FINITE ELEMENT ANALYSIS OF A BICYCLE HELMET John Hutchinson, Chuck Rogers, Jack Bish, Keith Friedman, Anthony Sances *, Srirangam Kumaresan * Friedman Research Corporation, SantaBarbara, CA, USA * Department of Computer Science, University of California, SantaBarbara, and Biomechanics Institute. SantaBarbara. CA. USA Bicycle helmets are widely used as safety devices, especially among children. The construction of these safety devices was, until recently, unregulated and the certification of these devices depended upon several different standards such as ASTM, ANSI and SNELL. These standards describe a series of qualifications a helmet must satisff in order to be certified. Improvements in helmet design can be determined early in the development stage through the use of finite element studies. This will result in safer helmets that are tested over the entire helmet and .range of feasible orientations. The present study was designedto examine the level of protection at various impact locations and the effects of altering the geometry and material. The finite element technique was used becauseof its capability to simulate irregular geometry and conduct parametric study [1-e]. A finite element model of a production bicycle helmet was developed. The surface geometry was obtained using a 3-D digitizer. The helmet geometry was then meshed with 31,000 tetrahedral elements using the HyperMesh finite element pre-processor [10]. Impact simulations were run using the LS-DYNA finite element program, which models the large-deformation, dynamic response of structures using an explicit time integration algorithm [l]. A Hybrid III crash-test dummy head model was inserted into the helmet, and assigned magnesium alloy material properties to simulate the metallic head forms used in helmet certification tests. The finite element model is shown in Fieure l. 3 deg. Figure 1: Front view of helmet/head finite element model illustrating the impact orientation with respect to a vertical rigid wall. The helmet foam was characterized using the "crushable foam" material definition. The nonlinear crush stiffrress profiles of various helmet materials were obtained from a series of compression tests performed on helmet foam samples at various strain rates. The model was then validated and examined in an impact configuration located below the helmet certification test line. The peak head accelerations obtained from the finite element simulations compare favorably with test data for each of the three tests, see Table l. For all tests, the finite elementresults are within 4yo of the test results. Table l: Validation results comparing the physical drop tests and the finite element simulations (peak S Droo Tests FE Simulations TestNo.I r99 207 Test No. 2 209 206 TestNo.3) l9l 195 The effects of altering the geometry and material properties on a portion of the helmet were examined. The helmet modeled in this study has a distinctive design feature in that the flap behind the ears tapers down in thickness. As a result. the rim of the helmet is rather thin compared to other helmets on the market. In order to investigate the protective capability of the thin rim, a 14- mph side-impact drop test simulation was re-run, but with the head and helmet tilted sideways an additional 3 degrees (counter-clockwise looking forward) to direct the impact lower down on the flap. This impact results in a peak head deceleration of441 g's, which far exceeds the criterion of300 g's prescribed by the standard. The geometry of the helmet was varied by increasingthe thickness of the flap behind the ear from 0.35 - 0.50 to 0.65 - 0.68 inches at the rim (Figure 2). This change removed the marked thinning of the helmet at this location and made it comparable to other production helmets. Under the same impact conditions as above, this helmet resultsin a peak head deceleration of 215 g's, a 5lYo reduction over the baseline helmet, and well below the 300 g certification criterion. The sharp increase in head deceleration for the baseline helmet model is due to the fact that the thin helmet flap crushes flat early on in er $ the impact resulting in a marked increasein the stiffness n ?\ of the helmet foam. The thicker helmet rim of tfrr [ $ modified helmet absorbs more impact energy before the l.X t.: foam material becomesrigid. * CO CD sa1 CD ''-