Long post for anyone interested in geeking out for a bit. If you think high shutter speed are irrelevant because we should always stop down to maximize DOF and only use slow shutter speeds, this may not be of interest to you. No need to comment, just enjoy the scenery I would actually like to use the faster speeds, and as I've been checking out the used lenses I've been buying, the phenomenon of the fast speeds measuring much slower than their labelled speed piqued my curiosity.

I've been doing a lot of searching to understand this, but my tests don't quite match up with what I've been reading.

Let's start with the basic shutter speed tester: phototransistor to detect the opening and closing of the shutter and measuring the elapsed time. Works great at slow shutter speeds where the time it takes for the shutters to transition is insignificant compared to the time the shutter is open, so the shutter is effectively fully open for the entire shutter duration. But at faster speeds (say about 1/125 and faster), the shutter transition time becomes significant.

At the fastest speed, using the basic shutter tester, people will typically measure a speed roughly half of the labelled speed. The claim is that this is because the tester measures from when the shutter first starts opening to when it's completely closed, but since it takes time for the shutter blades to transition from fully open to fully closed, they start closing again as soon as they're fully open. So if you integrate the light intensity over the entire time, it's actually 1/2 what it would be if the shutter transition where instantaneous. Therefore, if the shutter takes 1/200 from start to finish, the effective exposure is equivalent to 1/400.

Soooooo...

I made the basic shutter tester with an Arduino Uno and phototransistor. And sure enough, at the fastest speed (1/400 on all the lenses I've tested), the measured speed is roughly 1/200. So I decided to hook the phototransistor up to the ADC to try and plot the opening and closing of the shutter and see if it matched the claim. At the 1/400 setting I get this:

Click image for larger version. 

Name:	shutter.PNG 
Views:	31 
Size:	10.4 KB 
ID:	206701


I got similar results with several different lenses / shutters - LF with Copal and RB67 lenses. Seems that it really is slow, but it's so common, why would the manufactures label them as being faster? At least it's only about a stop over exposed which isn't too bad with negative film. I'm still curious what's actually going on here. I'd really like to see a slow motion video of the shutter. There are plenty of focal plane shutter videos, but I haven't found a good leaf shutter video.