BYOE: A Low-cost Material Testing Machine to Increase Engagement in a Materials Science Lab Course

As a field, engineering is a profession with rich and deep theoretical foundations in each of its numerous subject areas. Helping students understand these foundational theoretical concepts can sometimes be difficult, and it is not uncommon for students to "get lost" in the details and fail to understand the main concepts. One way to help overcome this problem is to use laboratory classes. Laboratory classes provide students with hands-on learning experiences that help them connect theory and practice. One way students do this is by running experiments, collecting data, analyzing it, and comparing the results to those predicted by theoretical models. This discovery process can help build students’ confidence in existing theories and help them understand these theories at a much deeper level. Although this sounds great, the reality for students in many engineering programs is different. Laboratory equipment is expensive, and even in relatively small laboratory classes (such as one with a dozen students), equipment can be overbooked. Although one or two students may get to run the equipment themselves, the rest may not get those experiences and don’t benefit in the same way as the students who did. This is the exact situation the author faced with a materials science lab course. The bottleneck in this case was the tensile testing machine, and feedback from students about the class included the fact that much of their lab time was spent "sitting around" while they waited to be able to use the (one and only) tensile testing machine. To address this problem, we developed a small low-cost tensile testing machine so that students could eventually work concurrently in groups of 2 or 3, each with their own tensile testing machine. The current tensile testing machine prototype has a crosshead with 18in of vertical travel and replaceable load cells of 5kg(11lb) to 500kg (1100lb). This prototype uses a dual leadscrew with a hand crank, an optical encoder to measure distance, a load cell to measure force, and a small 7" monitor to display the results in real time to the user. An Arduino board is used for data acquisition from the encoder and load cell, and this is connected to a Raspberry Pi computer, which is in turn connected to the monitor. A wireless keyboard with an integrated track pad was used to interface with the machine, whose output is shown on the small 7" monitor. 1 Pedagogical Context The field of materials science is focused on connecting the concepts of structure, processing, and properties of materials. Materials science textbooks [4] often begin with the topic of structure, then move on to properties and processing. Many students have difficulty seeing the importance of studying structure, even though the structure of materials fundamentally drives the macroscopic properties we observe, and determines the utility and potential applications. One solution to this problem is to begin with an exploration of material testing and properties, then to ask the question, “why?” A specific example is, “why does steel elongate about 25% before breaking whereas polyethylene stretches to over 500% of its original length?” The effort to lead with properties can only be successful with functional testing hardware. Past efforts to do this have been met with frustration because students spend most of the time waiting for the (one and only) testing machine to become available, or merely watch others conduct tests. Another approach is to forgo physical testing altogether, and to watch videos of testing, or to perform testing simulations [10]. While there may be some benefit in watching others perform tests, it does not have the same impact as personal hands-on experience. Simulations are limited by the pre-programmed material options, and don’t allow exploration beyond these limits. Universal testing machines from leading manufacturers such as Instron and Tinius Olsen cost tens of thousands of dollars each, so purchasing additional machines is not an option for most universities. The primary motivation for this work was to develop a testing platform that would enable more students to engage in hands-on learning of materials science concepts. As a tool for active learning, this builds on decades of literature on active [6,8] and experiential [5] learning. These both draw on the concept of constructivism [3], which in turn relies on discovery learning [1,2], cognitive conflict [7], and learner-centered teaching [9]. Specific educational outcomes for a materials science course that will be served directly by this effort include: 1. Determine mechanical properties from an engineering stress-strain diagram 2. Understand concepts of stress, strain, Hooke’s law, and Poisson’s ratio 3. Understand and calculate true stress and strain 4. Define flexural strength and the influence of porosity for ceramics 5. Understand stress-strain behavior in polymers 6. Define hardness testing techniques and computation methods 7. Understand concepts of property variability and design/safety factors 8. Describe mechanisms of brittle and ductile fracture 9. Define creep and conditions under which it occurs and calculate steady-state creep rate 2 Mechanical System Design and Construction 2.1 Testing Machine Structure The structure for the material testing machine was built exclusively with materials available from a local steel supplier. Plain carbon steel with a material thickness as close as possible to 1/8 in was used, which allowed for high strength and easy weldability. This meant choosing either a 1/8 in (0.125in) or 11 ga (0.120in) thickness. The list of materials used to build the steel structure and cost is shown in Table 1. Prices shown reflect those advertised by online supplier metalsdepot.com. This was done to provide an upper bound on price for anyone interested in building a similar prototype, but all the materials were actually purchased from a local supplier who had them on hand at less than half the cost listed (about $50 USD as opposed to $100 USD). Specific structural dimensions are shown in the CAD drawings given in Appendix A. Table 1: Steel Structure Materials List Profile Dimensions Thickness Total Length Purchase Length Cost* square tubing 1”x1” 11 GA 86” 8 ft $25.16 rectangular tubing 1”x2” 11 GA 22” 2 ft $12.08 channel iron 1-1/2”x3/4” 1/8” 49.5” 6 ft $28.36 plate/strap 12” 1/8” 16” 2 ft $29.50 flat bar 3” 1/8” 21” 2 ft $6.20 flat bar 3/4” 1/4” 8” 2 ft $3.88 $105.18 *Cost estimated from www.metalsdepot.com. Actual parts were sourced from a local steel supplier at a lower cost. The base consists of a 12×16in plate, supported on the two ends by 11/2× 3/4×12in channel iron. Four lengths of 1 in square tubing are welded in a vertical position on top of the base of the structure, each set of two tubes was placed 1in apart and the crosshead fits between them. The crosshead is made from 1× 2× 11in rectangular tubing oriented in the strong direction. The top of the structure is capped by a crossbeam comprised of a 3× 11in flat bar welded to a piece of channel iron with the open side facing downward. The top crossbeam is secured with four 1/4 in bolts to allow for easy removal of the leadscrews. Under the base plate, another piece of channel iron 141/2 in long is welded to provide additional support and prevent the base plate from bending. This channel iron is welded with the open side facing upward, against the base plate. Further support is provided on the top of the base plate with an additional 1× 2in section of rectangular tubing welded between the vertical supports in the strong direction. Tools used for this process were an abrasive chop saw for cutting the steel to length, a grinder to remove burs, and a gas metal arc welder for joining the steel.