A cutting power model for tool wear monitoring in milling

Abstract

This paper describes a cutting power model in face milling operation, where cutting conditions and average tool flank wear are taken into account. The cutting power model is verified with experiments. It is shown with the simulations and experiments that the simulated power signals predict the mean cutting power better than the instantaneous cutting power. Finally, the cutting power model is used in a cutting power threshold updating strategy for tool wear monitoring which has been carried out successfully in milling operations under variable cutting conditions.

Introduction

In metal cutting operations, tool condition must be monitored either by operators or by on-line tool condition monitoring systems (TCMS) to avoid possible damage to both products and machine tools. With the ever-increasing demand for cost saving and quality improvements, on-line tool condition monitoring is becoming more and more important in modern manufacturing. In the last three decades [1], [2], [3], [4], many efforts have been made to develop a reliable and cost-effective TCMS for applications on the shop floor. The suggested techniques for tool condition monitoring can be grouped into two main categories [1], [2], [3]: direct methods and indirect methods. The direct methods can be implemented using devices such as touch trigger probes, optical sensors, and proximity sensors to measure the geometry and/or the morphology of cutting edge. The indirect methods are based on the acquisition of process variables from which tool condition can be deduced according to certain known signal patterns. The former methods are reliable, but they cannot provide continuous in-process measurements because cutting edges are generally inaccessible during cutting; on the contrary, the latter methods can take measurements while cutting tools are actively engaged in cutting, which makes it possible to monitor the cutting tool condition monitoring on-line.

In these indirect TCMSs, process variables such as cutting forces, acoustic emission, temperature, vibration, spindle motor power (current), etc. are measured continuously and tool states are estimated according to certain signal patterns which are correlated to tool wear or breakage [4]. Among many on-line TCMSs, the spindle motor power monitoring system is considered to be one of the most applicable systems for shop floor applications because it is relatively simple and its mounting hardly affects the machining operations. Like many other monitoring systems, most power monitoring systems are usually based on the constant threshold monitoring strategy where the measured power signals are continuously compared with a preset monitoring threshold that is assumed to be correlated to tool breakage or certain level of tool wear [3], [5], [6]. Although the constant threshold monitoring strategy is easy to apply, it is only valid for a particular set of machining conditions. Indirectly measured power signals are generally affected by work-piece material variation, geometry and material of the cutting tool and cutting condition. In particular, the constants or thresholds suggested to relate power signals to tool condition are specific to a certain set of cutting conditions. This necessitates the development and storage of a set of thresholds as well as many other parameters for each process condition of interest. In addition, extensive wear tests must be carried out for the conditions or sets of conditions desired in order to obtain the various constants or parameters needed to predict tool breakage or tool wear level. These costly drawbacks make industry very reluctant to accept the technology for practical applications on their shop floors. Clearly, it is very demanding for new monitoring approaches capable of circumventing these drawbacks, especially the ones that can deal with variable cutting conditions.

The main objective of this study is to develop a cutting power model in which cutting conditions (such as cutting speed, feed rate, depth of cut, work-piece material and tool material) as well as tool flank wear will be taken into account. Based on the power model, a threshold updating monitoring strategy, which can deal with variable cutting conditions, is presented.