This proposed legislation presents the challenge of designing a bumper system which achieves the required performance in the leg-form impact-without sacrificing the bumper's primary function of vehicle protection during low-speed impacts. The first step in meeting this challenge is to understand what effects the front-end geometry and stiffness have on the leg-form impact test results. These results will then need to be compared to low-speed impact performance to assess if the two requirements are compatible.

This paper describes an investigation-using concept Finite Element models and a front-end variable geometry vehicle test buck-of the styling and engineering tradeoffs for a pedestrian safe bumper system.

Within the finite element models solid elements using material types such as honeycomb, advanced foam curvilinear recoverable, strain rate foam recoverable, and low density foam were evaluated as to their accuracy in simulating ConforTM foam on the pedestrian leg-form and polyurethane energy-absorbing foam on a bumper beam under extreme compression or deformation conditions. Extreme deformation which occurs after 80% compression can cause excessive hourglassing of certain types of elements. During this extreme event many solid element material types will not exhibit the correct foam behavior, consequently the results lead to an incorrect prediction. This study attempts to determine the best material type to use during this type of large deformation impact.

One of the challenges to vehicle design introduced by these tests is the impact of an adult pedestrian headform to the top of the fender. The proposed acceptance level for Head Injury Criterion (HIC) is less than 1000 during impacts at 40 kmlh. This paper uses the finite element (FE) method to predict the influence of proposed fender and shotgun design modifications aimed at meeting this target. In addition, the known issues with the implementation of these proposed changes are discussed.

Although the proposed changes are shown to meet the target in the theoretical analyses presented in this paper, these changes are also demonstrated to conflict with other aspects of vehicle safety (frontal visibility and frontal high-speed impact), vehicle manufacturing, and durability.

The most common method proposed for cushioning the lower limb in an impact uses an energy absorber (plastic foam or ‘egg-crate’) in front of a semi-rigid (steel or aluminum) beam. There are also proposals for ‘spring-steel’, steel-foam composites, crush-cans, and plastic beams. The most common method proposed for supporting the lower limb in an impact is a secondary lower beam, known as a ‘stiffener’ or ‘spoiler’. Most proposed lower stiffeners are plastic plates or metal beams supported by the engine undertray, the radiator support, or the front-end module. In addition to these concepts, there are a number of design proposals involving a deploying bumper or lower stiffener.

A simpler finite element model of human bones could be developed if it were shown that less complex finite element material modeling provides sufficient prediction of long bone macro-level strength. This study investigates the importance of including material anisotropy in the finite element model of a human femur. Four composite femur models were developed: linear orthotropic, linear transversely isotropic, linear isotropic, and non-linear isotropic. Each model was used to simulate anterior-posterior (AP) bending and external-internal rotation. Comparison of the results with physical tests indicates that the global elastic force-deflection response of the whole femur in AP bending is sufficiently described by isotropic material models of the two constituent tissues. The more complex (more detailed anisotropic) material models do not enhance the results of this simulation. However, the global response of the femur in external-internal rotation does indicate that increased material model complexity (or higher degree of detail in material anisotropy) can provide improved prediction capability.

The work presented in this paper first reviews current legforms developed for pedestrian impact testing, including the TRL impactor used in EuroNCAP tests, Honda's POLAR II pedestrian dummy, and JAMA/JARI's FLEX-PLI legform impactor. Component level testing shows the FLEX-PLI performance to be close to the human lower limb response corridors. However, there are still areas of potential improvement with this design. To address these areas, this research includes the phase 1 development of a new legform impactor incorporating adjustable ligament pre-load, direct ligament strain measurements, adjustability in knee flexion to account for the gait cycle, tuned composite bone cores that match the force-deflection curves of PMHS testing, and a condyle load plate that measures knee joint compressive forces. Development techniques include solid modeling, computer-aided manufacturing, composite design and analysis, sensor specifications, and data acquisition. In so doing, the injury specifications set by the EEVC will be modified to account for the more accurate injury assessment of the improved device.

Responsibility for making the most effective educational use of a design competition is shared between the students, the faculty advisor, and the competition organizers. Design competitions build student enthusiasm; however, there are some things they learn that we may not want to be teaching. Some of the educational shortcomings of these activities are highlighted, with suggestions on how to manage them. In particular, this article focuses on the risks of (a) distraction from classes, (b) a build-and-test approach, (c) advisor co-opted designs, and (d) design changes for their own sake. The influence of the advisor and the competition rules on each of these concerns will be discussed. Finally, the competitions themselves will be investigated to see how the form of the events may be improved to further enhance the learning opportunities for the students.

The mini-project consists of a design evaluation task which is investigated using three different methods: closed-form analysis, finite element analysis, and simple model build and test. The task is to select the better of two alternative support structures for a heavy-duty material-handling conveyor belt. Acceptance criteria in the form of maximum deflection and stress are provided. The closed-form analysis is conducted using Castigliano’s method. Beam and shell finite element models are built and analyzed in Abaqus CAE. Structural prototypes are constructed with PASCO Structures System components and tested with simple weights and scales.

The strengths of this combined approach are that the students (a) gain experience with the three different methods of stress/deflection analysis, (b) compare the different methods on a single problem, and (c) check or confirm their own results. By using existing finite element software licenses and available PASCO components, the project took no additional lab time and no additional cost to implement.

Since the first use of this lab project was with in a small class, no direct measures were used to capture improvements in student learning. Instead, Indirect measures (instructor observations, student comments) were used to evaluate the lab’s effectiveness. During the build and test phase of the project, students were much more engaged (discussing, changing approach) than in prior years when only the closed-form analysis was included. Student closed-form analysis results were also more complete, with fewer (20% versus 40%) of the student groups forgetting to include parts of the structure in their Castigliano’s analysis. On end-of-course evaluation forms, students commented that they enjoyed comparing results between the three methods, and appreciated having something physical to test rather than just performing calculations.

This paper presents a case study of a successful six-year partnership between the Automotive Bumper Project committee of the American Iron & Steel Institute (AISI) and a mechanical engineering department. The AISI Bumper Project has sponsored seven senior capstone design projects and three master’s projects, providing excellent educational opportunities for twenty-five students. The projects ranged from specific vehicle bumper designs to building and testing a high-energy pendulum impact tester.

The university benefited from this long-term relationship by gaining relevant student projects, supporting graduate students, and retaining a connection with industry. The industry consortium benefited by encouraging the study of topics of interest (steel design, impact analysis) at the undergraduate level, receiving ‘outside-the-box’ design concepts, and learning how bumpers may be affected by future trends. The costs on both sides were kept low, enabling most of the funds to go directly toward hardware so the students could build and test their designs.