No abstract

In machining processes, cutting forces measurement is essential to allow cutting process and tool conditions monitoring. Moreover, in order to have information about the quality of the milled part, the amplitude of the tool tip vibration would be very useful. Since both the measurements are extremely complicated especially in an industrial scenario, in this study, an in-process model based estimator of cutting forces and tool tip vibration was designed and properly tested. The developed estimator relies on both a machine dynamic model and on indirect measurements coming from multiple sensors placed in the machine. The machine dynamic model was obtained through an experimental modal analysis session. The estimator was developed according to the Kalman Filter approach. The fusion of multiple sensors data allowed the compensation of machine tool dynamics over an extended frequency range. The accuracy of the observer estimations was checked performing two different experimental sessions in which both the force applied to tool and the tool tip vibration amplitude were measured. In the first session, the tool was excited with different sensorized hammers in order to appreciate the broad-bandwidth of the performed estimations. In the second one, real cutting tests (steel milling) were done and the cutting forces were measured through a dynamometer, tool tip vibrations were measured as well. The experimental results showed that the indirect estimation of cutting forces and tool tip vibrations exhibit a good agreement with respect to the corresponding measured quantities in low and high frequency ranges. The contribution of this research is twofold. Firstly, the conceived observer allows estimating the tool tip vibrations that is a useful information strictly connected to the surfaces quality of the processed workpiece. Secondly, thanks to a multi-sensors approach, the frequency bandwidth is extended especially in the low frequency range.

No abstract

The capability of estimating the surface quality of workpieces in machining is still a challenging goal. The morphology of the processed surfaces does not only depend on nominal tool geometry and on machining parameters but it is also affected by several complex cutting phenomena and deviations from nominal conditions. In this paper, a framework model for estimating the surface texture in milling operations was developed. The model allows considering various tool geometries and the corresponding alignment/mounting errors. Since the back cutting phenomenon is adequately formalized, the model is particularly suitable for estimating the surface topography in face milling. Although the model does not consider the contribution due to the cutting forces, it is suitable for being fed by measured tool vibrations. The predicting capabilities of the conceived model were tested considering a highfeed milling operation that typically generates complex patterns on the processed surfaces. The model validation was carried out comparing the numerical and the real machined surface morphology. The analysis confirmed that the surface morphology can be predicted with negligible errors.