I am a researcher in the Department of Mechanical Engineering, at the University of Sheffield, in Sheffield, UK.
My current project, in conjunction with NTUA Athens, Mecalog, Fiat, Volvo, PSA ,Renault , Breed Italia and Bertrand Faure is to improve the accuracy of car crash test simulations.
Crashing cars is expensive; the aim of the project is to make simulated car crashes more accurate so that fewer tests need to be done on real cars.
My project is concerned with accurately measuring the properties of the materials used to construct car interiors and crash test dummies, in order to improve the accuracy of the interactions between them.
The difficulty is that these properties are dependent on strain rate, and so to determine how the materials will behave in crashes they must be tested at the speeds encountered during crash tests, in the region of 50mph.
This has led to a number of experimental difficulties, the tests last only about 2.5mS, much faster than the response time of hydraulic valves, and orders of magnitude faster than hydraulic systems. The machine needs to exert forces of around 10kN at velocities of 20m/s. Simple mechanics shows that this is an instanteneous power of 200kW.
The solution chosen was to use a cam operated device, incorporating a large flywheel to provide the instantaneous power required.
The motor spins the 125kg flywheel at between 40 and 3000rpm. Once up to speed a PC closes a relay, activating the pneumatic clutch which connects the cam to the flywheel. As the bottom platten rises it's position is measured by a capacitative transducer, and the compressive load is measured by the load cell. Load and displacement information is stored by the PC along with an internally generated time log.
Max Load, 10kN
Platten speed range 0.2 - 20m/S. velocity profile governed by cam shape.
Motor Power, 37kW