Modelling and Simulation of Gearset Behavior: Effects of Manufacturing Errors and Electrification

Due to electrification, new demands are imposed on gears regarding e.g., quality, noise, lubrication, and gear ratio.
Therefore, a better understanding of gearset meshing and operation is needed. Experiments are valuable tools, but
also tend to be cumbersome and time consuming. Moreover, contact pressure cannot be measured satisfactorily.
Thus, a simulation tool is needed. In this thesis, an LTCA simulation tool is developed, which simulates meshing
of gears with manufacturing errors (ME). Contact is found from common normal directions together with a
compliance condition, instead of using a predetermined load distribution. The LTCA accounts for non-Hertzian
pressure, contact outside the nominal line of action (LOA), and tip contact. Contact pressure of gearsets with
different combinations of ME tolerances is simulated, i.e., relating deviations in geometry caused by manufacturing
to performance of the gearset.
The transmission of motion should also be smooth. This is quantified by the transmission error (TE). TE is
simulated together with contact pressure. Since they tend to counter-vary, optimization is performed. Design
curves are presented to show how to choose tolerances to simultaneously optimize TE and contact pressure.
It is concluded that ME adversely impact gearsets. Too tight tolerances, however, increase the risk of unjustified
scrapping, which increases production time and cost, material waste, and environmental impact. To avoid this, it is
suggested to use tip contact threshold torque as a single metric to assess ME, i.e., basing assessment on performance
instead of only geometry. The feasibility of the method is shown in a case study, where some scrapping is shown
to be unjustified.
To account for lubrication, a thermal elasto-hydrodynamic lubrication (TEHL) method is developed. It finds
lubricant pressure and temperature by solving the Reynolds equation and the heat equation, and considers varying
viscosity and density, as well as cavitation. Load distribution found from the LTCA is used as input to include
ME. Results show that ME also cause an increase in temperature and a decrease in film thickness.
Apart from the single gear pairs, a study of a two-stage gear reduction used in an electric vehicle is presented.
Two stages open further possibilities for optimization. This is done by finding the optimum dog leg angle by
means of the lowest reaction forces, which in turn allows for smaller bearings. The decreased enclosed volume of
the gearset also results in a lower housing mass. Both propulsion and regenerative braking are accounted for