Abstract
Research and /or Engineering Questions/Objective:
NHTSA has been developing an advanced frontal impact test dummy called THOR. Various prototypes have been created in recent years and now the final components are being assessed for formal use in protocols. The objective of this study was to create a multi-body model of the dummy, in order to investigate the feasibility of using numerical simulation to predict the dynamic behaviour of the THOR dummy in crash situations. Numerical solutions are highly interesting to the automotive market as there are very few THOR hardware dummies available, making simulation a preferred tool for design development and evaluation.
Methodology:
NHTSA has made CAD data of the THOR available, making it possible to generate a multi-body model of the dummy that accurately captures its dimensions and mass/inertia properties. Several institutions have performed THOR tests in different test conditions for NHTSA to evaluate the physical behaviour of THOR in typical impact scenarios. By using the information from these tests and comparing the simulation results to those obtained experimentally, it was possible to validate the MADYMO model of the THOR. Component certification test data were used to check the model design and then to define and optimise local dummy parameters. The kinematic response of the whole dummy model was then validated against data obtained from full-scale tests of the hardware dummy.
Results:
This paper presents several time history curves which compare the response of the hardware dummy with the predictions made by the simulation model, for both tightly-controlled component tests and full-dummy tests representative of the dummy’s regular use. This study shows that a multi-body simulation technique can be used to represent the THOR test dummy’s response to dynamic loading accurately.
Limitations of this study:
The primary limitation of this study is the lack of test data specifically available for the latest version of the dummy. As the hardware has been continually redesigned and improved, much test data is rendered irrelevant by changes in the design and response. The extent of validation will be increased as more test data becomes available.
What does the paper offer that is new in the field in comparison to other works of the author:
The latest THOR dummy is a completely new test dummy. A multi-body model of this device is therefore a completely novel product and is of high interest to the market, as the physical dummy is not readily available (making simulation the preferred tool), and the solution times for a multi-body analysis are generally orders of magnitude lower than those for finite element simulations.
Conclusion:
A new model of the latest THOR test dummy has been developed in MADYMO. The validation of the model is sufficiently complete and accurate for the model to be used by the automotive industry for evaluation studies.
KEYWORDS – Frontal impact; THOR; MADYMO; Simulation; Crash testing