How Accurate is Your SLR's Shutter?

[doubs43]

How accurate is your SLR's shutter? I've evaluated a number of my own cameras with interesting results.

I've just spent a week traveling to Maryland for a high school class reunion and picture-taking. I took along two Ricoh Singlex II camera bodies and lenses from 20mm through 200mm. Before taking the trip, I tested both camera bodies to see how their metal Copal-Square shutters were functioning. All speeds were recorded three times and the results averaged. What I found was interesting.

From 1 second through 1/60 the shutters are essentially spot-on. X-sync is at the marked 1/125 speed but one tested at 1/100 and the other at 1/93. (1/100 is accepted as normal for Copal-Square X-sync from what I've read.) One was accurate at 1/250 while the other was at 1/225; still OK.

It was the 1/500 and 1/1000 speeds that were the surprises. Both were off by quite a bit with readings of 1/334 and 1/324 at the 1/500 setting and 1/505 and 1/538 at the 1/1000 position. That's more than enough to cause problems with over exposure. Knowing that is important for accurate exposures.

An old Pentax Spotmatic II with cloth shutter curtains turned in very similar readings.

Compare those readings to an Exakta RTL that I recently tested. It was a real surprise, turning in excellent readings across the entire range. They were: .8, .4, 1/5, 1/10, 1/20, 1/37, 1/74, 1/135, 1/255, 1/505 & 1/1149. Shutter settings are the standard 1, 1/2, 1/4, 1/8 etc. While ever so slightly fast, the shutter readings were very consistent for the three samples taken at each speed.

Another surprise is an original non-MD model Olypus OM-1 that is very close to all marked speeds across the board.

If you happen to have a shutter speed tester or access to one, check your speeds. It helps to know what your camera is actually doing.

Walker

[Randy]

Gee...how come you didn't test a Minolta? I'm curious to see how one with an electronic shutter would compare.

[doubs43]

Ron, most film - color or B&W - will have enough latitude that except for the most critical purposes won't show as much error as one might think. Slide film is the exception. It's pretty unforgiving. Even if film has the latitude to deal with errors in exposure, I'd prefer to be as close to the correct exposure as possible. The two Ricohs are not accurate at 1/500 and 1/1000 but knowing that helps me choose the proper f-stop.

Randy, I already have too darn many mount systems which is why I've avoided Canons and Minolta among others. However, I did buy a Minolta SRT-102 in a weak moment. It's the only one I own and when I have the time I'll try to check the shutter speeds...... just for you!

Walker

[Randy]

Danka Walker!

[pancake]

OM-1 rules!!! yeah!

[PeterW]

In my view the consistency of a shutter's performance is more important than whether or not the actual speeds agree with the marked speeds or not - provided you know within reason what the deviation is.

In the days before electronically controlled shutters, auto exposure, or even TTL metering, and before magazines like Amateur Photographer published actual vs marked shutter speeds in their camera tests, no-one expected to get to know a camera until at least a couple of rolls of film had been shot.

Compur, Compur Rapid and Prontor shutters were well known for over exposing at their two top marked speeds, sometimes by as much as 30% or 40%, even when new. As they got older the bottom speeds, 1/10 and below, also slowed down until the point where the shutter needed a CLA to be anything like accurate outside its middle range of speeds. Many older leaf-shutter cameras you find today are well overdue for a clean. Even on focal plane shutters the marked speeds are often suspect now the cameras are older.

Photographers always used to accept this, and made allowance for it.

Of course, making allowance assumes that you measure, or estimate, the exposure correctly in the first place. All built-in uncoupled meters and hand meters in reflcted light mode, selenium or CdS, give an average reading for the scene at which they're pointed. Similarly, most older TTL systems give an average reading over the whole frame. Even the Spotmatic didn't give a spot reading, it gave a centre-weighted general reading, which helped to avoid underexposure in a scene with quite a lot of sky in it.

That's fine in a scene which is fairly evenly lit, but often disappointing in a scene where there is an amount of shadow area. The shadows often show very little detail.

This is why, whenever possible, I like to take two or three readings - highlights, mid tones and shadows - and then try to estimate whether or not I need to adjust the indicated exposure. Even then I quite often get it wrong, and bless the exposure histogram in PS!

Adjusting the indicated exposure could be quite important in the days when I shot a lot of colour transparencies. Even Fujichrome was quite unforgiving, so I always bracketed exposures. Nowadays I use mostly colour print, with a lesser amount of black and white, both of which are much more forgiving than transparencies.

But even PS can't put non-existing detail into shadow areas, at least not with film scans - or I'm not skilled enough to make it do so. My limited experience of digital has indicated that it is excellent at recording shadow detail, even if the shadow is so dark that the detail is all but lost. It's still there faithfully recorded on the card, and with a spot of selective adjustment in PS you can bring that detail back surprisingly well.

I sometimes wonder how we managed without it. I suppose we took a lot more trouble to get the first exposure right, made test strips and then tweaked things a bit with different grades of paper and hand dodging under the enlarger.

A well-known dodge to bring out burned-out highlight detail was to use a warmer developer - and I mean physically warmer, up to 26 or 28 degrees C, and let the print stew for five minutes. Other people used to take the print out of the developing tray and use warm developer on a ball of cotton wool to bring out highlights.

All this was with black and white, of course. With colour prints you made a test strip, estimated the filtering and then worked completely in the dark - and used a lot more paper getting the final print right.

It all seems a lot of trouble and a long time ago.

PeterW

[doubs43]

Peter, as usual you are quite correct in that being consistent is the most important aspect of shutter function.

I believe the old axiom was "Expose for the shadows and develope for the highlights". The third method of bringing up an area on a B&W print was developer applied with a finger.

Randy, the Minolta shutter tested out as follows: (Three readings at each speed averaged)

1 Sec: 1.3 Sec.
1/2 Sec: 2/3
1/4 1/3
1/8 1/5
1/15 1/12
1/30 1/21
1/60 1/49
1/125 1/84
1/250 1/172
1/500 1/306
1/1000 1/549

For the most part the readings were pretty consistent with the higher speeds the most consistent. Still, I think that this camera is in need of a CLA. IMO good images could be taken with it now by adjusting for the known speeds.

pancake, the OM-1 shutter speeds are quite good but so are those for the Exakta RTL-1000 I tested.

Walker

[Randy]

I've been told the Minolta cameras with electronic shutters are more accurate than the older models with mechanical shutters, but in my experience the older models with mechanical shutters are more reliable. One thing that irritates me when I'm shooting is to compose a shot and have the camera not take the shot because it doesn't like the settings. I shoot Trains on the move often, and if I don't get the shot right then and there the opportunity is lost.

[pancake]

doubs43 said:

pancake, the OM-1 shutter speeds are quite good but so are those for the Exakta RTL-1000 I tested.

Walker

Yeah, but I'm an Olympus OM fan ;D

Although lately I've been mightily impressed with the Contax/Yashica SLRs.