LF shutter accuracy?

greenspun.com : LUSENET : Large format photography : One Thread

Until I recently purchased a hi-tech shutter tester (not Calumet) I suspected, but didn't really realize how inaccurate our traditional mechanical shutters were, especially at the higher shutter speeds...until I purchased this hi-tech solid state shutter tester. When I think of how man sheets of Ektachrome I used in bracketing, I shudder to think of the expense, over 40 years. Does anyone know the tolerances, percent wise, for shutter error in our traditional mechanical shutters. I had heard that it was 10% up to 1/30th. and 25% above that. ALso, has anyone heard anything about a more accurate shutter that might be in development, or are we just the old sytle 'traditional' types, that have been passed by,..by the high tech world of 35mm and medium format?

-- Richard Boulware (boulware-den@att.net), August 15, 2001

Answers

Richard,

I don't know the specs for mechanical shutter error, but could you please reveal to us the discrepancies for marked speeds for any of the shutters you have tested?

I'd be interested in knowing the difference.

-- ed (ekang@cse.nd.edu), August 15, 2001.


If I'm not mistaken, there may be some info on shutter accuracy here on this site - I recall seeing some kind of test that involved the two or three types of common shutters in different sizes, and their accuracy at various speeds - I seem to recall the range was fairly tight - nothing like being off by 25% - but that's a while ago, so I could be mistaken. I'm curious - do you have the results of your testing available?

-- Michael Mahoney (mmahoney@nfld.com), August 15, 2001.

Richard: I've got a digital timer and from time to time check my shutters. I've had a number of older shutters repaired by pros and checked them before and after to see how much improvement I could get. After doing this for 15 years, I've got a few observations:

1. The Copals, unless they're broken, should be very accurate (1/6th stop off or less) on the lower speeds. This would include up through 1/60th, which is the range most LF photographers are using them in. Several I have are within 3% on the lower speeds. A few manage to be within 1%, which is remarkable for a mechanical device. Despite Copal's own specification (I think the sheet which comes with them says 5%, but I'm going from memory, knowing that any error on this web site will be corrected...) they don't do quite so hot on the higher speeds in my experience. You'll see 125th which is more like 100th, higher speeds are faster than lower speeds but how much faster just depends.

2. Error (if they're NOT in need of fixing) is very consistent error. If speeds are off more than 3-5% from firing to firing (usually it is MUCH less than that) they need to be fixed and can be fixed.

3. Older compur shutters (and this is true of most older shutters of all makes in my experience, including supermatics, betax and compounds and compurs) don't go anywhere near as fast as they're labeled. Make your peace with that. If you are a perfectionist and think there is somebody somewhere who can tweak them so the higher speeds are as fast as they're supposed to be, well, I think you're in for disappointment. Again, through 1/6oth they can be fixed and adjusted and work quite well. Things start slipping after that on every one I have, even after servicing. As a run of thumb 1/250th will be about 1/125th, 1/500th will be about 1/250th or so. About a stop off has been the rule with very few exceptions. If the shutters are gunked up and need cleaning the speeds can be much further off, I'm talking about those which have been CLA'd or don't need it.

The really interesting question (to me, anyway) is whether the shutters in their prime really did any better. My compur-rapids are pushing 50 years old. Could they go as fast as advertised back in 1954? The marketing temptation would be to suggest they went faster and faster and faster. We didn't have reasonably priced and very accurate gizmos to check them with, so unless you were repairing them 50 years ago and had a then high-tech way of checking them you're not really going to know. I suspect people didn't know one way or another and exposure precision wasn't great to begin with, which masked shutter error and/or compounded it. The important thing, of course, is to know what the actual speeds are, to make sure they are consistent on those values whatever they are, and then go use them. To answer your final question (or to not answer your final question) there are electronic shutters out there. I've never used one or tested one. I suspect (based on testing electrong shutters on medium format cameras) that they are really accurate. How fast they go I can't say.

-- Kevin Crisp (KRCrisp@aol.com), August 15, 2001.


The laid down standard for all shutters is that they should be within 1/3 stop of their marked speed. This translates to a tolerance of about -20% to +25% on the marked speed.
Leaf shutters present a special problem, however. Because they take a finite time to open and close, the effective speed varies with the aperture set. The shuuter blades open from the centre outwards, so the centre of the lens might be fully revealed, while the outer aperture is still obscured. This usually results in a variable error at the higher speeds, depending on the aperture chosen. This variation is almost certain to take the exposure tolerance outside of the 1/3rd stop limit at the top speed of the shutter, but it may not show on a shutter tester, no matter how 'hi-tech'.

Yes, there were electronically controlled shutters introduced in the late 1960s, but they just didn't seem to take off. Also, controlling a leaf shutter with solenoids doesn't get round those opening and closing time errors, so the electronic shutter still wasn't dead accurate.

I've always wondered why the advances in metal focal plane shutters weren't adapted to leaf shutters too. The use of materials like Titanium would surely be advantageous? If you can get a great clunking thing like a 35mm FP shutter to give 1/10,000 of a second with quartz accuracy, you'd think it'd be a piece of cake to give us a reliable 500th on a #3 shutter.

