The ultimate shutter tester is a oscilloscope. It can measure flash sync timing and shutter efficiency. It can detect problems like shutter bounce. Learning to use an oscilloscope is certainly easier than learning to use most cameras well.
That's why I didn't build it. I already have an oscilloscope.
"People who say things won't work are a dime a dozen. People who figure out how to make things work are worth a fortune" - Dave Rat.
There is nothing special about mine. It is a grid of super bright LEDs with load resistors to make sure they don't pop. A simple detector diode (I forget the exact part number) and a power supply and Oscope. The diode is reverse biased because it operates better that way, though you may be able to use it in photovoltaic mode (I never tried it). Keep an eye on the proper voltages from your detector, that is important for the speed of the device. The one I used is something like 10 volts DC for optimum speed. That makes it a little hard on the scope that I have since I can't offset that much DC, so I run it in AC mode which is why the picture looks like it does. This is also what you will see on a sound card, but with the sound card you can not use 10 volts!
The data sheet for the detector showed exactly how I needed to connect the power for this to work, there are VERY few components in this and almost any should be able to build it. If you want to get really tricky, use IR LEDs for this detector to get the largest voltage change. The trick is determining if the LEDs are actually turned on, which is why I went with the white LEDs. Cost me about $25 in parts, and I had to buy more than I needed to beat the service charge.
I followed Denis' link, constructed a tester and have tested every shutter I own, MF and LF... all in the past 10 days. As a result I have sent 1 shutter to Carol Miller, with another ready to go when the first returns. I have also decided that my MF TLR lenses are all fine, within 1/3 stop. Thanks to Denis for the advice and link!
The most analog (and traditional DIY) way was to photograph a rotating phonograph turntable with a marker on it.
Great for between-the-lens shutters, but unpredictable for slow moving focal plane shutters. This method probably upset a few Speed Graphic users.
Another way to check moderate speeds is to photograph a neon or mercury vapor lamp while moving the camera to spread the image out during exposure. These lamps flash 120 times a second. For focal plane shutters, move the camera at a right angle to shutter movement.
Viewing a TV screen through the shutter is a quick way of testing speeds from about 1/30 second to high speeds.