Interesting Tidbit about exposure compensatiion with leaf shutters at small apertures

[sharktooth]

I got an old copy of the "Kodak Professional Photoguide" (1979), from my local camera shop today. They have lots of useful info, especially for large format, and I like to have one handy. It's got a grey card, color patches, and lots of handy calculator dials for things like lens extension exposure factors and depth of field.

I did some reading when I got home, and discovered an interesting tidbit about exposure compensation with leaf shutters at higher shutter speeds and small apertures. This can result with up to 1 stop overexposure even if the shutter speed is accurate. I wasn't aware of this effect, but it would be something to keep in mind for large format, since we're almost always using very small apertures.

There's a table with exposure compensation. Basically, if you're using any aperture of f16 or smaller, and shutter speeds of 1/125th sec, you'll get 1/3 stop overexposure. At 1/250th sec it'll be 2/3 stop overexposure, and at 1/500th sec it'll be 1 stop overexposure.

This is only for leaf shutters, and does not impact focal plane shutters.

The guide doesn't give an explanation, but I think I can explain why this is necessary. When a leaf shutter opens, it starts opening from the center much like the aperture, and then gets larger and larger until it is fully open. Once it's fully open, there is a time delay until it starts the closing cycle. The opening and closing cycles happen very fast, so the time delay between these opening and closing cycles determines the shutter speed. For slow shutter speeds, the opening and closing cycles are much much faster than the delay time, so they have negligible affect on the overall exposure time, but as the shutter speeds get faster, the opening and closing time starts to become significant.

I think the issue revolves around how shutter speed is actually measured. I'm pretty sure it's measured from the midpoint of the opening cycle to the midpoint of the closing cycle. With small apertures, however, all the light is coming through almost right at the start of the opening cycle till right near the end of the closing cycle. This means more light is actually coming in than the shutter speed would imply. It's only important when the shutter speeds are fast enough that the opening and closing speed is a significant proportion of the overall exposure time.

[ic-racer]

As one closes the aperture, the plot of intensity vs time becomes more like a rectangle or a box, rather than a trapezoid. Thus the Te (exposure time) becomes longer and longer, until it nearly matches the To (time shutter is just open to just closed)

[BrianShaw]

Shutter efficiency.

[Kiwi7475]

Why wouldn’t shutters include this in their speed? They could just be tuned to be a bit faster when fully open, given their opening and closing speed. In other words, the real exposure time is the integral of the trapezoid not just the area of a rectangle that ignores the opening and closing triangles.

[sharktooth]

Why wouldn’t shutters include this in their speed? They could just be tuned to be a bit faster when fully open, given their opening and closing speed. In other words, the real exposure time is the integral of the trapezoid not just the area of a rectangle that ignores the opening and closing triangles.

In theory, this could work if both the lens and the shutter were designed to use a well defined fixed aperture. With that scenario you could calculate a "real" shutter speed. That's impractical, however, since lenses use variable apertures, and shutters are designed to work with various lenses with different aperture settings. With leaf shutters at fast shutter speeds, the "real" shutter speed is a function of the aperture, as well as the shutter dynamics. That's what I was trying to point out in the original post.

[Kiwi7475]

Agreed, but who would design it for maximum error at small apertures given that that’s where most of the users will use it? One could do it so that both ends eat some of the error, with a mid aperture between fully open and the minimum having zero error.

[BrianShaw]

One would have to go a long way back in history to find the source of this alleged engineering blunder.

[Tin Can]

You guys crack me up

Look up allowable plus or minus leaf shutter data

[sharktooth]

At the higher shutter speeds, most leaf shutters, especially older ones, are slower than rated. This compounds the problem at small apertures, since even more light will be getting in. The Kodak guide indicates one stop overexposure at 1/500 sec and apertures of f16 or smaller. If the shutter is slower than rated, then the overexposure will be greater than 1 stop. That's getting into serious problem territory with transparency films, although it may turn out to be an advantage for negative films where overexposure isn't such a big deal.

This Guide is from 1979 when Kodak was the bible, and film was king. The old folks knew what they were talking about, so at least be aware of the issue.

[BrianShaw]

Shutter efficiency and shutter accuracy (including allowable speed tolerances) are two different things.

And, yes, their effect can compound