Skip to main content
To KTH's start page To KTH's start page

Towards improved motorcycle helmet test methods for head impact protection

Time: Mon 2019-11-11 10.00

Location: T2, Hälsovägen 11, Huddinge (English)

Subject area: Applied Medical Technology

Doctoral student: Shiyang Meng , Neuronik

Opponent: Associate Professor Mazdak Ghajari, Imperial College London

Supervisor: Professor Svein Kleiven, Neuronik; Peter Halldin, Neuronik

Export to calendar


Head injury is the leading cause of death and major trauma for users of powered two-wheelers (PTWs). Helmet use can prevent and reduce head injuries when a crash is inevitable. However, today’s motorcycle helmets do not protect equally against all types of head injuries. All helmets available on the market today are designed, manufactured and tested to meet certain standards. Current test standards evaluate helmet performance by dropping a helmeted headform perpendicular to an impact surface, and passing or failing a helmet based on the peak linear acceleration between 250-400G. Yet, real-world head impacts, being either linear (perpendicular) or oblique, impart both linear and angular head acceleration. Oblique impacts, which are known to be more common than linear impacts from in-depth analysis of motorcycle accidents, can transmit the tangential impulse to the head and hence cause the head to rotate. Head rotation has been hypothesised to be the main cause of traumatic brain injury (TBI) ranging from mild injuries such as concussions to more severe injuries such as acute subdural haematomas and diffuse axonal injuries. Therefore, there is a great need to develop test methods that replicate real-world accidents and reproduce realistic head impact responses. A number of potential test methods that subject the head to rotational insults are available today. However, there are still several questions that need to be answered: At what speed and angle should the helmet be tested? How boundary conditions of the head in the test methods, i.e. free, partially constrained or a surrogate neck, affect the kinematics of the head?

To answer these research questions, both experimental and numerical methods, such as finite element (FE) methods were used. Experimental tests for the helmet were performed using multiple test methods, providing a comparison between the test methods and data for subsequent validation of the FE helmet model coupled with anthropomorphic test devices (ATDs). FE Human body models (HBMs) with accurate anatomical structures and material properties were employed to evaluate the biofidelity of current test methods. Brain tissue strain of a head model resulting from direct impacts or inertial loadings were used to provide a direct causal link between the mechanical insult and the brain injury.

The first study in the thesis showed that both the US and European helmet standards lacked consideration for head rotation in linear impact tests. The US helmet standards use a partially constrained headform, which does not permit head rotation and hence not rotation induced TBI. European standards, on the other hand, adopt a free headform but the head rotation is not measured or assessed. The brain tissue strain resulting from the European standard tests at which rotation is allowed was up to 6.3 times higher than that in the US standards. In the second study, 300 simulations of possible motorcycle accidents were performed to understand the effect of impact velocity angle on impact severity. The results indicated that a 30o or 45o impact angle produced greater brain tissue strain than other impact angles, i.e., 15o, 60o and 75o. In the third study, it was found that when the helmeted head impacted the ground from low to high tangential velocities, i.e., 0-216 km/h, the motion of the helmet exhibited rolling and sliding phenomena. Since the helmet rolling and sliding phenomena govern impulses transmitted to the head-helmet system, and consequently the brain tissue strain, it is desirable to test helmets at speeds covering both the rolling and sliding regime. The tangential velocity at which motion transitioned from rolling to sliding was identified to be 10.8 m/s (38.9 km/h), given that the normal velocity is 5.66 m/s (20.4 km/h) and the coefficient of friction between the helmet outer shell and the impact surface is 0.45. In the final study, simulations with and without the experimental neck (Hybrid III) were compared to the HBMs. The results showed that the Hybrid III head-neck ATD used in the laboratory setting proved to correlate less with the head responses of the HBMs than the free headform. In particular, the Hybrid III head-neck ATD correlated poorly with the HBMs in axial (inferior-superior) acceleration and over-predicted the maximum angular velocity by up to 75%. However, the free headform was also limited in replicating the chin-neck and helmet-torso interactions. The need for a more biofidelic surrogate neck, especially under axial compression, is evident.

In summary, this thesis demonstrates methodologies for a reason and objective based decision making process and provides important information in the design of future helmet test methods and standards. Some of the major findings in this thesis, despite focusing on motorcycle helmets, can also be applied to other types of helmets.