Testing shutter speeds without a shutter tester.

While working on a Podcast devoted to large format photography, I decided to record the sound of one of my shutters firing to use as a sound effect in the intro. When I did that, I immediately realized that I had a graph that showed me how accurate my shutter speeds were. I used the free audio program, Audacity, which only took me about 15 minutes to figure out how to use.... and I'm slow! I then tested all of my large format shutters and the results were very good up to speeds of about 1/60th of a second which is fine because I rarely, if ever, use speeds faster than that anyway. Fortunately, I had just received a lens back from S.K. Grimes that was in for a CLA and the results that I got with Audacity matched perfectly the data sheet that Grimes included with the lens, confirming the value of this method. Someone then asked me if it would work with focal plane shutters so I gave it a try. I did not expect it to work but I was very suprised to find out that it worked extremely well for my 35mm shutters too but, again, mainly for the slower speeds. Attached is an image of a test. I apologize for the double graphs for each speed. It was recording in stereo so there are two channels. I will have to find out how to shut that off next time. Of course, the graphs can be scrolled up to line them up with the timeline at the top and also scrolled left-to-right to align it that way (as I have done in this example.) The graphs can also be expanded or contracted as much as you like to study them in greater detail. You almost certainly could do this with any audio editing program.

For me, this is a real boon because I don't have a shutter tester and I am never really sure if my shutter speeds have drifted. For that reason, I generally send them in for a CLA every year or two even though they may not need it so this saves money and allows me to be confident that I know the actual shutter speeds of my lenses. Now I can test them in just a minute or two! I hope someone here finds this method as useful as I have. I have no doubt that other people have stumbled upon this or a similar method but I hadn't so, hopefully, there are others who may find this useful.

You can download Audacity for free at http://audacity.sourceforge.net/

The attached graph should be self-explanatory. This is the results from a test of a large format lens for shutter speeds of 1 second, 1/2 second, and 1/4 second. Of course, you can test all shutter speeds and simply scroll up and down to them with the program.

Do you know how much of the duration of the audible event represents the amount of time when the shutter is open? You would have to. Much of that sound event occurs before and/or after the time when the shutter is open.

For speeds faster than 1/60, you can also do a photo of a TV screen. The screen is interlaced, and will paint the entire screen in 1/60 sec. If half of the screen shows in the image, the speed is 1/120, 1/4 screen is 1/240 etc.

After having painted the screen in 1/60, the TV will spend the next 60th of a second painting the lines in between the first set of lines that were painted (odd vs even numbered lines), close examination of this will allow you to measure down to 1/30 sec.

Note that these numbers are assuming that your TV is a CRT based TV, in North America, standard definition - those of you with HD plasma sets can probably afford a shutter speed tester .

As for the duration of the audible event.... I was concerned about that too. Of course, I had to know when the shutter actually opened and closed as you correctly point out. However, as I mentioned, I had a shutter that I had just that day received back from S.K. Grimes so I could compare the graph to their results to determine where the actual opening and closing of the shutter occured. To my suprise, the actual opening and closing was easily identified in the graphs for my LF shutters and my 35mm focal plane shutter. Take a look at the graph posted and you will see what I mean. The event (shutter opening) begins at the sharp spike and ends at the other sharp spike. I also compared the graph of another lens to the test sheet I got back from S.K. Grimes several months ago and they were all right on the button except the 1 second speed which is now about 1/3rd stop off. I simply made a note of that and now compensate when using that speed now.

The 35mm focal plane shutter that I tested had also just been in for a CLA. However, International Camera did not give me a data sheet when they returned the camera but the graphs were perfectly clear with sharp spikes aligning virtually perfectly with the marked shutter speeds.

