When the D-500L/600L models were introduced in 1998, they were the only consumer-grade digital SLR available. We decided to try to use an Olympus D-600L to photograph balloons being shot by BBs from an air rifle. Some features that made the camera attractive for the experiment at that time were a 1.4 megapixel CCD for high-resolution images, manual exposure, and manual focusing controls. Like other digital cameras in the sub-$1500 category, it had no shutter speed control. This meant that one wouldn't be able to manually hold open the shutter and then shoot the balloon. If the experiment were to work, therefore, the opening of the camera's shutter would have to be synchronized with the entry of the BB into the balloon. 

Matthew Moore, a senior at the North Carolina School of Science and Mathematics (Durham), carried out the experiment in the fall semester of 1999. Moore solved the synchronization problem using a modification of an apparatus that had been used by Matthew Hinshaw in 1990 for high-speed sweep photography.  (Hinshaw's apparatus is described here.)  He also used a triggering technique described in an article by Davidhazy.  Moore encountered a number of obstacles that he had to surmount. These included finding (and rebuilding) an air rifle in order to achieve a firing sequence repeatable to within a thousandth of a second, adapting Hinshaw's timing equipment to the experiment at hand, learning how to trick the D-600L to open its shutter in the dark (something the camera generally refuses to do), and timing the interval between the discharge of the camera's red-eye reduction flash and the opening of the shutter.

The red-eye reduction flash was the key to initiating the timing sequence, as described in Davidhazy's article. The sequence for Moore's experiment was the following:  a) the camera's shutter button was depressed to set off the red-eye flash, b) a photocell detected the flash of light, c) a computer detected the voltage transient from the photocell, d) the computer generated a predetermined delay interval and then activated a solenoid, e) the solenoid pulled the air gun's trigger, and f) the BB reached the balloon as the camera shutter opened and the flash discharged. A difference between this experiment and Davidhazy's was that Moore used the camera's built-in flash unit to illuminate the subject. Davidhazy used an external microflash unit in order to provide the microsecond exposure necessary to stop the action of the supersonic bullets he was using.)

In order for the experiment to be successful, the interval between the discharge of the red-eye flash and the opening of the camera shutter had to be greater than or equal to the time taken for the solenoid to be activated and the BB to reach its target. Moore measured both time intervals in advance in order to ensure that this condition was met.  He verified that the time from activation of the solenoid to balloon burst was repeatable to within a millisecond. This reproducibility was important, since the balloons would take only a few milliseconds to burst. With results of the timing experiments in hand, Moore was able to set the time delay generated by the computer necessary to achieve synchronization.

Moore's experimental setup is shown to the right. We present this photograph not only to show the arrangement of the components but also to show that appearance is irrelevant to a successful experiment. Note that the gun, which is mounted securely in a wooden framework, fires downward toward the balloon. The BBs are captured in a trap on the floor. A small tripod on which the camera would be mounted is taped to the floor in front of the balloon. The computer that controls the timing is to the left.  Balloon fragments littering the floor provide evidence of the trial and error testing that was necessary to fine tune the timing.

Moore also found a use for motion video in his experiment. As mentioned above, he had to measure the time interval from the red-eye reduction flash to the opening of the flash shutter and the main flash discharge. For the measurement, he aimed the camera flash at a rotating disc and discharged it. A digital video camera recorded the flash sequence. As it turned out, there were two red-eye flashes before the main flash. Each of the three flashes appeared on a different frame of the videotape. That can be seen in the following 4-frame video clip:  

VFW (avi format)      QuickTime (mov format)

Moore had measured the frequency of the disc with a stroboscope. He used that value together with a measurement of the angle between the first and last image of the marker line to determine the needed time interval.  (For more information on the use of motion video for high-speed photography, the Filming page in the Video section.)

Photographic results 
Complete text of Moore's paper in Acrobat PDF 
Back to top