Accuracy of shutter speeds, test results...

I just peformed a ss test on 7 LF lenses. There has been several posts asking a bout the Calumet ss tester. For the money, $80, I was very impressed by this li ttle gadget. More importantly I was very impressed with the results. I tested 7 new LF lenses, ranging from 75 to 600mm, and between 1/2 second and 1/500, the furthest reading off was about 5%, or less than 1/10 of a stop. And that only occured a few times. All in all, things tested almost perfect... In this field, I find that hard to beleive, could have I screwed the tests up? I checked my m ath several times... anybody ever try this?

Hi Bill;

I think that a lot of the concerns people have about shutter accuracy stem from the many older, used shutters on the market. I've tested my new Copal shutters fairly extensively (using equipment lying around here at work rather than the Calumet or a similar tool) and they're all pretty close. They aren't to within 1/10 of a stop at the faster s[eed, but they're all _well_ within 1/3...

-- Patrick

One additional comment: My favorite testing method is to put a light source on one side of the lens and a photocell on the other, and hook an oscilloscope to the photocell. Focus an image of the light source onto the photcell, and start shooting. The nice things about this approach are that:

1. It's very easy to look at things like the variation in effective shutter speed as a function of aperture (simply vary the aperture and look at the changes in the waveform)

2. It only costs a few dollars (assuming that you can borrow an old 'scope from somewhere - You certainly don't need a fancy one for looking at events as slow as shutters opening...)

-- Patrick

Bill, I have been using the same device each year to validate my shutters prior to starting the summer season. I do not have my numbers with me at work, but I have found very little variation from year to year. I also noticed that long shutter speeds were more accurate than shorter ones. Sometimes I would get a 30% error, but most were around 5% (I think). I do not put much value in this test because I cannot remember when I shot a scene shorter than 1 sec.

I have just purchased a pronter timer that connects to the shutter using Bulb. It ranges from 2-32 seconds. B&H sells them for $650. I got mine from ebay for $62 used. I am real excited about this because trying to watch the flower movements in the wind and your rist watch at the same time can be a hit or miss experience. This will free my eye up to focus exclusively on the vegetation and will provide me with more consistant exposures.

Bill, your results are pretty amazing. I've had the Calumet shutter tester for about a decade and, during that period, have tested fifteen brand new lenses, all in Copal shutters. These were mostly 0 size; a few were size 1. Up to 1/125, they are usually within 1/4 stop of accurate, but never less than 1/2 stop slow at 1/250 and always at least 2/3 stop slow at 1/500. I tape a small table on the lensboard with compensating aperture corrections for each speed that is more than 1/6 stop off. In recent years this table has been printed using 8-point type in Arial Narrow font.

A comment to Sal:

Be VERY CAREFUL about how you measure timing accuracy at the faster shutter speeds. The reason is because if you measure at small apertures (i.e. looking only at light rays which pass through the center of the shutter, then the shutter speed will appear to be slow as you describe. If you measure at wider apertures (i.e. including rays from the edges as well as the center) then the speed will appear to be accurate or perhaps a bit fast.

Regards,

Patrick

...unless, of course, you shoot almost exclusively at very small apertures.

All my testing has been at f/22. Almost all my shooting is done at f/22 and smaller. Shutter efficiency is much higher at this aperture, and it makes no sense to test wide open unless one uses that aperture for real pictures. Hardly anyone working in large format does. Also, note that it is important to precisely follow directions on the Calumet tester regarding illuminant type and distance. Not doing so can skew results. The ultimate test of my calibrations has been field use, where, even with TMY in X-TOL (EI 500) under full sunlight, exposures are exact.

The Copal specification for shutter accuracy is stated as follows in their documentation (for the Copal Press shutter):

"This shutter has passed rigid inspections for the following standards.

1. Exposure time: Within +-30% for all shutter speeds. ..."

I don't know if their spec is different for the non-press shutters.

I second the note of caution when testing fast shutter speeds. If memory serves me, you have to start paying attention to this factor at about 1/125 and faster.

Even a perfect leaf shutter will be effectively slower at its higher settings and small apertures, as much as a full stop, since the blades spend virtually all of their time in transition at the highest setting. Luckily, we rarely use the combination of highest speed and small aperture because we don't shoot on bright sunny days with fast film, right? :) Don't forget the effect of temperature on shutter calibration. More than once I've had old shutters with thickened lubrication that work fine at cozy room temperature but slow down substantially out in the winter wonderland. I've got the dense negs to show for it.