Of course, different shutters function differently and the graph may be different for different types so a little careful examination is all that is needed when you try this with a different shutter. For example, a self-cocking No. 5 Ilex Universal Syncro shutter that I tested does record sounds both before and after the shutter actually fires but the points where the shutter snaps open and closed are perfectly sharp spikes in the graph and they also match the S.K. Grimes tests that were made by them the last time the lens received a CLA. The bottom line is that once you know how to identify the actual opening and closing on the graph, you are all set. All of my Copal shutter graphs look virtually the same and it is extremely easy to see when the shutter actually opens and closes. Residual noises are easily identifiable. In every lens and shutter that I tested, the event was perfectly clear. But, ideally, you should first try this with a lens that you have recent test data for.

Mark, I am familiar with the method you desribe and, in fact, I did it years ago. Somewhere I even have my results filed. However, I found it more of a curiosity since it involved some effort and the counting of lines on film. That method is often used in college photography courses as a learning project. With the method I describe here, you don't have to expose film at all and you have results in a matter of seconds. Of course, you could just buy a shutter tester but my intent in posting this is to give those who may be interested a quick method of testing their slower shutter speeds without needing any equipment at all and without any expense. I am getting virtually perfect results and I'm very glad that I stumbled upon it. I did post this idea in another forum and someone else had plans for using this method with a simple light sensing device that cost less than $3 to build. (If anyone is interested in that, email me. I think I have the plans saved in a file.) The device apparently sends a signal to the sound program when a light passes through the shutter when it is open. I haven't tried it yet myself but I may. By combining the methods, it may be possible to very accurately measure even the highest shutter speeds. The device was made up of just a box to put it in, a switch, a resistor, and a light sensing diode, if I remember correctly.

If you are concerned about identifying the event (shutter opening and closing) I suggest simply taking a minute and giving this a try and see how it works for you. It's so easy!

Test with focal plane shutter

Attached is a test done with a Canon F1 that I had only exposed about four rolls of film with since it had a CLA. I have a motordrive mounted on it so I did the test with the motordrive turned off. There is more residual noise with it's focal plane shutter but the graph is still easy to interpret especially since I knew the shutter was accurate when I did the test. This test is for 1 second, 1/2 second, 1/4 second, and 1/8th second as marked on the left of the graph. I tested faster speeds but, of course, they won't fit on the screenshot. I have scrolled the first sharp spike to the zero point. Remember, the graph can be expanded as much as you like to see the shorter durations in greater detail.

I want to stress that I am not suggesting that this method is a substitute for tests done with shutter speed testers. What I am suggesting is that it is a very useful, accurate, and easy method to check your slower shutter speeds making it especially useful for testing LF shutters. If you have a PC or MAC (and you do if you are reading this), you can do it without incurring a single penny in cost. And it's so fast and easy! Other photographers who I have told about this have reported that they are getting beautiful results.

I've used this method as well, and it works, as long as you have an idea of how the shutter operates and what sounds it's producing. With some leaf shutters, the slow speed retard mechanism can be loud enough to make the sonogram hard to read. In my Bronica S2a, the shutter is pretty complicated (mirror drops rather than rises as it is covered by a metal sheath, cloth shutter covers the focusing screen to prevent stray light from entering through the eyepiece, then the focal plane shutter opens and closes, mirror returns and focusing screen blind opens), but you can pick the focal plane shutter sound out of the graph fairly easily.

Somewhere on the net, there are instructions for building a shutter tester by plugging a circuit with a photocell into the microphone jack of a computer and reading the graph produced in a sound recording program, measuring the light directly, rather than the sound of the shutter.

Your method as well as the one using the photocell instead of a microphone work well. However, as you mentioned they are only accurate for the slower speed and they are fine for checking large format shutter and older 35mm cameras. I found that modern electronically controlled shutter 35mm SLR's are very accurate at their slower speeds because at these slow speeds the mechanical components contribute very little to their accuracy. Their speeds are off only for the higher speed i.e. 1/500 and up.

Audacity has the ability to stretch out the time scale so that you can measure to milliseconds. This zooming capability is under the View menu.

Lee