2016 Volume 10 Issue 4 Pages JAMDSM0059
We developed a new method that can diagnose damage on a gear tooth surface using a laser beam without a rotary encoder. The method procedure is as follows: 1) The tooth bottom, the tooth tip and their two median values are detected using the differential values of the laser reflection data. 2) The gear rotation speed is calculated with these four positions, and interpolated according to the rotation fluctuation. 3) Using the calculated gear rotation speed, the measured data can be converted to the corresponding gear rotation angles. Thus we can diagnose the damage of the gear tooth surface precisely and can easily set up the experimental measurements without being influenced by rotational fluctuation. To confirm the validity of the method, we conducted the diagnosis experiment and we created contour maps to show the diagnosis accuracy variation according to the fluctuations of the amplitude and cycle. Based on these maps, we found that the diagnosis accuracy of the damage size is the same irrespective of the presence or absence of a rotary encoder. The diagnosis accuracy of the damage location without using a rotary encoder is lower than the result obtained using a rotary encoder because we assumed that the detection of the damage start point is delayed using this new method. Furthermore, we defined the limit using the conditions of this method from the sampling theorem; the validity of this definition could also be confirmed from the contour map.
