Are the higher shutter speeds on large format lenses always off?

[Alan Klein]

The whole point of using lenses wide-open these days is for the effect, not because we're sports or news photographers tying to stop action with fast shutter speeds. Not that lf can't do that, just that I'm pretty sure most sports and news photographers are now using digital, not their 1940s press cameras.

So you spend mega bucks buying, say, an f3.5 Xenar or Planar or whatever in shutter, you head out to your favorite stream or meadow or rock pile, or bribe your kid to sit for an outdoor portrait, and proceed to shoot the lens wide open, at top shutter speed. Your lens is no longer effectively wide open. What effect does this have on image, given that your effective aperture is 1) moving, 2) not circular?

I propose the following two tests, that require a few sheets of film, a fast lens in shutter, and an ND filter: shoot the same scene wide-open at 1/15s and 1/400 or whatever the shutter's top speed is. Shoot the proverbial ruler used to check focus, but in this instance for measuring depth of field, and shoot something with an OOF background with pinhole light sources. I expect the DoF will be increased using the shutter at max speed, negating the money you put into that fancy lens, and I expect OOF highlights will be very undefined but in some scenarios might be starfish-shaped?

The test could also be done with a Speed Graphic, using the same shutter speed on the leaf shutter and then with the lens open, using the roller-blind shutter. That might remove 1 variable for the OOF highlights (reflections off the ND filter). I'm not doing this myself for the very good reason that I don't own any fast lenses in shutter, unless you want to count a Verito in a non-working Studio shutter. The same effects will be present in slower lenses, but the demonstration will not be as interesting.

Wider lenses give more light through the viewfinder or ground glass to see the subject, even if the camera stops down when your shoot the shot. Also, with slow film, at least on 35mm, you're able to capture shots that would be blurry otherwise due to slower shutter speeds. Of course today, with digital, those big open heavy lenses are not necessary due to higher ISO settings and display views that compensate for low light electronically. The DOF aadvantage is overplayed unless you want 1" of DOF on a portrait . Then everything is blurry eside the person's eyeball.

[Alan Klein]

Jody, I've never noticed vignetting from my lenses in shutter. And I sometimes use high speeds. What am I doing wrong?

How can you get vignetting? That means at smaller f stops, you'll also get vignetting which doesn't happen. The lens is designed to avoid that.

[Pere Casals]

Not with chromes, Pere... On a large even white background, 2/10th of a stop stands out...

Steve K

Of course, with chromes we need all precision we can get. But meters themselves have a 1/6 stop variability, depending on brand, beyond spectral sensitivity differences. I found that with slides best is learning from bracketings while spot metering different subjects, many meters underrate blue so sky tends to be more overexposed than spot meter says.

[Pere Casals]

Pere: How does it work? Is it using light through the shutter or sound? Or something else?

Hello Alan

Yes, it works. Those sensors do include a photocell (photoresistor, photodiode or phototransistor), then don't take sound but light.

There is some confusion because those sensors are plugged in the audio input of a PC or of an smartphone. Usually audio inputs have a capacitor in series to remove DC so when you record the photocell signal with Audacity audio recorder-editor (freeware) you may see a weird signal, still you see the opening/closing flanges and you can calculate speed, the exposure interval is seen in the Audacity audio editor, on in the smartphone App.


Me, I use a sensor and a USB oscilloscope (Picoscope brand) that shows a clean signal, knowing if the it is the opening or the closing ramp, or both, gives a clue of what it has to be fixed. Last fix I made was removing oil that reached the blades of a compur, blades were looking clean, but the erratic oily behaviour in the ramps made me suspect that problem, so I let fall some ether drops on the blades and after firing reveral times I removed ether with a cloth and extreme care: the cloth get black !!! after repeating that several times the shutter performed like new since then. Probably at Flutot's they see that at first sight, but viewing the ramps was useful to me, I saved openning tha shuter and probably destroying it


If you don't like electronics then best is purchasing the standalone version ($84) that delivers the readings in its own screen. IMO those readings are pretty accurate.

Ciba didn’t just have “idiosyncrasies” it could suffocate you!

I Was an Agnecolor rep way back when and had the 16x20 processor. The first part of the processing was a piece of cake. But once you opened the cover- WOW!!

Ciba is still alive, at least one man (of a kind) still stands in the trench serving a 16-pounder, we all know who is.

