Development of a vibration fatigue failure prediction model for automotive lamps made of polymers

This thesis presents a technique for numerically modelling and predicting the fatigue life of automotive lamps made of thermoplastic polymers. The primary aim is to use the fatigue failure prediction technique developed in this research to address fatigue challenges in the design of automotive lamps. The research combined numerical simulation and experimental studies. As part of design verification, automotive lamps are subjected to accelerated random vibration tests during design to assess their integrity over a life-time exposure to mechanical vibration loading. Often, the lamp assemblies experience fatigue failure during testing. Currently, in the design phase, it is difficult to identify the weaknesses within the assembly. The product development is hence based on trials. This results in a costly and time-consuming design cycle. Having a robust mathematical failure prediction model (a virtual prototype) will eliminate many of the shortcomings listed, enabling reduction in the design cycle, eventually successfully meeting vibration specification economically and efficiently. However, there are challenges in numerically predicting the fatigue of polymers. Modern automotive lamp assemblies have complex geometry and are composed of different parts made of polymer materials. Polymers exhibit hyperelastic behaviour and they also exhibit inter-sample variation due to manufacturing processes. Some of the material properties required for dynamic and fatigue life analysis are not readily available. The key to robust fatigue analysis of lamp assemblies is the availability of reliable input parameters, such as identifying the reliable hyperelastic material model from the existing models, fatigue data (S-N curves) and variability in the behaviour of manufactured constructions. Also, dynamic and fatigue models need to be established. A study was conducted to identify from the existing models, hyperelastic material model that can reliably describe the hyperelastic behaviour of polymers including the ability to represent inter-sample variations. Subsequent to the study of hyperelastic material models, the effect of material nonlinearity on the random vibration response and fatigue behaviour of the polymers were assessed. The fatigue data (S-N curves) used in the fatigue life evaluation were experimentally determined. The true fatigue property of Polycarbonate (PC) could not be obtained using the standard fatigue test machine, this was attributed to thermal softening. A novel test rig for measuring bending fatigue curves of materials based on specimen resonant behaviour was developed and used to experimentally characterise the fatigue data for the PC material. Prior to modelling the dynamic response and the fatigue life of the lamp assembly, modelling the response and fatigue life of individual components was performed using a simple beam. Various fatigue failure theories were explored to find an appropriate model describing the fatigue behaviour of polymers of automotive lamps. The experimental testing covered characterisation of baseline material properties and validation of dynamic response and fatigue life. It was found that the level of error in the vibration response and fatigue life resulting from material nonlinearity depended on the level of strain the components are operating at. The higher the strain, the larger the error. The lamp studied was operating at small elastic strain (within the materials linear region) under the standard automotive random vibration loading. Therefore, linear material model was used to model the vibration and fatigue behaviour of the lamp assembly. Steinberg theory predicted the fatigue life of the lamp assembly with high degree of accuracy. This developed fatigue life prediction technique has made a significant contribution in the design of automotive lamps.

Grantor dc:publisher

Oxford Brookes University

Year dc:date

2020

Author dc:creator

• Okeke, Chukwudi

Contributors dc:contributor

• Thite, Anand
• Durodola, John
• Greenrod, Mike

Statement dc:rights

• All rights reserved

Language dc:language

en

DOI dc:identifier

https://doi.org/10.24384/zcbe-bd15

OAI identifier oai:identifier

tle:fbd5b887-6c0c-4592-bd04-bf1f90febe40:d6bd9758-527a-46cd-bfe2-c433766e8fca:1