspect is the capacitor. If the camera has lain unused with a dead battery, then the capacitor may have become "leaky" - not physically, but leaking electrical current instead of holding it.​
    That capacitor ("condenser" in older parlance) would be the "exposure time control condenser (A)" in this schematic from the M645 user manual. The rate at which capacitor (A) is charged, or its ability to hold all that charge, is where the fault lies with Damien's camera.
  13. And just to show how accurate the M645 shutter can be when the charge timing is properly calibrated, this is from a test review in 1975:
  14. Okay, the negs came back from the lab today. I've asked to have them scanned to check for inaccuracies on my computer screen and at first, highlights looked pretty blown out. I was devastated, as I am quite frankly a beginner and the learning curve is quite steep here: 1) first time shooting medium format 2) quite new to the use of a handheld light meter 3) and all that with a camera that I don't know and that showed seemingly inaccurate shutter times to begin with.
    Now, as I still don't have more precise instruments, I checked again tonight with a stop watch and the shutter times are way more accurate. Maybe it's the camera coming back to life, but it's getting better the more I use it. So why the blown-out highlights?
    Then I checked the negs themselves. There is so much more informations in there! Clouds have texture and skies have shades while it's just big white zones on the scanned image. Could it be that the lab seriously botched the scanning process and that my camera is fine?
  15. Oh, and I should add that I used an 81A warming filter for the test to compensate for the slightly longer shutter times. Well, this is somehow counterproductive for the test but I somehow wanted the shots to turn alright as those are my last days in Japan for quite some time and there are still so many things I wanted to capture here.
    Thank you all for the informations on the electronic issues that an old Mamiya 645 may face. I will definitely have the camera CLA'ed the next time I am in Tokyo, but where I am heading for the coming year, finding a repair shop is nearly hopeless.
  16. Damien, from your description of the negatives, the exposures were fine.
    The scans are clipped in their dynamic range. What format are the scans in? If they're just jpegs, I would not be surprised at the blown highlights - it's just like the way that labs making machine prints from C41 negatives tend have always tended to blow out the highlights. But a 16-bit TIFF or DNG file should not be blown out.
  17. Thank you for your answer, Ray.
    Yup, those are jpeg files and a look at the histogram on Gimp shows that the dynamic range is badly clipped. I can't believe I am paying for that...
    Well, I'm leaving Japan and all of the second hand market goodness but back home, I'll have at least a proper dark room and a good scanner to do things right.
  18. The other thing I just thought of is something you hinted at ("maybe its coming back to life"). In Lecia M's it recommended to work the shutter through all the speeds several times (without film).
  19. The other thing I just thought of is something you hinted at ("maybe its coming back to life"). In Lecia M's it recommended to work the shutter through all the speeds several times (without film).
  20. I did this:

Share This Page