-- Pete Andrews (p.l.andrews@bham.ac.uk), August 16, 2001.


For 0 and 1 shutters very tight accuracy is possible but expensive.

the Rollei Linear Motor view camera shutter system, available from Rodenstock, Schneider, Rollei, Silvestri, Imacon and others offers speeds to 1/500 on 0 and 1/300 on 1 shutters. Each can be adjusted in 0.1 stop steps for aperture.

A controller is also required to control the shutters.

Another shutter system that is available currently is the Horseman ISS 2 system in 0, 1 and 3 sizes.

these shutters are also available through lens manufacturers.

-- Bob Salomon (bob@hpmarketingcorp.com), August 16, 2001.



The most accurate way to test a shutter speed is by experiment with film. Be sure of all the parameters except the shutter speed. Expose film and by densitometer or other method backtrack to tell what the shutter speed actually was.

This is an excercise that used to be conducted at photography schools to determine a "working ASA" for any given film which would integrate all errors of shutter, glass transmission, developement techniques, etc. Its still a useful experiment to see what's going on.

Shutter speed testing with instruments has always been a problem. Although it would seem simple the problems of "efficiency" (That's the percentage of the time the blades spend opening and closing, yeilding a greater length of time open at the center than at the edge) and just the way light seems to shine around render this a more difficult challenge than one would think. This problem seems to show up mostly at the fastest speeds.

The film density after developement method allows you to test your shutter tester. I have never done this test myself, not being a photographer but remember it as part of the first year curriculum at photo schools.

I have, on experiment (some ten years ago on a modern Kyoritsu tester) taken a dozen brand new Copal #0 shutters and they have all tested very closely to each other at 1/200 second or a full stop slower than marked. It seems to me that this is more a test of the test than a test of the shutters, since Copal is a well established brand made in modern factory and designed by skilled engineers.

Trying to get a shutter to do something that it just won't do leads to a lot of grief.

SKG

-- Steve Grimes (skgrimes@skgrimes.com), August 16, 2001.


The major problem with most shutter testers is that, with lens shutters, they cannot measure the upper speeds accurately, since all they are doing is measuring the total time the shutter is open.This is, unfortunately, is not the same as exposure, at the higher speeds, and will always be slower than the rated speeds. I check all such shutters using an oscilloscope and transmission tester. This gives me a "picture" of the shutter operation from opening to closing. With shutters in good shape, after CLA, the upper speeds will nearly always be very close to the rated values.

-- Henry Paine Camera Repair (hpaine@inreach.com), August 16, 2001.

Richard - As mentioned by Steve Grimes, you're seeing the effects of shutter efficiency at work. There is likely nothing "wrong" with the shutter, assuming the speeds are consistant. The marked speeds are only for the amount of time the shutter spends at full aperture, not counting the amount of time it takes to get there and closed again. Ansel refers to this as a shutter's "dwell time" in The Camera (along with some drawings which make the effect clear).

However, there IS a way to test for this using your electronic shutter tester. Put the camera (lens mounted) on a tripod and find a way to mount the tester behind the ground glass. Set up a light (incandescent lamp) with an electronic dimmer so it shines through the lens in question and falls on the tester. (Do this at night - no other light hits the tester.) Set the aperture of the lens to almost wide open - maybe 1/3 stop down from full. Now adjust the dimmer on the lamp so that the shutter tester just comes on. (Play with the aperture and dimmer settings to make sure the tester doesn't start until the lens is almost wide open.) Now the tester will tell you the "official" speed of the shutter.

As an example, my big old Betax #5 tests at 1/30 at the highest speed (by the normal method) even though it's marked 1/50. However, tested as described above it consistantly yields 1/50. This info by itself isn't very practical (other than a reassurance to someone who thinks his shutter is damaged) because we almost never shoot wide open. However, you can do this same test at various common taking apertures (ex: f/16-f/45) and get an idea of what exposure your lens/shutter is really giving you at each aperture. And as Mr. Grimes mentioned, a complete test of the whole system with film is the best.

Cheers!

-- Mark Parsons (polar@thegrid.net), August 16, 2001.


I would much prefer that a veiw camera shutter be accurate at 1/25 or 1/60 than at the faster speeds. I find that hardly ever use a shutter seed above 1/125 due to the fact that I use filters most of the time and usually rate my film at 1/2 the published speed. Faster shutter speeds are seldom accurate on older 35mm cameras, with most of them maxing out about 1/750. If the shutter is consistant, its speeds should be noted and compensation made if necessary. Consistancy is more important. Incidentally, I saw an article some years back that showed how to test shutter speeds by photographing a piece of tape on an LP record. I can't recall the formula for figuring the speeds, but it can be worked out. Has something to do with how far the tape moves at 33-1/3 rpm.

Regards,

-- Doug Paramore (dougmary@alaweb.com), August 16, 2001.


Directly from a new Copal 0 manual (on a Rodenstock Apo Sironar S): "this shutter has passed rigid inspections for the following standards, 1. Exposure time: Within +/- 30% for all shutter speeds..." This is a standard? :) 60% swing over mean?! A 1/2 stop error, hmm.

-- Paul Coppin (coppin@execulink.com), September 25, 2001.


Moderation questions? read the FAQ