You are on page 1of 7

Home

Search

Collections

Journals

About

Contact us

My IOPscience

Simulation work of fatigue life prediction of rubber automotive components

This article has been downloaded from IOPscience. Please scroll down to see the full text article. 2010 IOP Conf. Ser.: Mater. Sci. Eng. 11 012009 (http://iopscience.iop.org/1757-899X/11/1/012009) View the table of contents for this issue, or go to the journal homepage for more

Download details: IP Address: 59.162.23.79 The article was downloaded on 27/03/2012 at 12:57

Please note that terms and conditions apply.

9th National Symposium on Polymeric Materials (NSPM 2009) IOP Conf. Series: Materials Science and Engineering 11 (2010) 012009

IOP Publishing doi:10.1088/1757-899X/11/1/012009

Simulation work of fatigue life prediction of rubber automotive components


M. S. A. Samad1 and Aidy Ali2* Automotive Engineering Unit, Institute of Advanced Technology, University Putra Malaysia, 43400, UPM Serdang, Selangor, Malaysia
2 1

Department of Mechanical and Manufacturing Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 UPM, Serdang Selangor, Malaysia *E-mail: aidy@eng.upm.edu.my

Abstract. The usage of rubbers has always been so important, especially in automotive industries. Rubbers have a hyper elastic behaviour which is the ability to withstand very large strain without failure. The normal applications for rubbers are used for shock absorption, sound isolation and mounting. In this study, the predictions of fatigue life of an engine mount of rubber automotive components were presented. The finite element analysis was performed to predict the critical part and the strain output were incorporated into fatigue model for prediction. The predicted result shows agreement in term of failure location of rubber mount.

1. Introduction Rubber is an elastomeric material that can absorb a large number of energy compared to other type of material such as metal, wood, and plastic. Due to its large energy absorbing properties, rubber is one of the best choices of materials that can act as a damper. Because of that, the applications of rubber components are always related with fluctuated load, random vibration or large deformation. An example of implementation of rubber is an engine mount, which is subjected to both static and cyclic loadings. Overtime, the component will experience fatigue because of the operating condition. Therefore, it is necessary for manufacturer of rubber component to be able to predict their products life cycle given the actual operating condition to prevent failures during operation. Fatigue life determination is also an important factor to determine the extent of warranty of any rubber component. In a literature surveys conducted by Mars and Fatemi [1], they stated that there are two ways of predicting the fatigue life of rubber. The first method is the crack nucleation method, and the second method is the crack growth method. They also mentioned that the first method is popular among the researchers and engineers. This is due to the determination of fatigue life is correlated with quantities of continuum mechanics, such as strain energy density. In finite element analysis (FEA), the easiest method to be implemented into a user subroutine is the crack nucleation method. The result from the FEA is post processed in order to get the fatigue life cycle. Numerous of studies [2-4] have been conducted in predicting the fatigue life of engine rubber mount using FEA and the crack nucleation method. In order to verify the result of FEA it must be compared with experiments. Fatigue life experiments can be conducted, either by using load controlled method, or displacement controlled. For the load controlled experiment, the minimum load was chosen based on point at which the specimen had finite life. Up to 7.06 kN of load with a different mean load has been experimented [3]. In a paper written by Woo and Kim [5], they
c 2010 Published under licence by IOP Publishing Ltd
1

9th National Symposium on Polymeric Materials (NSPM 2009) IOP Conf. Series: Materials Science and Engineering 11 (2010) 012009

