Search Results

In the previous paper(1), the authors reported the calculation method they developed for predicting the idling rattle in manual transmissions. This method provides data that represent noise levels to which human ear is not sensitive by numerical values. In the study described in this paper, the authors attempted to produce audible noise through a speaker by the following process: create time-series data of fluctuation in the angular acceleration obtained by the calculation (which is considered to correspond to rattle noise); create next-stage data by applying convolution of a transmission case's vibration transfer characteristics filter obtained by the experiment to the above-mentioned time-series data; convert the filtered data into a wave file; and then input the file to a personal computer to obtain audible sound as output. The audible noise thus produced provides a means of evaluating the level and nature of noise in the way humans naturally experience it.

It is generally known that the idling rattle in manual transmissions is caused by gear tooth portions that make repeated impact-generating vibrations in the backlashes. These vibrations result from rotational fluctuations of the flywheel induced by combustion in the engine. In the study reported here, the authors constructed an experimental setup using rotary encoders and a transient torsional angle converter that allowed the long-awaited direct measurement of impact-generating vibrations in the backlashes. Using this experimental result, the following ideas that the authors must pay attention for the numerical simulation are obtained. That is, transmission drag torque is to be input and treated as the offset value in the torque value of the torsional characteristics in the clutch disc, and coefficients of attenuation have great influence upon the calculation result.

To reduce the idling rattle of manual transmissions, the computer simulation has been utilized. However, the conventional simulation model could not express properly the relationship between the transmission oil temperature and the rattle noise level, especially in case of transmission with backlash eliminator in constant mesh gears. In this study, the authors carried out detail experiments investigating the motion of each part in the transmission. Based on the experimental results, an additional mass representing all constant mesh speed gears supported on plain or rolling element bearings was introduced to the simulation model. Using the improved model, it was confirmed that the calculated RMS value of the fluctuation in countershaft angular acceleration corresponds to the experimental rattle noise level.