There was some talk on DPReview about how to measure shutter travel time as analogue oscilloscopes become harder to find. Turns out that there are more accurate and precise ways of measuring shutter travel time (and shutter speeds).
That image was created by photographing an out-of-focus LED, driven at 10.000kHz.
More specifically, I set a signal generator to 10kHz, square wave, 12.5% duty cycle, and used it to drive a single leaded warm-white LED via a resistor.
The camera was set to 1/8000 s (minimum exposure time), mechanical shutter.
I used RawDigger to generate a 16-bit RGB TIFF. This was quite noisy, so I applied a 100x5 gaussian blur. I also applied a 30-pixel unsharp mask, with "ammount" 60% and threshold 0. Converted to monochrome, and stretched the range.
The annotations are estimates of positions of the minima at x=3500. The recorded RAW image is 6024x4016.
A list of the estimated y locations of the minima is:
66.5, 243.5, 427.5, 599.5, 782.5, 957.0, 1133.5, 1310.5, 1481.5,
1651.5, 1816.0, 1980.5, 2145.0, 2303.5, 2457.5, 2594.5, 2741.5, 2879.0,
3011.5, 3138.0, 3260.0, 3376.0, 3487.0, 3590.5, 3690.0, 3786.5, 3877.5,
3962.5
The shutter speed increases from about 5.0m/s at the bottom of the frame (top of sensor) to about 10.5m/s at the top of the frame (bottom of sensor). Most of the shutter acceleration is in the first half of the frame: half-way across the sensor, the shutter is travelling at about 9.7m/s.
Using those estimates of the starting and ending shutter movement speeds, and the estimated locations of the first and last minima, I estimate the total shutter travel time as 2.80ms.
Shutter travel speed versus shutter travel, from measurements above:
This implies significant variation in the slit width across the frame, at high shutter speeds.
Shutter speed versus time:
Here's an earlier measurement attempt on the same camera, using an analogue 'scope (Tek 2235):
It seems that my methodology was somewhat flawed. 😞
It appears obligatory to record the lens used in such measurements. I used a Samyang 135mm T2.2 ED UMC. That's like the F2.0 ED UMC, but with clickless aperture, and gears on the aperture and focus rings. Marianne Oelund had some comments in this DPReview thread.