IOP Publishing doi:10.1088/1757-899X/11/1/012009

suggested a method of predicting rubber components by test of material properties. The first test to be conducted is to determine the hyperelasticity constant. Afterwards, the second test is followed by 3D dumbbell fatigue test. The second test was compared with the FEA to obtain the values such as stress and strain with the purpose to correlate with the fatigue life. Throughout a numbers of comparisons, they proposed that maximum Green-Lagrange (G-L) strain provides good correlation with fatigue life. In this study, crack nucleation method will be used to predict the life cycle of the engine rubber mount. 2. Crack Nucleation Method The crack nucleation method is most favorable method in incorporating the subroutine into finite element software because of its life is determined by the history of stresses or strains at any point. In the crack nucleation method, there are two widely used as a parameter to determine life of rubber, namely maximum principal strain and maximum energy density [1]. These two methods are popular because of the measurement of strain is easily determined from experiment by calculating the displacement from the initial point, while the energy density can be determined from the hyperelasticity strain energy function. Previous studies [2, 3] in comparison between the accuracy of determining the life cycle of a rubber shows that maximum Green-Lagrange strain provides a better correlation with fatigue life compared to maximum energy density. The following correlation obtained by Kim [3] is used to determine the life of a rubber component is programmed into the ABAQUSs user variable (UVARM) subroutine. (1) is fatigue life cycles. In the case of UVARM subroutine,

Where by

is the G-L strain and

the program will be called at every time step and accessed the material point data. Therefore, the material point data can be manipulated at user level. In this particular case, the value of the principal strains (G-L strains) at every material points will be called at all time step, and then be used to calculate the life cycle. Ultimately, the quantity of the life cycles value is returned back to ABAQUS as a field output. 3. Methodology 3.1. CAD data generation The study was undertaken on an engine rubber mount which is used in Malaysian local automotive cars. It consists of the rubber section and metal section as shown in Figure 1. This component is subjected to both static loading and fatigue loading during idle and during operation. Therefore it is a suitable component to be tested out. Before the simulation can be started, it is important that the Computer Aided Design (CAD) data is generated beforehand. Since this component is a commercial part, it is impossible to get the CAD data due to proprietary. Hence, to overcome this issue, a Coordinate Measuring Machine (CMM) and 3D Scanning Machine were used to generate the CAD data as illustrated in Figure 2.

Figure 1. The Engine Rubber Mount


2

9th National Symposium on Polymeric Materials (NSPM 2009) IOP Conf. Series: Materials Science and Engineering 11 (2010) 012009

IOP Publishing doi:10.1088/1757-899X/11/1/012009

Figure 2. The coordinate measuring of engine rubber mount (left) and the CAD data produced (right) CMM is capable to measure some of the features, such as length, point, and circle in term of coordinates. The coordinates measured can be exported to common CAD data and thus can be modified. In order to ensure the accuracy between the actual parts and the CAD data, 3D Scanning machine is also used to assist the generation of surface on the component as shown in Figure 3. Besides that, it allows the comparison between these surfaces drawn on CAD and surfaces scanned by 3D scanning machine. The final produced CAD data are shown on Figure 4.

Figure 3. The 3D Scanning machine (left) and the CAD data generated (right) y-axis

Figure 4. The final CAD data. 3.2. Finite Element Analysis The finite element analysis is conducted using ABAQUS 6.8 EFI. The geometry is meshed using 10 nodes linear tetrahedron which is the C3D10 element. In order reduce the analysis time, symmetry
3

9th National Symposium on Polymeric Materials (NSPM 2009) IOP Conf. Series: Materials Science and Engineering 11 (2010) 012009

IOP Publishing doi:10.1088/1757-899X/11/1/012009

boundary condition is applied. On the other hand, the material constant is obtained based on previous experiment conducted by Kim [3].Therefore, in this FEA, it is assumed that the material that has been tested by Kim is the same. A displacement on y direction ranging from 6 mm to 14 mm is applied as depicted in Figure 5. The result is post-processed to yield the fatigue life prediction as displayed in Figure 6. For future comparison between experiment and the FEA result obtained from this study, the load ratio in the experiment must be equal to zero.

Figure 5. Finite element analysis result of 8 mm displacement (left side) and 14 mm displacement (right side).

14 mm 12 mm 10 mm 8 mm 6 mm 4 mm 2mm -2 mm -6 mm -4 mm

Figure 6. The fatigue life predicted on y direction of engine rubber mount.

