Designing Experimental Procedures: A Low Cost, Hands On Project That Students Actually Enjoy

Because many students are unfamiliar with the development of experimental procedures, a project was sought that would allow students to use the design process to generate an optimal experimental procedure. An experiment that required students to determine the coefficient of drag for a small, dollar-store parachute figurine was selected. After a brief introduction to the pertinent theory, students were tasked to determine a governing equation for the experiment, identify measurable variables, generate concepts for experimental procedures, evaluate them based on customer requirements, conduct the experiment, analyze the results, and suggest improvements for subsequent iterations. Survey data indicates that most students enjoyed the project even though it required hard work, and that most learned how to design an optimal experimental procedure. Introduction Practicing engineers are often forced to answer complex questions by conducting an experiment and analyzing data. However before the experiment can be conducted, the engineer must devise an experimental procedure that will yield accurate results and not require an excessive expenditure of resources. The ability to develop an appropriate experimental procedure is so vital that the Accreditation Board for Engineering and Technology (ABET) requires that, “Engineering programs must demonstrate that their graduates have an ability to design and conduct experiments”. Given the number of experiments conducted throughout their undergraduate studies, most students that have graduated are well prepared to conduct an experiment that has an established set of experimental procedures. Unfortunately, students are not nearly as well prepared to design experimental procedures. Occasionally, laboratory exercises require students to develop experimental procedures. But even this requirement fails to satisfy the criterion set forth by ABET because design includes both the generation and evaluation of several potential solutions. Indeed, it is rare that students are ever required to design experimental procedures. Background With varying degrees of success, the Fluid Mechanics course at the United States Military Academy has traditionally been tasked to formally address the design of an experiment. This is somewhat challenging given that the course is one of the first engineering courses taken by cadets and is typically taken during the third year by both engineering and non-engineering majors. Further, it is highly unusual that any of the cadets have been introduced to a formal design process. Proceedings of the 2002 American Society for Engineering Education Annual Conference & Exposition Copyright 2002, American Society for Engineering Education P ge 783.1 Several of the previous projects that were used to address the issue of designing experimental procedures were theoretical exercises that focused on the development of an experimental procedure as opposed to the design of an experimental procedure. For instance, the previous project required cadets to develop an experiment that could be used to determine the viscosity of a fluid. While this project had some redeeming qualities, it did not illustrate that an experimental procedure can be designed in a manner not that dissimilar to an aircraft or an automobile. Additionally, students expressed the desire for the projects to be more interesting and hands on. Design of an Experiment Rather than attempting to fix an existing project a new project was sought. While there were many requirements for the project, by far the most important was that the project had to allow the students to focus on the design of experimental procedures. Operationally, this meant that the project had to be straightforward and have well defined relationships between the variables. The only other firm requirement was that the project must not be too costly. Beyond these requirements, potential projects should be interesting, hands on, and challenging. Because many cadets are interested in parachuting, an experiment was chosen that required cadets to determine the coefficient of drag for a small, dollar-store paratrooper. This task is fairly straightforward using the concepts of terminal velocity and dimensional analysis and is inexpensive enough for each group of cadets to be given their own test specimen. With a promising experiment identified, an abbreviated design process was introduced to the cadets. In its most basic form, design includes four phases, (1) plan for the design process, (2) develop engineering specifications, (3) develop concepts, (4) develop products. Although design is unquestionably an iterative process, emphasis was not placed on the iterative nature of design. Although much of the problem definition phase was given to the cadets, they did have to develop a schedule and try to find a time when all of the group members could get together. During the second phase, develop engineering specifications, cadets were required to identify governing equations and measures of goodness. Measures of goodness focused on minimizing uncertainty in the dependent variable and the availability of required resources. The third phase, concept generation, consisted primarily of modifications of a basic freefall procedure. There were two basic ways that the procedure could be modified, the cadets could modify the operating conditions or the techniques used to measure the parameters of interest. Modifications of the operating conditions consisted of altering the weight of the paratrooper, the height from which it was dropped, or even the fluid in which it was dropped. Additionally, alternative measurement techniques were considered. The evaluation of these concepts was accomplished by objective and subjective means. The uncertainty of the dependent variable was calculated using the following equation: