Metodologia de avaliação da performance acústica experimental e virtual do sistema de exaustão automotivo

Abstract

Acoustic noise generated by vehicles is one of the main noise pollution factors in large cities. Among the systems applied to the automobile, the exhaust system is the major contributing factor to the generation of this noise, so its attenuation must be very efficient. The acoustic attenuation of the exhaust system is performed by reactive type passive silencers that generate wave reflection due to the impedance caused by the geometric variation and dissipative ones that generate absorption of the acoustic wave transforming the acoustic energy into heat. Performance evaluation of silencers is performed in two steps. The first through the transmission loss bank. The second step is the evaluation of the silencer from the application to the exhaust system that is connected to the engine in order to evaluate the noise emitted throughout its operating range. This paper aims to present a methodology for evaluating the experimental and virtual acoustic performance of the automotive exhaust system considering its analysis in the transmission loss bank and application to the vehicle. In order to predict the acoustic behavior in both cases during the design phase, the one-dimensional virtual simulation software GT-Suite was used. Experiments were performed using silencers applied to the transmission loss bank to validate the virtual model, which showed good correlation between the 250 Hz and 1000 Hz ranges. The virtual simulation of the application of the exhaust to the vehicle takes into account the engine model and the air intake system. The engine is modeled in steady state, which generates the correlation problem of the acoustic exhaust model, since the experimental validation is done in state transient. With this information, this work also aims to evaluate the acoustic effects of the exhaust system when it is applied to the transient and permanent model experimentally and, finally, compared to the virtual permanent state model. The use of the roller dynamometer is necessary to perform the test in the permanent condition, which is stabilized in a fixed rotation with each experiment. Another condition evaluated on the dynamometer is run-up, where the acoustic condition is monitored at each transient rotation. In addition to evaluating the conditions applied to the dynamometer, the run-up test is compared to the simulated model in order to evaluate the different acoustic impacts on the different measurement forms. The dynamometer experiments did not show great differences in the permanent and transient mode. Thus, it can be concluded that the evaluation performed under run-up conditions in the roll and constant rotation presented small variations, both in the general level and in the evaluated orders. Thus, the validation of the model from the transient method does not present satisfactory variation when evaluated by the permanent method.