Modeling of a Gear Test Rig : An Investigation of Static Loads and Dynamic Excitation of Vibrations

Sammanfattning: Today the automotive industry is going through big changes and facing many new challenges due to the transformation to a sustainable transport solution in combination with tougher legal demands. Gears have been around for a long time and it is one of the most efficient ways of transferring rotary motion from one shaft to another. Transmissions is a key factor in order to transfer the rotary motion from the engine to the wheels. Both traditional combustion engines and electric powered vehicles needs a transmission with gears. The continuously increasing demands on efficiency, noise and durability requires increased knowledge about gears, especially about gear dynamics and how vibrations are excited and transferred to the gearbox for different mesh frequencies. In this report, a theoretical background about gears is presented as well as the mechanisms behind the excitation of vibrations. The goal with this master thesis has been to create a validated model of a gear test rig. For the static validation, two types of gears have been used in the test gearbox. The gears in question are FZG standardized type-A gears and type-C gears. The model includes gears, shafts, bearings and housings. This model has been used for simulating static torque fluctuations as well as dynamic excitation of vibrations that is transferred to the housings, both at constant speed and with continuously increasing speed. In the dynamic analysis, only type-C gears have been used. To validate the model, vibrations have been measured on a FZG gear test rig using accelerometers and tachometer. For the static validation, torque has been measured while running the test rig at 5 rpm. The results show that it is possible to get the load clutch in the model to behave as in the test rig. It also shows that the model can register the static torque fluctuations similar to the fluctuations in the test rig. The type-C gears are better suited for the simulations and gives a better result than the type-A gears. For the type-A gears there are some numerical problems related to tip contact during meshing. The investigation of the tooth contact pressure pattern shows a good correlation between the simulations and the used type-C gear. The patterns have the same shape on the tooth flank which indicates that the contact between the teeth behave similar in the model and the test rig. For the validation of the dynamic behaviour, the gear mesh overtones have been investigated, both at constant velocity and with continuously increasing velocity. For the simulations, there are some issues related to the FEprobes placed on the gearboxes so instead the data from the inner bearings in the model have been used for the validation. The analysis of the dynamic simulations shows that it is possible to identify the tooth mesh overtones but the resonance peaks are less amplified compared to the measurements from the test rig. For the simulations with constant velocity the overtone trend correlates well with the experimental data at high speeds but at lower speed there is an amplitude peak for the firstovertone that doesn’t correlates with the measurements. As a conclusion, the model has beenstatically validated with good results while for the dynamic validation, there is still some aspects that need to be improved in order to get a good correlation between the simulations and the measurements from the test rig. Improvements suggested is to run the simulation with continuously increasing speed using more time steps in order to get more data points for the rpm spectra. It is also suggested to perform an impact hammer modal testing on the test rig in order to get a better understanding of the damping in the system.