We used a Tektronics 5 mhz data storage oscilloscope and a pick-up device much like a phonograph cartridge in the lab. The pick-up device was taped to the firearm and the o'scope was set to store the trace in the time domain mode. After the gun was fired, you recalled the trace and "read" it on the screen. The first "click" or peak on the screen would be the sear release when the trigger was pulled. Then there would be some static looking peaks that came from internal friction. A second peak would be from the hammer striking the firing pin. In the lab, we just dry fired.
The calibration on the o'scope would allow the trace to be spread out where each centimeter represented a given period of time. The scope was capable of measuring time in units of .0000005 seconds/centimeter so accuracy was excellent. A typical lock time on a S&W revolver was about 38 milliseconds or .038 seconds. A Ruger OM SBH was about 75 ms. The fastest lock time I can remember was from a Remington bolt gun and was just under 10 ms.
That was back in the late 60's so I'm sure technology has advanced. They probably have a dedicated tester that has a digital readout. The current concept is much the same, just more modern test equipment.
In the "old" days, lock time was a measurement of time from when the sear broke until the hammer struck the primer. Now, most tests include the time it takes for the primer to ignite, powder to burn, and the bullet to exit the barrel. This is a more realistic measurement because it is the total time where you have an opportunity to move off target while the gun is "processing".
