Ever tested the shutter speeds on your DSLR?

Eye-level

Banned
Has anyone ever tested the shutter speeds on a dslr like you would on a film camera? It would be easy to do. I wonder if you could ever find a shutter speed off enough to tell?
 

Eye-level

Banned
I had forgot about this. Basically all you do make a series of tests on an evenly lighted wall. For example f/2.5 1000, f/4 500, f/5.6 250, f/8 125, f/11 60, f/16 30 using ISO 200. All of the exposures should be identical. Then you run some series using each shutter speed and go through all the stops. Eg. f/2.5 30, f/2.5 60, f/2.5 125, f/2.5 250, etc. then go to the next stop f/4 30, f/4 60, f/4 125, etc. This series should make for some nice uniform patterns. If a shutter speed is out of whack it is going to show up with these tests.

I do not know if this would be useful for DSLR's but this is one method of testing shutter speeds in old mechanical film cameras.
 

Bob Blaylock

Senior Member
I had forgot about this. Basically all you do make a series of tests on an evenly lighted wall. For example f/2.5 1000, f/4 500, f/5.6 250, f/8 125, f/11 60, f/16 30 using ISO 200. All of the exposures should be identical. Then you run some series using each shutter speed and go through all the stops. Eg. f/2.5 30, f/2.5 60, f/2.5 125, f/2.5 250, etc. then go to the next stop f/4 30, f/4 60, f/4 125, etc. This series should make for some nice uniform patterns. If a shutter speed is out of whack it is going to show up with these tests.

I do not know if this would be useful for DSLR's but this is one method of testing shutter speeds in old mechanical film cameras.

That would test that the shutter is consistent from one speed to the next, that 1/30 is really twice as long as 1/160, that 1/60 is really [close to] twice as long s 1/125, up to 1/2000 being twice as long as 1/4000.

What it would not test is for the shutter to be off by a consistent factor across its range—for example, if “1/30” is really 1/25, and “1/60” is really 1/50, and so on.
 

WayneF

Senior Member
That would test that the shutter is consistent from one speed to the next, that 1/30 is really twice as long as 1/160, that 1/60 is really [close to] twice as long s 1/125, up to 1/2000 being twice as long as 1/4000.

What it would not test is for the shutter to be off by a consistent factor across its range—for example, if “1/30” is really 1/25, and “1/60” is really 1/50, and so on.

You might test that part at 30 seconds, long enough to time it to some percent. I suspect it will be 100.0%, subject to our own testing error. But 30 is actually 32 seconds.

For the f/stops to hold true, our actual shutter speed stops simply have to be:

1/4096, 1/2048, 1/1024, 1/512, 1/256, 1/128, 1/64, 1/32, 1/16, 1/8, 1/4, 1/2, 1, 2, 4, 8, 16, 32

Other than logic, the proof is that 15 actually is 16, and 30 actually is 32. This necessarily HAS TO BE TRUE for the concept of 2x stops to work.

Each step is exactly double the previous, which is the whole idea. The marked numbers are just rounded nominal markings, to be easier for us humans, marked that way for maybe 100 years, when it was much harder back then (when mechanical shutters could not do longer than 1 second, and error was also terrible at 1/1000 second). But todays camera knows to do it correctly.

Third stop shutter speeds are intervals of cube root of 2, or 1.12992x multiples from the full stops (aperture thirds are 1.12246 steps, the sixth root of 2).
More at Understanding Photographic Aperture f-stop, Shutter Speed, ISO


An old mechanical shutter was more issue, but our DSLR focal plane shutters (today) are quartz crystal controlled, like a watch, electronically timing the delay between the start times of the two curtains. Both curtains are driven by the same motor and same gears, and all shutter speeds move the two curtains at the same one speed, just with a timed delay before starting the second curtain... easy for a digital clock.

As the first curtain is opening, the second curtain is closely following, shutting. The width of this open slit is the shutter duration for any pixel that it passes, same for all pixels. However, the moving slit does expose the early pixels a couple of milliseconds before the slit reaches the later pixels.

The mechanical motion of the curtains is one single speed to be designed (also driven by electronic pulses), but if it was off a little, it simply doesn't matter.... since both curtains are necessarily the same speed, and since the shutter duration is instead the timing between the two curtain starts, timed electronically. Simple elegant beauty. Mechanical shutters ain't that way. :)

Simple cameras today (compacts and phones) use a CCD sensor, and their shutter is the timing of the duration that the CCD sensor is enabled, also electronic. Has downsides, but is free, cheaper than a complex mechanical system. A few inexpensive DSLR (D70, D40, D50) also did it this way, but the current D3x00 line uses a real focal plane shutter.

So really, any test of equal wall brightness at each stop is more likely testing the compensating aperture, more so than the shutter speed. The aperture is mechanical.
 
Last edited:
Top