Search Results

This paper contributes to the Committee on Commonized Aerodynamics Automotive Testing Standards (CAATS) initiative, established by the late Gary Elfstrom. It is collaboratively compiled by automotive wind tunnel users and operators within the Subsonic Aerodynamic Testing Association (SATA). Its specific focus lies in automotive wind tunnel test techniques, encompassing both those relevant to passenger car and race car development. It is part of the comprehensive CAATS series, which addresses not only test techniques but also wind tunnel calibration, uncertainty analysis, and wind tunnel correction methods. The core objective of this paper is to furnish comprehensive guidelines for wind tunnel testing and associated techniques. It begins by elucidating the initial wind tunnel setup and vehicle arrangement within it.

The development of the WorldSID 50th percentile male dummy was initiated in 1997 by the International Organization for Standardization (ISO/SC12/TC22/WG5) with the objective of developing a more biofidelic side impact dummy and supporting the adoption of a harmonized dummy into regulations. More than 45 organizations from all around the world have contributed to this effort including governmental agencies, research institutes, car manufacturers and dummy manufacturers. The first production version of the WorldSID 50th male dummy was released in March 2004 and demonstrated an improved biofidelity over existing side impact dummies. Full-scale vehicle tests covering a wide range of side impact test procedures were performed worldwide with the WorldSID dummy. However, the vehicle safety performance could not be assessed due to lack of injury risk curves for this dummy. The development of these curves was initiated in 2004 within the framework of ISO/SC12/TC22/WG6 (Injury criteria).

Efforts by vehicle manufacturers to reduce road testing have resulted in an increased reliance on the simulation methods for loads measurement and validation, including increased emphasis on methods to characterize and digitally represent test road inputs. Accurate terrain models are especially important in the case of large dynamic road inputs, and for evaluation of vehicle suspension loads and durability. In contrast to direct terrain topology measurement, methods to estimate test road input using only vehicle suspension measurements and a tire dynamic model will be presented. Applications of terrain models for generic simulation and testing will also be discussed.

Competitive pressures, globalization of markets, and integration of new materials and technologies into heavy vehicle suspension systems have increased demand for durability validation of new designs. Traditional Proving Ground and on-road testing for suspension development have the limitations of extremely long test times, poor repeatability and the corresponding difficultly in getting good engineering level data on failures. This test approach requires a complete vehicle driven continuously over severe Proving Ground events for extended periods. Such tests are not only time consuming but also costly in terms of equipment, maintenance, personnel, and fuel. Ideally multiple samples must be tested to accumulate equivalent millions of kilometers of operation in highly damaging environments.

Shudder vibration of a hydraulic power steering system during parking maneuver was studied with numerical and experimental methods. To quantify vibration performance of the system and recognize important stimuli for drivers, a shudder metric was derived by correlation between objective measurements and subjective ratings. A CAE model for steering wheel vibration analysis was developed and compared with measured data. In order to describe steering input dependency of shudder, a new dynamic friction modeling method, in which the magnitude of effective damping is determined by average velocity, was proposed. The developed model was validated using the measured steering wheel acceleration and the pressure change at inlet of the steering gear box. It was shown that the developed model successfully describes major modes by comparing the calculated FRF of the hydraulic system with measured one from the hydraulic excitation test.

The intention of this work is to illustrate a method used to overcome limitations of tire models developed during an evaluation study of an Empirical Dynamic™ (ED) damper model. A quarter vehicle test system was built to support the evaluation, and a model of the test system was also developed in ADAMS™. In the model, the damper was represented by a polynomial spline function and by an ED model separately. Vehicle level comparisons between the physical measurements and the model predictions were conducted. The actuator displacement signal from the physical test was used to drive the virtual test system. Spindle acceleration, spindle force, and other signals were collected for comparison. The tire model was identified as a significant source of error and as a result, the direct vehicle level correlation study did not illustrate any advantage of the ED damper model over a spline damper model.

