Search Results

Statistical Energy Analysis is an accepted method for NVH model development at high frequencies. This paper explores hybrid methods to expand SEA to the mid and low frequency range where the assumption of high modal density is not valid. The newly developed hybrid method is particularly useful for structure-borne noise and vibration studies. The method is based on use of mobility functions to improve calculation of power input, modal densities and coupling factors. Comparison of the hybrid model predictions with measured data show good agreement down to 100 Hz.

Statistical energy analysis models are often used to predict vehicle noise. These models are generally successful at high frequencies, above 500 Hz, where transmission of airborne noise from the vehicle exterior to the interior is the dominant source of noise. At mid-and low frequencies the noise transmitted by structureborne paths becomes more important. SEA models can be extended to study both airborne and structureborne noise transmission in this frequency range. The modeling is more complex, however, because of the variety of structural wave types and the spatial irregularity of structural parameters. This paper presents the techniques required to develop SEA models for predicting structureborne noise in vehicles. Particular attention is given to the calculation of modal densities and coupling factors in the mid-frequency range from 100 to 500 Hz. Attention is also given to the calculation of the statistical variance of the SEA prediction.

Airborne sound transmission through vehicle panels with penetrations and sound insulation is a major component of high frequency interior noise in cars and trucks. Accurate analytical models of interior noise require high fidelity simulation of these paths in order to perform upfront design of the sound package. This paper describes a modeling approach based on Statistical Energy Analysis (SEA) that provides a general and flexible capability for incorporating sound package parameters within an analytical model of high frequency interior noise. Validation of the model for sound transmission through panels with holes and with typical sound insulation material is achieved through innovative testing methods that reveal dynamics of the decoupler and barrier layers. Refinements of the general approach that consider more deterministic features of the specific decoupler material are also suggested.

In order to effectively use CAE to meet wind noise NVH targets, it is important to understand the main wind noise transfer paths. Testing confirmation of these paths by means of acoustic wind tunnel test is expensive and not always available. An on-road test procedure including a “windowing” method (using barriers) was developed to measure wind noise contribution at important higher frequencies through the main transfer paths, which were shown by test to be the glasses at a typical operating condition in which wind noise was dominant. The test data was used to correlate a full-vehicle SEA (Statistical Energy Analysis) model that placed emphasis on the glass properties, main noise transfer paths, and interior acoustic spaces while simplifying all other transmission paths. A method for generating wind noise loads was developed using measured glass vibration data, exterior pressure data, and interior acoustic data.

Statistical energy analysis has become an accepted method to predict noise, vibration and harshness in motor vehicles. SEA provides a statistical estimate of the vehicle response. In most cases, the mean response prediction is used to evaluate different designs. It has generally been found that SEA predictions of the change in mean acoustic response due to a specific design change are in good agreement with the change in the actual measured results. However, there may be significant differences between the absolute value of the predicted mean and the value observed from measurement. These differences are associated with the statistical variability of the prediction. This paper explores the use of the variance of the SEA prediction. Determination of confidence intervals for absolute prediction accuracy is described. In addition, the variance associated with SEA prediction of design changes is discussed.

Vibration and noise generated by the mesh of gears rotating under load are complex phenomena that are difficult to analyze. Recent advances in digital signal processing make it possible to use measured vibration signals on gear bearings to construct the time history of the excitation of the gear mesh. Also, recent advances in numerical models of gears make it possible to predict the load distribution on the teeth in the mesh, and the associated vibration excitation. Combining these two techniques provides useful input to the design of quiet gears, and the monitoring of gear vibrations to detect failures.