Search Results

Acoustical laboratories for vehicle testing have specialized requirements which differ from those for most conventional buildings and facilities. As a result, the normal design and building process takes on added dimensions which need to be carefully considered and addressed. This paper presents an overview of the process that starts with conceptualization of the laboratory and ends with the validation and qualification of the laboratory, and includes particular emphasis on the inherent peculiarities. Case studies are provided of several potential perils and pitfalls that may be encountered in the process which can adversely affect the usability of the laboratory as well as the validity and repeatability of test results obtained by that laboratory. The paper concludes with suggested courses of action which will help either to avoid or minimize the compromises that imperil the functional effectiveness of a laboratory.

There are a number of factors that dictate the size, type, and resultant over all cost of a controlled acoustical environment in which measurements can be made accurately and reliably. The type of acoustical environment is generally specified in the appropriate SAE, ISO, ANSI or ASTM standards. The purpose of this paper is to concentrate upon the design considerations of a properly engineered anechoic chamber. Anechoic is defined as “free from echoes or reverberations”. An ideal chamber would contain no reflections of sound from its walls, ceiling, or floor and an acoustical free-field condition would exist. Probably the best testing environment is outside with no boundaries to cause reflections. However, temperature, pressure, humidity, and wind can significantly and unpredictably disturb the uniform radiation of sound waves. In an ideal free-field environment, the inverse square law would function perfectly.

A frequency domain order tracking method is developed that is able to separate both close and crossing orders. This method is based upon the multiple input H1 FRF estimator. The method can use either the actual tachometer pulse train or a simulated chirp function as an assumed input to formulate the FRFs which are actually order tracks. The advantages and disadvantages of using each type of assumed input are discussed. The performance of this method in both simple and complex order tracking cases is evaluated. Analytical datasets are used to evaluate the performance of these order tracking methods under a variety of operating conditions which include close orders. Finally, this paper will develop the necessary derivation to show the analytical relationship between the Time Variant Discrete Fourier Transform (TVDFT) and the FRF based techniques, one being an order domain method and one being a frequency domain method.

Practical issues to consider when making measurements for Nearfield Acoustical Holography (NAH) analysis are addressed. These include microphone spacing and placement from the test surface, number of microphones and array size, reference microphone number and placement, and filtering of the data. NAH has become an accepted analysis tool so that several commercial packages are available. Its application is limited to test surfaces that are fairly planar, lending itself well to tire testing, front of dash testing, engine face testing, etc. In order to achieve accurate NAH results, the measurement and analysis process must be clearly understood on a practical level. Understanding the advantages and limitations of NAH and the measurement parameters required of it will allow the user to determine if NAH is applicable to a particular test object and environment.

We propose the use of an active element in conjunction with a passive, reasonably-damped Suspension to make an engine mount capable of satisfyingBoth damping and isolation requirements while avoiding 1) the loss of performance due to detuning of tunable (such as hydraulic) engine mounts and 2) the undesirable on/off switching associated with the decoupler in hydraulic mounts. In the proposed scheme, the damping will be provided by the proper choice of damped elastomeric material and isolation will be achieved by controlling the active element using a novel kalman-estimator based algorithm developed by the authors. To demonstrate the effectiveness of the proposed mounting scheme the distributed parameter models of a typical engine/chassis system was developed. Finite element analysis of the chassis was performed to find its natural frequencies and mode shapes which in turn were used to construct the state space control model.

Engine noise is still one of the dominating sources to vehicle interior noise. Among the engine components, engine covers are often the significant contributors to overall engine noise, requiring in-depth acoustic investigation to achieve substantial reduction. Ford Motor Company has acquired a 150 channel Nearfield Acoustic Holography (NAH) system for powertrain NVH development. This system provides new acoustic information with various metrics and visualization of non-stationary sound field in time domain to facilitate better understanding of noise generation/propagation mechanism. This paper focus on investigating the design of engine covers which radiate chain whine, fully utilizing the capability of this system including spatial transformation. Based on reconstruction of noise sources, effective design change to achieve significant reduction of chain whine is derived and then verified in very short time compared to previous methods.