This paper describes an approach to simulate spindle coupled full vehicle durability tests for the purpose of completing virtual durability evaluations on components and full vehicles before a prototype is available. The reproduction of measured spindle loads was achieved on a virtual model of a passenger car coupled to a 4 Degree of Freedom (DOF) and 6 DOF spindle coupled test system. The tools and process improvements developed here will aid both test and analysis engineers in working closer together in solving their durability problems. By using Remote Parameter Control® (RPC®) technology in the virtual world, analysts have a new method to understand the virtual model by reproducing field-measured or generic road predicted signals for a variety of road surfaces. With newly created test rig models and a user friendly RPC™ iteration process, virtual testing that accurately replicates laboratory tests are now a reality.

The quest to simulate noise problems has led to the building of larger and more detailed finite element models in order to perform vibration solutions to higher frequencies. This leads to the building of solid finite element models of complex geometries, such as castings, which might previously have contained less detail or even been built with shell elements. Unfortunately, detailed geometric representations used to build models do not always agree with as built parts and lead to discrepancies between analysis results and test data. This paper presents an approach that reduces the time and cost necessary to identify these differences.

The role of NVH testing has evolved from a firefighting role and a period of exploration to a well defined standard test role in the product development and validation process. Integral to this process is robust engineering, which drives the need to execute many tests quickly, efficiently and accurately. This allows the NVH specialist to concentrate on interpretation of results and spend less time on the acquisition of data. As the volume of data grows, this creates the opportunity to data mine an NVH database to compare results from large sample sizes and focus on product variation. Today's NVH laboratory is accountable for producing high quality, consistent, timely, and cost effective test reports. The basic core of the test has to be easy to set up and execute for a novice, yet still allow for exploratory tests by specialists as necessary. The NVH laboratory is now subject to the same budgetary pressures and quality audits as other testing operations.

Specimen grips in an axial load frame typically have a small misalignment that imposes bending strain on the clamped specimen. The bending strain causes variability in the material test results, especially in fatigue testing of brittle materials. This paper introduces new techniques for aiding load frame alignment. Examining the source of the bending strain identifies how much of the bending strain is due to the specimen imperfections versus the machine misalignment. Quantifying the misalignment components provides criteria for automating the setscrew adjustment selection of the alignment process.

The essential concepts for developing a moving deformable barrier that may serve as a vehicle surrogate in assessing vehicle compatibility are described in this paper. Although moving deformable barriers have been used for assessing other safety criteria, their purpose in those cases is to reproduce a limited set of responses in the struck vehicle. An MDB for vehicle compatibility however, needs to be able to reproduce the responses of both the vehicles. The present study describes the concept of developing such barriers by generating ‘response corridors’ for the significant variables by nonlinear finite element simulations and then selecting design parameters such that the MDB response is within this corridor. It is observed that the response of the equivalent MDB representing a light truck vehicle is reproducible when response corridors are utilized.

This paper documents the process used to correlate the secondary belt degradation experienced on the test track with the secondary belt degradation experienced during laboratory tests using a Secondary Drive Test System. Two different software products were used to produce this correlation: nCode's pseudo-damage functionality was used to estimate the proportional belt degradation and MTS's RPC Pro functionality was used to edit the field data, create a time history file, and to shift the frequency domain of the vehicle into the usable range of the servo-hydraulic actuator (time stretching). For purposes of this paper, the test data and information presented in this paper is based on two different secondary drive belts that were used on the test track as well as in the laboratory tests. As will be shown, the plot information that resulted from these tests showed very good correlation.

To cope with the conflict requirements between the stability and handling performance, and the high-order and complex vehicle-trailer plant, a model tracking method is proposed. With this approach, a feedback control is designed to “decouple” the vehicle and the trailer plant, such that each tracks a well-defined second-order reference model independently yet coordinately. A feedforward control is designed to maintain its system steady-state performance. As a result, the proposed approach not only improves the system transient responses, but also its steady-state performance. This approach further yields a simple yet analytical control derivation that provides more insight to the system dynamics.