[Alan Klein]

I note the spread of several readings per speed, as well as the general deviation from nominal rated speed. I've been quite fortunate in that this too has been quite predictable with all my lenses, so small in fact that I don't have to worry about any of them except at the highest speeds. When I did mostly chrome work, I'd factor in anything 1/3 stop, but that was the worst any setting got. So now, when shooting color neg and b&w neg, I really don't worry at all. But over the long run, it is indeed important to periodically "exercise" lenses that might otherwise be completely neglected. Someone a couple weeks ago wanted me to look at a lens in Copal 3S from an estate sale that probably hadn't been handled in decades. Even though it was in clean condition, the low speeds were completely frozen, and most of the hypothetical profit selling it will disappear into a professional tuneup. It certainly wasn't a cult lens, but someone like me would have bought it just for the shutter itself if it didn't have a problem.

As you know Drew, I just started shooting 4x5. My first lens I bought from Japan was a Schneider 150mm Symmar. So I checked it with my phone recording the sound and then viewing it with a program that I could check the speed. It worked up to about 1/125 generally. But the signal becomes very chopped at faster speeds. So it's hard to check it.

So the first time I checked, most of the speeds were fast by 1/3 of a stop up to 1/30. Then I dropped the lens from about 6 feet high right onto my garage's concrete floor. Smack!. I thought it was all over. So I checked the lens and it looked OK. Then re-checked the shutter. It got more accurate! See the summary below. The numbers on the right were from the first test before I dropped it.

The second photo is a the wave chart using Audacity program that examines the waveform for a 1/8 second. The time .101 seconds is circled.

As an aside, I check two other lenses for large format and they were more accurate. I also check a Nikon N6006 35mm film camera from 30 years ago. It was on the money. Exact. Must be an electronic shutter.

[Alan Klein]

Here's a test for a 90mm Nikkor f 4.5 large format. Shutter is very accurate.

[Alan Klein]

Last one. Fujinon 75mm f 5.6 large format 4x5
Some speeds were right on. Others maybe 1/6 stop too fast like the 150mmm Schneider. But overall pretty accurate.

[Joseph Kashi]

FWIW - I just reviewed the late 2019 speed test data that I have for the 30 lenses in shutter that I have in various 4x5, 5x7, and 11x14 kits. I did these tests so that I could put together a set of laminated field-use index cards with true speeds for each lens.

The tested lenses include multiple examples of Compur 0 an d 1, older and also more recent Copal 0,1, and 3, Ilex Acme 3 and 4, Rapax 0 and 1, Seiko O, multliple sizes of Compounds, and single examples of Prontor 1, Compur 2, Kodak Supermatic 1, and Volute on the 12" Protar VIIa set.

About 85% +/- seem to be factory OEM mounts. These shutters mount a wide variety of modern Rodenstock, Nikko, Schneider, and Fujinon optics and a total of 8 earlier 30s Protars VIIa and 50s-60s US-made Dagors and Ektars. Probably a pretty decent range of ages, optics and shutters overall.

The only firm conclusion is that in 90% of these shutters, including those with very recent Grimes and Flutot CLA, the fastest speed is almost always 1/2 to 1.5 stops slower than marked, with the next fastest speed usually 1/2 to 1 stop slower than marked. On smaller size 0 and 1 shutters, speeds up to 1/125 are usually reasonably close to marked speeds, though typically off somewhat.

Luckily, at f22-32 with ISO 100 film, those accurate slower shutter speeds are the ones that we will most likely use with everything but a hand-held Speed Graphic, etc.,, even in bright sunshine. So as long as the true speed is known and used in determining the aperture to use, it's all OK.

How tested: I have tested all of these shutters at all speeds multiple times with my Calumet shutter speed tester under repeatable conditions with the same light source to be sure that I was not recording outlier results. Most shutters at slower speeds were repeatable within a tight band.

At least a third of these lenses have had recent CLA by Carol Flutot and Grimes, who both also provided a chart of tested speeds for each CLA shutter. The fastest two marked speeds for lenses that have a recent CLA were not reliably closer to marked than non-CLA lenses. No brand was reliably closer to marked speeds at the fastest shutter speeds. Age did not make a discernible difference if the shutter was reasonably clean. Older Compounds mounting 145-183mm Protars were at least as accurate as newer Copals.

The Flutot-Grimes post-CLA test results and the results of my shutter speed tests made after CLA agreed closely, indicating that my own Calumet shutter speed tester was providing reliable results for the lenses that were not recent CLA.

Conclusion: About 90% of those 30 tested shutters, of every brand, age, and size, were 1/2 to 1.5 stops slower than marked at the two highest marked speeds.

Testing every shutter and the using the actual tested speeds seems necessary where exposure is critical, as in chromes.

This isn't any different than the practical use film-based Zone System tests that were the norm mid-20th Century, except that we can now do this electronically without wasting a lot of film.

[Pere Casals]

Excellent report