Tools with built-in thin film thermocouple sensors for monitoring cutting temperature

https://doi.org/10.1016/j.ijmachtools.2006.09.007Get rights and content

Abstract

The ability to monitor in real time, the thermal activation and thermal impact to cutting tools has been very appealing to the manufacturing industries. Such responses can be measured with appropriate sensors such as thin film thermocouples (TFTs) built in cutting tools. The challenges have been to instrument the tool, equipment and sensors, which can withstand high stress and temperature in machining process. In this work, the sequence of fabricating the built-in TFTs and experimental setups are proposed and demonstrated. The cutting experiments are carried out under different cutting conditions for A6061-T6 aluminum alloy and finally cutting temperature is measured at very high cutting speeds up to 16 m/s.

Introduction

In manufacturing process, especially in machining, there has been a surge of interest to know more about the tool temperature. Temperature has critical influences on machining: it accelerates tool wear shortening tool life; it causes the thermal deformation of the workpiece, cutting tool and machine tool, which degrades the machining accuracy; it affects the subsurface layers of machined parts through phase transformation, thermally induced residual stresses and increase in thermally activated defects. However, it is hard to predict the intensity and distribution of the heat sources in a machining operation due to the complexity of the mechanics and heat transfer of machining. Especially, because the properties of work materials are dependent upon strain, strain rate and temperature, the mechanics and heat transfer are tightly coupled together in machining. Although lots of efforts in theoretical analyses and experiments [1], [2] have been made to understand this phenomenon, many problems are still remaining unsolved.

Many experimental methods have been devised to measure the temperature at the tool–chip contact surface: tool–chip thermocouple technique [3], embedded thermocouple technique [4], [5], infra red radiation measurement [6], measurement of metal microstructure and microhardness variation [7], utilization of melting points of thermo-sensitive materials [8], and thin film sensor technique, which is based on the property of pure platinum film that its ohmic resistance changes with temperature [9], [10]. Each of them has its own advantages and disadvantages. Thin film sensors have a great potential to industry applications [9], but measurement of temperature at the tool–chip interface was limited to a free machining aluminum alloy among the engineering alloys and measured temperature was approximately 200 K higher than solidus temperatures of aluminum alloys [10]. By contrast, the numerical methods such as finite difference method [4], [5], finite element method [11], [12], hybrid analytical-FEM method [13], boundary element method [14] and inverse method [15] have developed to solve the problems related to cutting temperature. However, rather simple assumptions are implemented into analysis codes due to the complexity of thermal problems in cutting processes.

Sensors in a form of thin film thermocouples (TFTs) have been developed by the authors for measuring temperature at the tool–chip contact zone under high temperature and pressure. A prototype tool with a Pt—(Pt-13%Rh) TFT and a hard protective coating of Al2O3 and AlN had been applied to machining of 0.45%C carbon steel S45C at cutting speeds 1.67–5.0 m/s and feed rates 0.08–0.25 mm/rev [16], but the width of TFT 1.8 mm was too wide to measure the temperature distribution and cutting experiment had to be finished in a few seconds to avoid the injury of TFT. In this paper, fine TFTs were made of Ni and Ni–Cr (80:20 in mass%) layers deposited on Al2O3 substrate tool. The TFT layers approximately 0.5 μm thick were insulated by a PVD layer of hafnium oxide (HfO2). Then, a TiN hard layer was deposited on the HfO2 insulator for protecting the built-in TFTs against the wear and high stresses in high-speed cutting. Calibration of the TFTs was followed by machining experiment and temperature measurement with only an insert with TFTs at different cutting speeds.

Section snippets

Working principle

A schematic view of TFT application to orthogonal machining and the longitudinal cross section of a tool with built-in TFTs are shown in Fig. 1. The hot junctions of Ni and Ni–Cr thermocouples are fabricated very close to the cutting edge; the cold junctions, which are virtually terminals to a recorder, are placed away from the edge, but on the rake face. For this arrangement of the sensor, the electromotive force (EMF) of a TFT is due to the temperature difference between the hot junction and

Device fabrication

The TFT sensors to be used for measuring the cutting temperature must fulfill following requirements: (1) hard coating layers for protecting the sensors have enough adhesion forces to the beneath layers to exhibit good wear resistance and reasonable durability during machining operations; (2) the sensors are small enough both in dimension and mass to achieve high sensitivity; and (3) fabrication and installment of the sensors do not cause changes in machining processes. To meet these

Calibration of TFTs

All the TFTs on all the tool inserts, more than 30 thermocouples, were calibrated to obtain Seebeck coefficients of TFTs. As a result, it was found that Seebeck coefficients of TFTs were not constant and smaller than that of a Ni—(Ni–Cr) wire thermocouple 0.5 mm in diameter. This is chiefly because the TFTs have very large ohmic resistance as compared with the wire thermocouple. Measured Seebeck coefficients and ohmic resistances of TFTs showed the following exponential relation:St=exp(3.199-

Experimental setup

The orthogonal cutting experiment was performed with a lathe. Force and temperature measurement systems were set up on it, as shown in Fig. 5. The force measurement system consisted of two piezoelectric quartz dynamometers, a three channel charge amplifier and a personal computer. The dynamometers were set under the base of tool holder to detect the cutting and thrust forces. The detected signals were amplified with the amplifier and then collected by a computer through an A/D card. The EMFs of

Results and discussion

All the data for six different cutting speeds were obtained using only one tool insert with TFTs without breakage of TiN hard and protective coating. Fig. 6 shows cutting and thrust forces, and temperatures in two zones at two different distances of 0.3 and 0.5 mm from the cutting edge as well. Only temperatures measured with the second and third TFTs in order of distance from the cutting edge are shown because temperature measured with the first TFT was very unstable. The unstable formation of

Conclusions

In this study, reactive DC magnetron sputtering and helicon sputtering, physical vapor deposition methods, was used to deposit the TFTs of Ni–(Ni–Cr), insulating layer of HfO2 and protective hard coating of TiN on an alumina tool insert. Chemical machining and photolithography were used to shape the thermocouple circuit in the tool. Different patterns and different widths of hot junctions were fabricated to measure the cutting temperature in the regions of the tool–chip contact surface. The

Acknowledgments

The authors would like to thank Mr. Utsumi, who was a graduate student of Tokyo Institute of Technology, for his contribution in the TFT fabrication and cutting experiment. They also acknowledge Onward Ceramic Coating Co., Ltd. for hard coating of TiAlN or TiAlSiN. This paper is partly based on researches supported by Grant-in-Aids for Scientific Research (B)(2), 14350068, 2002 and (B)(2), 17360060, 2005, and Young Scientists (B), 15760077, 2003, Japan.

References (19)

There are more references available in the full text version of this article.

Cited by (102)

View all citing articles on Scopus
View full text