In December 2015 NHTSA announced the intention to introduce a new, modified USNCAP vehicle test protocol. Among other proposed changes NHTSA has announced the introduction of a new full vehicle crash test, frontal oblique, and the introduction of a new ATD - THOR. In the THOR, ankle xversion and dorsiflexion injuries may be predicted based on the bending moments calculated from the THOR lower tibia load cell readings. These readings are transformed to the ankle joint location, and corrected through inertia compensation from acceleration readings measured at the mid-shaft of the tibia.

This approach is subject to the following assumptions. First, the mid-shaft tibia mounted accelerometer is assumed to read the same acceleration as the ankle potentiometer block. Second, the mid-shaft tibia, distal tibia load cell, and ankle potentiometer block are assumed to move as a rigid body. Third, there must be no alternative load paths applying force or moment to the distal tibia between the tibia

load cell and the ankle joint. The goal of this study was to examine oblique crash tests performed by NHTSA to observe mechanisms of loading of the ankle and distal tibia to elucidate the validity of the ankle joint moment calculations. We examined 35 USNCAP oblique crash tests with THOR dummies seated in both the driver and the passenger seat. From each crash test, we compared the calculated ankle joint moments to the joint angles read. Results were also compared to a series of oblique sled tests and component tests using THOR.

The results indicate that there is often an inconsistency between the component/sled tests and full vehicle tests featuring THOR ATD when predicting ankle injuries. In several cases the calculated joint moment does not correspond with the ankle rotation angle expected from the biofidelity component requirements. This observation was made for both dorsiflexion and xversion.

A case by case analysis of data points located away from the biofidelity corridor revealed multiple mechanisms responsible for the lack of comparable results between the biofidelity requirements and full vehicle crash tests. First, in some cases an alternative load path was present, applying a load to the distal tibia between the ankle and the load cell. Second, in multiple cases there was an interaction between the mid shaft of the tibia and interior of the vehicle that resulted in a short duration spike in the recorded tibia acceleration. Due to the proposed inertial compensation, the spike in the acceleration was carried over into the ankle moment calculation resulting in artificial moment prediction when no ankle rotation was present. Third, in several tests data acquisition problems were observed in the NHTSA tests (spiking channels, lost channels, or polarity errors) that resulted in incorrect or incomplete moment calculations pushing the results away from the biofidelity requirements.

In conclusion, alternative load paths at the distal tibia, and acceleration spikes in the tibia can cause an inconsistency between the moment and angle read for the THOR LX ankle in crash tests. Thus, the ankle moment calculation should be verified prior to applying to injury risk prediction to ensure that the results are not artefactual.