Search Results

The object of this paper is to develop a random vibration laboratory test specification for automotive subsystems using the Power Spectral Density (PSD) method. This development is based on the 150k mile field data collected from vehicle proving grounds. The simulated vibration bench test will be used to simulate the energy of the 150k mile field data. The developed specification will include 3 axis random vibration profiles of appropriate duration. The Power Spectral Density method converts the time-domain field data into the frequency-domain data. The Enveloped Energy method groups the similar road PSD profiles to produce a generic PSD profile. The Inverse Law allocates an adjusted duration to the desired PSD energy level. The Road Test Specification provides the duration time for the developed bench test. The n-Soft tool [1] is utilized for data reduction analysis. The Bench Test Specification of the Fuel subsystem is a pilot for this development.

Global acoustic sensitivity analysis [1] is a technique used to identify structural modifications to a component that can reduce the total radiated power of a vibrating structure or the sound pressure levels at specified field points. This report describes the use of global sensitivity analysis within SYSNOISE to determine what structural changes are required to reduce radiated noise from flexible structures in an open duct system. The technique can help optimize design parameters that define the behavior of a flexible structure such as shell thickness and Young's Modulus. The sensitivity analysis approach consists of separately evaluating structural and acoustic sensitivities. A structural finite element model (FEM) of an open duct system is used to compute the sensitivity of the structural response to changes in thickness. A boundary element model (BEM) is then used to relate changes in the calculated acoustic response to changes in the structural design variables.

A simple CAE method of predicting the performance of a component during sine testing has been developed and applied to the practical case of an automotive component. The slow frequency sweep rate during a test is represented as a sequence of steady state conditions. Direct frequency response analysis at the limited number of frequencies is conducted and results used as a basis for prediction of fatigue damage using the Palmgren-Miner rule. The total damage during the test is calculated by linear summation of the damage during each frequency interval. This technique is completely general and can be applied even if there are multiple inputs to the component. A simple extension enables application to engine testing and other cases where excitation may be expressed as a Fourier series expansion of periodic excitations.

This paper describes a CAE methodology for optimizing the performance of induction system resonators on the basis of their location within the standing wave modes of the system. An acoustic modal analysis utility of NASTRAN [2], and a duct system acoustics model, LAMPS [1], are used interactively to determine resonator location and resonator size requirement. The NASTRAN code predicts the standing wave mode shapes for the resonant frequencies of the induction system which includes the air inlet system and the intake manifold. Resonator location is determined on the basis of the pressure anti-nodal points of the mode shapes. Maximum attenuation is achieved at or near the anti-nodal points and minimal attenuation at the nodal points. The LAMPS code is used to predict the acoustic response characteristics of the system with and without resonators. Induction noise measurements made on an operating engine agree fairly well with the computer predicted results.