Transmission Error as Gear Noise Excitation

Traditionally, transmission error (TE) has been used in order to asses the noise properties of gears. Measurements of gear noise for a complete truck gearbox have been used to correlate noise from a gear pair with the concept of calculated static transmission error as noise excitation. Two gear pairs with very similar macro geometry but different micro geometry was used. Both transmission error as excitation and the excitation proposed by P. Velex and M. Ajmi which is the difference between the loaded and unloaded transmission error, are compared with measured noise. The result shows that the difference between the loaded and unloaded TE correlates well with measure noise for gear pair A but no excitation correlates with the measurement result gear pair B. A big difference between gear pair A and B can be seen in the contact pattern. The contact pattern of gear pair B shows that despite a large tip relief, edge contact occurs where the tip relief starts. This can be one explanation to the lack of correlation between TE and the measurement result for gear pair B. Another explanation can be other excitations such as friction and bending moments. The results show the limitations of only considering transmission error when designing quiet gears.