Interior noise in motor vehicles is essentially influenced by the engine which may contribute via both, structural and acoustic transmission paths. This engine related interior noise components may be controlled deliberately by active control measures without changing any source or transfer paths characteristics. Besides attenuating dominant noise components, the approach may equally be used to optimise interior sounds with respect to sound quality. Based on general considerations how active noise and vibration control measures may effect subjective criteria, the paper gives examples how different, sometimes extremely contrasting noise characteristics may be realised in a given car.

This paper deals with the results of an investigation of moulded cases such as acoustic deflectors housing noisy equipment of the vehicle. The main advantage of the deflector is to decrease the level of noise, but the phenomenon of self-excitation is important, too, because it becomes a new source of noise. Readjustment of the vibration modes spectrum is possible by creating special structural variations, which increase the internal dissipation of energy of the vibrations in the deflector material. This approach is related to the introduction of design variables, the kind of material or to the choice of the points of fixation. The principal indication of a vibration mode of such a structure as a deflector is the number of nodal points. This characteristic gives the possibility to select several groups of modes, which are effective from the viewpoint of decreasing vibration amplitudes. Other possibilities are related to the means of active vibroinsulation.

A silencer to attenuate engine exhaust noise using active control methods has been developed. The device consists of an electrically driven valve, combined with a buffer volume, which is connected to the exhaust outlet. Using the mean flow through the valve and the pressure fluctuations in the volume, the valve regulates the flow in such a way that only the mean flow passes through the exhaust outlet. The fluctuations of the flow are temporally buffered in the volume. To carry out optimization and validation experiments, a cold engine simulator has been developed. This device generates realistic exhaust noise as well as the matching gas flow using compressed air. The simulator allows quick and reliable acoustic and fluid dynamic experiments on exhaust prototypes. The silencer is developed using electrical equivalent circuits, wherein at first instance a feedforward control is applied.

This paper describes a project on the dynamic characterization of automotive shock absorbers. The objective was to develop a new testing and analysis methodology for obtaining equivalent linear stiffness and damping of the shock absorbers for use in CAE-NVH low- to- mid frequency chassis models. Previous studies using an elastomer test machine proved unsuitable for testing shocks in the mid-to-high frequency range where the typical road input displacements fall within the noise floor of the elastomer machine. Hence, in this project, an electrodynamic shaker was used for exciting the shock absorbers under displacements less than 0.05 mm up to 500 Hz. Furthermore, instead of the swept sine technique, actual road data were used to excite the shocks. Equivalent linear spring-damper models were developed based on least-squares curve-fitting of the test data.

This paper presents a study on using rubber bushing as a sensor for the identification of forces transmitted onto the car body. The method starts from the idea that the transmission forces can be related to the deformation of the rubber bushing multiplied by its stiffness. Deformation of the rubber bushing is estimated from relative vibrations across the bushing. Simple theories are presented to deal with modeling of the rubber bushing and processing of the vibration mesurements on the link and car body to identify the transmission forces. Then, validity of the proposed approach is shown by applications to a suspension system under several driving conditions.

This paper reviews the philosophy of large-data-acquisition-system design as it has evolved over the past 20 years and discusses recent developments in applicable hardware and software tools. PIRANHA, a 320-channel system that embodies many of the features discussed and takes advantage of the new technologies, is described.

Five examples of HMIs, Human Machine Interfaces, (formerly Man-Machine-Interfaces) are presented to illustrate the variety of operating solutions for computer-aided dynamic data acquisition systems. The examples range from a simple on-off control with a simple terminal -type interface for a sophisticated instrumentation recorder to a 384-channel automated Wind Tunnel test system. The design challenges, the evolution from GUI’s to HMI’s, the potential pitfalls and benefits of software-based HMI’s and a means to speed development and to get timely end-user input are covered.

Because of the recent trend for testing products in multi- directions and to higher and higher frequencies there is a need to develop excitation systems with high natural frequencies, i.e. having a stiff construction with small moving mass(es). In this work, a 3- degree of freedom (DOF) excitation mechanism with closed-chain kinematics is proposed that can satisfactorily fulfill these requirements. Closed loop control allows this device to be used for testing parts under realistic combined loads that include not only vertical but also two axial horizontal loads. The 6-degree of freedom version of such device can, in addition to 3 axial forces, also load a part with 3 moments. The use of this device can be extended to active vibration control applications such as active seats for off-road vehicles.