As a result of the severity of occupant injuries during a side impact collision, there has been an escalating demand for accurate component level side impact simulation. Three major components for accurate simulation are accurate door velocity, door to seat relative velocity, and door deformation. This paper shows data demonstrating accurate door velocity reproduction, presents test methods to passively and actively control relative seat to door velocity in a non destructive manner, and presents test methods to simulate real time door deformation in a destructive manner. All side impact waveforms include a negative acceleration, high positive accelerations, high jerk, and high frequency content that add to the complexity of this simulation. The simulated door velocity is produced by means of a MTS deceleration brake that only applies a braking force during the deceleration portion of the waveform to maximize acceleration capacity.

Recent studies have shown that complex vehicle components such as shock absorbers, rubber bushings, and engine mounts can be accurately modeled by combining laboratory measurements with neural network technology. These nonlinear dynamic blackbox models (also known as Empirical Dynamics1 models) make it possible to predict nonlinear and hysteretic component behavior over wide ranges of amplitude and frequency. The models can handle realistic input waveforms as well as multiple inputs and multiple outputs. These techniques have now been applied to rolling pneumatic tires, to enable high accuracy predictions of tire and vehicle handling behavior. Models that predict high amplitude force components (three forces and three moments) using up to four randomly-varying inputs (radial deflection, slip angle, and camber angle, and slip ratio) have been successfully generated, using data obtained from MTS Flat-Trac III tire test equipment.

Hyundai Motor Company has combined physical and virtual testing tools to validate a full vehicle virtual prototype. Today a large number of physical tests are still required because the cycle of “design-build-test-change” relies on complex models of components and systems that typically are not easily validated. In order to shorten the development cycles, engineers perform multi-body simulations to dynamically excite components and systems and thereby estimate their durability under dynamic loads. The approach described herein demonstrates the feasibility of correlating the output from the corresponding physical and virtual prototype. Both synthetic and road load events are employed to excite physical and virtual vehicles, reveal difference in response, and ultimately improve the predictive capability of the model.

The sound field in a vehicle is one of the most complex environments being a mixture of multiple, correlated and uncorrelated sound sources. The simulation of vehicle interior sound has traditionally been produced by combining multiple test results where the influence of one source is enhanced while the other sources are suppressed, such as towing the vehicle on a rough surface for road noise, or measuring noise in a wind tunnel. Such methods are costly and provide inherent inaccuracies due to source contamination and lack of synchronization between sources. In addition they preclude the addition of analytical predictions into the simulation. The authors propose an alternative approach in which the component sounds are decomposed or separated from a single operating measurement and which provide the basis for accurate sound synthesis.

The quietness of the interior of automobiles is perceived by consumers as a measure of quality and luxury. Great strides have been achieved in isolating interiors from noise sources. As noise is reduced, in particular wind and power train noise, other noise sources become evident. Noise reduction efforts are now focused on components like power steering pumps. To understand the contribution of power steering pumps a world-class noise and vibration test stand was developed. This paper describes the development of the test stand as well as it's objective to understand and improve the sound quality of power steering pumps.

Brake groan noise and vibration occurs in a stopped vehicle by the simultaneous application of torque to the wheel and the gradual release of brake pressure. Eventually the torque load breaks the friction between pad and rotor causing slippage and energy release. If the torque load is not large enough to maintain slippage, a sustained stick-slip vibration, called groan, can occur which transmits a low frequency noise to the vehicle interior. In some cases the noise levels caused by groan can be objectionable, thus procedures for developing remedial designs are needed. To this end, a project was performed to analytically simulate groan vibration in a vehicle with a McPherson strut type suspension. The goal was to demonstrate that analytical models could be used to simulate groan behavior and to identify suspension components that affect the groan behavior. The ADAMS software was used to model a brake/suspension system.

Many structures of practical interest exhibit a significant degree of nonlinearity. In such cases, the modal frequencies, damping, and amplitudes will change depending upon the excitation force level, response level and spectrum shape. When reporting the measured modal parameters from an artificial excitation test, the excitation conditions and response levels should be specified, and different modal models may be needed to represent the structural dynamics at different response amplitude levels. If the frequency responses are measured by moving accelerometers in multiple test runs, then it is important to maintain a consistent response level for all test runs. This paper describes a method to eliminate the variability of the response level between data sets by means of closed-loop control of the RMS level. The amplitude control program uses a nonlinear gain estimation technique to set the gain on a “proportional-integral” controller.