Measuring film camera shutter speed with an iPhone

Most film cameras are old, ancient even. My cameras were manufactured around 50 years ago, 25 years before my birth.

My cameras

The shutter mechanism is the part of the camera that does the most mechanical work. At the same time, the correct operation of the shutter is essential for obtaining technically accurate shots. It is important always to check the shutter speeds of every film camera you are buying. It also doesn’t hurt to check your camera shutter speeds every 10-20 rolls of film.

However, checking shutter speeds by eye is not an easy task. You can check that every next shutter speed is roughly twice as long as the previous one, but I doubt that you’ll be able to notice even a 30% difference from the expected value. And 30% difference might have a significant impact on the look and feel of a picture.

In this post, I want to tell you how to reliably (with the precision of 4 ms) measure shutter speeds of up to 1/250 with the iPhone slow-motion video recording feature.

The idea is to film the shutter and then count the number of frames in which the shutter is open. When filming in 240 fps mode, each frame takes 4.17ms. So, the fastest shutter speed we can measure is the 1/250, which takes 1000 / 250 = 4 ms or precisely one frame. At any speed faster than this, the shutter will remain open for less than one frame, which is unmeasurable.

If your phone works with different FPS, you can calculate the frame duration by dividing 1000 by the FPS. If the FPS is 120, one frame takes 1000 / 120 = 8.33ms.

Here is the table of the most common shutter speeds and their duration in milliseconds and frames.

Shutter speed Duration, ms Duration, 240 FPS, frames Duration, 120 FPS, frames
1/1000 1 - -
1/500 2 - -
1/250 4 1 -
1/125 8 2 1
1/60 16.67 4 2
1/30 33.34 8 4
1/15 66.67 16 8
1/8 125 30 15
1/4 250 60 30
1/2 500 120 60
1 1000 240 120

How to

Step 1

Place your camera on a flat surface and prepare an iPhone. It is better to use a remote shutter release for the camera and a tripod for the iPhone, but you absolutely can do that by holding the phone in one hand and using another hand to press the shutter button. That way, you would get a slightly shaky video.

Camera on a table Camera is on the table, iPhone is on the tripod

Step 2

Start recording slow-motion video and go through every shutter speed. You can film every shutter speed several times if you suspect your camera is inconsistent.

Step 3

Export video and analyse it frame by frame. Unfortunately, there is no way to calculate the number of frames on an iPhone reliably, but a standard video player on macOS (QuickTime) can easily handle this task. Use right and left arrows to navigate the video frame by frame.

QuickTime window

You are looking for a frame where the lower shutter curtain is already moving and is at least halfway down. It would be your first frame. Then find a frame where the upper shutter curtain is moving. It would be your last frame.

QuickTime window Example of a good first frame

QuickTime window Example of a good last frame

QuickTime window Example of three consecutive frames. I would count that as 1.5 frames.

After finding the first and the last frames, count the number of frames between them. To convert that number of frames to milliseconds, multiply that number by the duration of one frame. For 240 FPS, by 4.17ms. For 120 FPS, by 8.33ms. So, 64 frames in a 240 FPS video would take 64 * 4.17 = 266.88ms. I use a spreadsheet for calculations.

Spreadsheet window

Now you can compare shutter speed in milliseconds with the expected value in the table above and calculate the relative error.

Example

Let’s say you are checking the 1/60 shutter speed using slow-motion video at 240 FPS. You found the first and the last frames and counted 5.5 frames between them (including them). The shutter was open for 5.5 * (1000 / 240) = 22.9ms. Expected value is 1000 / 60 = 16.67ms. So, the shutter is (22.9 - 16.67) / 16.67 * 100% = 37% slower than it should be.

My results

I checked my Minolta XD and discovered some bizarre behaviour. Shutter speeds from 1/4 onwards are precisely 10% faster than they should be. My best guess is that some resistor or capacitor was replaced with another with different characteristics during a repair. 10% is not a big difference, and I can’t counteract that by adjusting exposure compensation or aperture value, so I’ll probably keep things as they are.

Here is the short video with the results.