9th National Symposium on Polymeric Materials (NSPM 2009) IOP Conf. Series: Materials Science and Engineering 11 (2010) 012009

IOP Publishing doi:10.1088/1757-899X/11/1/012009

Table 1. Predicted fatigue life cycle

Displacement [mm] -6 -4 -2 0 2 4 6 8 10 12 14

Max G-L Strain [m/m] 0.569196 0.364346 0.172402 0 0.086224 0.182612 0.287798 0.397986 0.510461 0.624616 0.885005

Predicted Nf [cycles] 1.58E+08 7.63E+09 5.11E+12 NIL 2.11E+15 3.10E+12 5.93E+10 3.54E+09 4.07E+08 7.03E+07 3.40E+06

4. Discussion The finite element analysis provides a clear view of the critical area that prone to crack. The comparison between the failure location of crack conform to the actual location of crack occurred on the engine rubber mount during operation. Furthermore, the maximum principal strain obtained from unique nodal solution can be considered the location of the first occurrence of crack, thus under the influence of fatigue loading causes the crack to growth. During operation, if it is assumed that normal engine operates at 3000 rpm during running, therefore the frequency of the forced vibration is 50 Hz. If the forced vibration has the maximum displacement, thus minimum fatigue life predicted in the finite element analysis is 3,400,000 cycles, as shown in Table 1. It can be translated to 19 hours of operation. However, an engine rubber mount can stand in a long period of operation in actual condition, which means that the displacement must be less than 14 mm. On the other hand, since the early assumption is the rubber tested by previous researchers is the same in our studies, therefore when the experiment is conducted, it might not yields the same result. Rubbers mechanical properties rely heavily on the formulation of the rubber itself, the content of fillers, the amount of carbon blacks, and other materials. The percentage of carbon black is also one of the factors affecting the fatigue life [6, 7]. Nevertheless, the prediction still lacks in term of comparison with the actual experiment. However, based on preliminary results, it is rational to support that FEA has successfully pin point the exact critical location or region of fatigue crack initiation in engine rubber mount as illustrated in Figure 7.

Figure 7. The fatigue life predicted on y direction of engine rubber mount. The left side of the superimposed image was adopted from previous research [8].
5

9th National Symposium on Polymeric Materials (NSPM 2009) IOP Conf. Series: Materials Science and Engineering 11 (2010) 012009

IOP Publishing doi:10.1088/1757-899X/11/1/012009

5. Conclusions Finite element analysis performed on Malaysian local car manufacturers rubber mount has been successfully conducted. The comparison between the finite element analysis and the failure location on the rubber component shows promising result in predicting the fatigue life of engine rubber mount. In the near future, experiment on the actual component is necessary to compare the predicted life on the analysis. Material properties study might as well be conducted to ensure accurate result can be achieved. 6. Acknowledgements The author wishes to thank to Malaysian Rubber Board (MRB) on the support and Universiti Putra Malaysia (UPM) for scholarship awarded to M S A Samad as one of the authors. References [1] Mars W V, Fatemi A 2002 A literature survey on fatigue analysis approaches for rubber International Journal of Fatigue 24: 949-961. [2] Woo C S, Kim W D, Kwon, J D 2008 A study on the material properties and fatigue life prediction of natural rubber component Materials Science and Engineering: A 483-484: 376381. [3] Kim W D, Lee H J, Kim J Y, Koh S K 2004 Fatigue life estimation of an engine rubber mount International Journal of Fatigue 26: 553-560. [4] Li Q, Cai J, Zhao, Zhao B 2009 Fatigue life prediction of a rubber mount based on test of material properties and finite element analysis Engineering Failure Analysis 16: 2304-2310. [5] Woo, C S, Kim W D 2008 Interlaminar crack initiation and growth under modes I and II in a carbon-fibre epoxy composite subjected to fatigue WIT Transactions on the Built Environment PP. 285-294. [6] Saintier N, Cailletaud G, Piques R 2006 Multiaxial fatigue life prediction for a natural rubber International Journal of Fatigue 28: 530-539. [6] Kim J H, Jeong H Y 2005 A study on the material properties and fatigue life of natural rubber with different carbon blacks International Journal of Fatigue 27: 263-272. [7] Woo C S, Kim W D, Kwon J D 2008 A study on the material properties and fatigue life prediction of natural rubber component Materials Science and Engineering: A 483-484: 376381.