High-Speed Photography

Authorship: Mark Warren, Martin Oltmanns

Department of Physics and Astronomy, Augustana College, Rock Island, IL 61201

Abstract: The following report is an exploration in the strength of high speed photography in the observation of fast occurrence and small durations of time. In a controlled environment, cameras are far more powerful than observer and can be used to isolate a window of time for detailed analysis. In this exploration the motion spinning belt was recorded and the average translational velocity over a short time and noticed small fluxuations that would otherwise be impossible to discover. High speed photography was also used to observe a balloon popping and could accurately detect average velocity of the tear as it moved down the edge of the balloon

I. Introduction

The use of high speed photography is very useful for events which the human eye cannot observe due to either the motion of an object or its extremely minimal existence in time. For example, measuring the velocity of a moving object is very error-prone when relying simply on a measuring tool and a stop-watch. The experimenter’s reaction speed must be taken into consideration which weakens the strength of any possible conclusion to follow.

The first demonstration is done by using a long exposure digital camera and a tunable strobe-light. The strobe light will flash at a frequency (f) chosen by the experimenter. If a fixed camera takes a long exposure over several flashes, the picture will show several iterations of a moving object at different times. The difference in positions of each iteration can be used to find an average velocity.

This method was used to find that the belt on a Vandergraaf generator has a translational velocity of .233 m/s.

High speed is more useful in occurrences that do not last long enough for the human eye to notice. The popping balloon is a great example. The apparatus is much more intricate however the theory and goal is the same. From two successive pictures of a balloon “mid-pop” can be used to discern the speed in which the balloon tears.

II. Experimental Setup

The measurement of velocity was demonstrated with a belt for from a Vandergraaf generator. By placing a mark on the white belt, a single point can be observed as it cycles around the machine. Next to the belt a meter stick was held upright for later reference. A digital camera was held stationary on tripod, and the lights were turned off. Using a strobe-light with a tunable frequency, the camera recorded several flashes on a long exposure. This process was repeated until four marks were visible in the resulting picture. The picture then could be used to find the average velocity over a distance or period of time.

The photography of popping balloon is more sensitive. This time, instead of a strobe light, the flash was triggered by sound. Two microphones were placed at different distances with respect to the balloon. They fed into a tape recorder which gave output to a silicon switch. The switch shorts the electrodes in the camera’s flash device which makes a flash. When two microphones are used, we can achieve two separate flashes which are delayed apart by the difference in time that it takes the sound to reach the microphones. This time difference is not easily to calculate, so a rotating with a known period is implemented to establish a time a reference between flashes. Both flashes can be recorded on a digital camera set to a long exposure. The resulting picture has two superimposed images of the balloon popping at different times.


III. Results

The following picture is the long exposure taken from the strobing process. On the right is the white belt, and each black “<” represents the position during the flash. Arrows gave been drawn to connect the belt and the ruler and the left.

The positions were tabulated in the x (m) column below, as well as the frequency and period of each flash. Using the position, three average velocities were calculated between each position. The mean was found between the three which matched the average velocity from point 1 to point 4.

f(s^-1) / 4.333333 / period(s) / 0.230769
part 1b / x(m) / V(m/s) / average / -0.23256
1 / 0.296 / mean / -0.23256
2 / 0.246 / -0.21667
3 / 0.19 / -0.24267
4 / 0.135 / -0.23833

The following images recorded the process of popping the balloon with the sound-trigger apparatus. In the first figure the two images occurred in the time that it took the wheel to rotate approximately 20 degrees or 0.056 revolutions. The wheel spinning at 7.02 rev/s corresponds to a 0.0077 second window of time. In this time the tear moved 7.4 cm. The average tear velocity is then 9.346 m/s

Using the same method we find that the second tore at a velocity of 7.98 m/s

IV. Discussion

The strobing process was actually more difficult than initial consideration suggested. The belt was moving at a fast enough rate that the picture often did not take until the mark on the belt was out of view.

However we still get a very clear picture of 4 separate positions. What is also worth mentioning is that each velocity is different from the mean on the order of a few cm/s. We can still say that the belt turns with essentially constant velocity, but we can also say that the belt may slip or have slight bursts now and then. This could be the actual resistance due to charge building in the generator. What is certain is that this would be unnoticeable for human observer with just a ruler and a stop watch.

In recording the popping of a balloon, the apparatus is very sensitive to small fluxuations in the instruments. The rotating wheel was a considerable source of uncertainty seeing that angle of rotation in the picture is hard to determine by inspection. The size of the balloon also affects how quickly it tears. It was a goal to make balloon approximately 25 cm long but the balloons could have still had different volumes. With that said the tear velocities were reasonably close for an object moving that quickly. It is important to keep in mind that these average velocities are far more powerful because they give a more detailed look at the motion through time.

*estimate and propagate experimental error

*comparisons of rip speed and balloon properties?

*late -10

*90-10=80%