Effects of sensor noise in digital signal processing of the three-point method

The three-point method is a technique that uses three displacement sensors to separate the out of roundness of a reference sphere or cylinder as well as the spindle run-out. This method has high potential for ultra-high-precision measurements. However, applications of the three-point method have been impeded because the setting angles of the sensors have not been clearly established. In this paper, the effects of uncorrelated noise in sensor outputs on the calculation results of the roundness are analyzed by applying the process of the three-point method as a digital filter. Then, the optimum sensor angles, which minimize the effects of the noise, are computed where there is uncorrelated white or pink noise in sensor outputs. In addition, the effects of correlated noise in sensor outputs are analyzed