Signal processing for the identification of tire faults

In this work, different signal processing techniques are used to identify the tires’ radial runout. The presence of tires grooves (designed to allow water to be expelled) actually prevent from using the discrete Fourier transform on the tire radial profile measured acquired with noncontact technique (laser working with triangulation principle). We have therefore developed an adaptive algorithm that iteratively erases the groves with a procedure similar to the erosion filter already used in the literature for the same problem. The efficiency of the proposed method has been verified by comparing the runout computed with our method coupled with a moving average procedure and the measurements performed on a commercial balancing machine. The signals were measured in different positions on the tire inflated at different pressures ranging between 200 and 300 kPa. In addition, experiments with two locking mechanisms were performed. Results showed that the PkPk errors and the first four harmonics are more repeatable with respect to the method actually used.