Knock is known to cause acoustical cavity resonances in the combustion chambers of engines. Understanding this phenomenon is important for control of engine noise and optimization of knock-detection systems. Cavity resonances were investigated in detail for six open-chamber diesel engines of different sizes. Spectral data were obtained from cylinder pressure-time traces and compared with predictions from finite-element calculations of the cavity resonances. Good agreement was found.

Cavity resonances in engine cylinders are caused by combustion events such as the rapid rate of pressure rise that occurs during compression ignition in diesels or from knock in gasoline engines. These resonances generally occur at frequencies greater than 4 to 5 kHz where the engine structure is not an efficient acoustical radiator. However, when they occur at lower frequencies such as in engines with a large bore or in indirect injection diesels, they can be important in the noise generation process. They are also important for knock detection in gasoline engines. Current knock detection systems are tuned to the frequency band of the lowest cavity resonance in the combustion chamber. It is shown in the paper that higher order resonances can also be detected by a knock vibration sensor on the surface of the engine. Another use for the cavity resonances is to determine the bulk temperature of the gas in the combustion chamber as a function of crank angle. This technique is demonstrated in the paper for a heavy-duty two-stroke diesel. Also, the results of several fundamental investigations of cavity resonances in engine combustion chambers are reported briefly. Good agreement is obtained between theoretical prediction of the resonant frequencies and experimental observation. The splitting of degenerate modes into two components is discussed.

No abstract

A structural-acoustic finite element model of an automotive vehicle is developed and experimentally evaluated for predicting the structural-borne interior noise in the passenger compartment when the vehicle travels over a randomly rough road at a constant speed. The structural-acoustic model couples a structural finite element model of the vehicle with an acoustic finite element model of the passenger compartment. Measured random road profile data provides the prescribed power spectral density excitation applied at the tire-patch contact points to predict the structural-borne interior road noise. Comparisons of the predicted and measured interior noise for laboratory shaker excitation, tire patch excitation, and vehicle travel over a randomly rough road are used to assess the accuracy of the model.

No abstract