A significant challenge in successful implementation of large channel count dynamic test systems, like multi-channel modal systems, has been efficient and error free test set-up and data collection. Newly developed instrumentation that digitally communicates self-identifying information (including transducer type and serial number, calibration value and physical location) greatly enhances system performance and drastically reduces the opportunity for human documentation errors. Sensors featuring this Transducer Electronic Data Sheet (TEDS, consistent with the impending IEEE P1451.4 standard), in addition to such techniques as sonic digitization for geometry definition and automated data acquisition with computer controlled, bank switching signal conditioners are continually improving accuracy and reducing the per channel price of multi-channel dynamic test systems.

The objective of the present work was to establish a correlation between steering pump cam ring profile location and steering system performance for noise, vibration and harshness (NVH). Once this correlation was established, the secondary objective was to determine acceptable cam profile position tolerances from the standpoint of NVH performance. These objectives were accomplished through the use of binaural measurement and jury evaluation of vehicle interior noise. Cam rings were manufactured for this study with profiles shifted a predetermined distance away from the nominal position. These cams were built into steering pumps and these pumps were in turn installed in a vehicle. Vehicle interior noise and pump housing vibration measurements were made to quantify the steering system noise performance associated with each cam ring. The interior noise recordings were played back for a jury comprised of engineers familiar with steering system noise.

In order to optimise the vibro-acoustic behaviour of panel-like structures in a more systematic way, accurate structural models are needed. However, at the frequencies of relevance to the vibro-acoustic problem, the mode shapes are very complex, requiring a high spatial resolution in the measurement procedure. The large number of required transducers and their mass loading effects limit the applicability of accelerometer testing. In recent years, optical measuring methods have been proposed. Direct electronic (ESPI) imaging, using strobed continuous laser illumination, or more recently, pulsed laser illumination, have lately created the possibility to bring the holographic testing approach to the level of industrial applicability for modal analysis procedures. The present paper discusses the various critical elements of a holographic ESPI modal testing system.

The Percentile Frequency method originated in an attempt to quantify the frequency content of door slam sounds. The method is based on the Specific Loudness Patterns of Zwicker Loudness. Zwicker states that the area of the Specific Loudness Pattern is proportional to the total loudness. The method summarizes each Pattern as seven frequencies identifying the contributions of fixed percentages of the total area (i.e. 10%, 20%, 30%, 50%, 70%, 80% and 90%). Applying the method to each Pattern in a time series generates a family of curves representing the change in relative frequency content with time. The process, in effect, normalizes the frequency content of the impulse for loudness and reduces the data to a two dimensional plot. On a Percentile Frequency plot a simple impulse appears as a pattern of “nested, inverted check marks.” More complicated impulses, such as rattles, have more complicated shapes that are still “nested” together.

This paper focuses on psychoacoustical experiments for the assessment of roughness by using vehicle interior noise. The experimental design is carried out carefully to derive reliable data for further analysis with objective parameters. Apart from the acoustical properties of the recording/playback system the different meanings of the word roughness are taken into account, because each person has its own interpretation of ‘roughness’ differing between the phenomenons of roughness, r-roughness, rumble, harshness, fluctuation strength, etc.. An important preparation for psychoacoustical experiments is a clear definition of the sound attribute under investigation by using typical examples. Furthermore, accidental influences of other psychoacoustical parameters like the influence of loudness have to be avoided.

In this paper an algorithm is developed for combining finite element analysis and boundary element techniques in order to compute the noise radiated from a panel subjected to boundary layer loading. The excitation is presented in terms of the auto and cross power spectral densities of the fluctuating wall pressure. The structural finite element model for the panel is divided into a number of sub-panels. A uniform fluctuating pressure is applied as excitation on each sub-panel separately. The corresponding vibration is computed, and is utilized as excitation for an acoustic boundary element analysis. The acoustic response is computed at any data recovery point of interest. The relationships between the acoustic response and the pressure excitation applied at each particular sub-panel constitute a set of transfer functions.