21-06-2013, 04:07 PM
FINITE ELEMENT ANALYSIS OF WINDSCREEN WIPER CHATTER NOISE AND ITS SUPPRESSION APPROACH THROUGH STRUCTURAL MODIFICATIONS
Abstract
This paper presents an approach to predict automotive windscreen wiper chatter noise using finite element method. In
this study, a 3-dimensional finite element model of a real wiper blade is developed and then validated using
experimental modal analysis. In order to assess stability of the wiper blade assembly, complex eigenvalue analysis
that made available in ABAQUS is performed. The positive real parts of complex eigenvalue indicate unstable
system. The baseline model is first simulated and stability of the system is examined. Having found a predicted
unstable frequency which is chatter noise, various structural modifications are proposed in order to reduce it.
Introduction
Windscreen wipers are indispensable components to the maintenance of a safe and comfortable field of vision
when driving on rainy days. Today, almost all automobile are equipped with windscreen wiper, often by legal
requirement. Clear vision for the car driver is an important prerequisite for safety in road traffic. A conventional wiper
as shown in Figure 1 generally consists of an arm, pivoting at one end and with a long rubber blade attached to the
other [1]. The blade is swung back and forth over the windscreen, pushing water from its surface. The mechanical
structure of the wiper blades is attached to the arm tips, holds the rubber blade, which drains the water off the
windscreen or to smooth the water on the surface of the windscreen in order to create a thin film that allows light to
pass through it without refracting or bending as shown in Figure 2.
Experimental approaches
This section describes in details experimental approaches to study vibration of the windscreen wiper. The first
experimental approach is using experimental modal analysis to validate the developed 3 dimensional FE model in
terms of natural frequency and its associated mode shape. The second experiment is carried out to study vibration of
the wiper in the wet condition.
Experimental Modal Analysis (EMA)
EMA is performed on the individual wiper components in order to determine its dynamic characteristics [9]. The
EMA is performed at free-free boundary condition for the rubber blade, primary lever and yoke and secondary lever,
whilst at fixed boundary condition for the windscreen. The impact hammer test method is used to obtain natural
frequencies of those components. In doing so, a Kistler type 9722A500 impact hammer is used to produce the
excitation force while a Kistler Type 8636C50 uni-axial accelerometer is fix-mounted onto the tested components [10]
as shown in Figure 3.
Modeling of a Wiper Blade
A detailed 3dimensonal FE model of a passenger car windscreen wiper assembly is developed as shown in Figure
5. The FE model consists of a windscreen, a rubber blade, primary yoke and a wiper arm as shown in Figure 6. The
windscreen is simplified as a flat surface in the FE model in order to reduce computational time. The assumption of a
flat surface is made based on the previous studies [6]. The windscreen and vertebrata are developed using 8-node
(C3D8) linear solid elements while the rubber blade are developed using combination of 4-node (C3D4) and 8-node
(C3D8) linear solid elements. The levers (primary, second and third) are developed using 4-node (C3D4) linear solid
element while the wiper arm are developed using combination of 4-node (C3D4), 6-node (C3D6) and 8-node (C3D8)
linear solid elements. Element types in brackets show the notation in ABAQUS nomenclature. Details for each of the
components are given in Table 1.
Conclusion
This paper presents a numerical analysis to predict wiper blade chatter noise using FE method. Complex
eigenvalue analysis that made available in commercial software package, ABAQUS is used to predict chatter
frequencies. From the experiment, the result shows that the windscreen wiper produces a low frequency noise called
chatter, at dominant frequency of 11Hz. It is found that, at different wiper average speeds, the chattering vibration is
generated before and after the wiper turnover. The complex eigenvalue analysis has been utilized in a finite element
analysis to study chatter problem. The measured chatter frequency has been successfully replicated in the analysis.
Initially, there are eleven unstable frequencies (chatter) appeared in the baseline model. Based on the baseline results,
various structural modifications have been proposed and simulated using complex eigenvalue analysis in order to
suppress the chatter frequency.