11th European Conference on Non-Destructive Testing (ECNDT 14), October 6-10, 14, Prague, Czech Republic More Info at Open Access Database www.ndt.net/?id=16643 Ultrasonic Method for Measuring Frictional Temperature Rise in Rubbing Surface Ikuo IHARA 1, and Shingo AOKI 2 1 Department of Mechanical Engineering, Nagaoka University of Technology; Nagaoka, Japan Phone: +81 258 4797, Fax: +81 258 4797; E-mail: ihara@mech.nagaokaut.ac.jp 2 Graduate Student of Nagaoka University of Technology; E-mail: s113002@stn.nagaokaut.ac.jp Abstract An ultrasonic thermometry is applied to temperature measurements during the friction process between an acrylic resin and felted fabric plates. Temporal variations in the temperatures at and near the friction surface of the acrylic plate has been examined. Both the temperature at the friction surface and temperature distribution beneath the surface are quantitatively determined by the ultrasonic thermometry. The ultrasonic pulse echo measurements at 2 MHz are performed for an acrylic resin plate of 10 mm thickness whose single side is being heated by rubbing with a felted fabric plate. The influence of the loading pressure between the acrylic resin and fabric plate on friction temperature rise has been examined. Rapid temperature rise more than 30 K at the friction surface is observed within 0.5 s immediately after rubbing starts. It has also been observed that the temperature at the friction surface increases with the loading pressure. Thus, it has been demonstrated that the ultrasonic thermometry is a useful tool for measuring temperature rise in friction surface. Keywords: Ultrasonic thermometry, Frictional temperature, Rubbing surface, Acrylic resin, In-situ observation 1. Introduction Accurate knowledge about temperature rise due to friction between rubbing surfaces has long been a topic of great interest in the field of tribology as well as any other relating fields in science and engineering. This is basically because such temperature rise is closely related to the mechanical and physicochemical properties of the rubbing materials. Such temperature also influences tribological properties and lubrication efficiency of the rubbing surfaces [1]- [4]. Although there have been many theoretical and experimental works on the topics, few studies on direct measurements of the temperature rise at the friction surface have so far been made. This is because of extreme difficulty in measuring such temperature rise at the rubbing surface. Unfortunately, little is known about the temperature measured at the interface of two rubbing materials. Therefore, measuring such temperature rise due to friction and examining its variation comprehensively are important issues. In particular, in-situ and real-time monitoring technique for the temperature could be quite beneficial not only for basic research in tribology but also for making an effective quality control of material manufacturing with tribological processes. Ultrasound, because of its capability to probe the interior of materials and high sensitivity to temperature, is considered to be a promising tool for measuring internal temperature of materials. Since there are some advantages in ultrasonic measurements, such as non-invasive and faster time response, several studies on the temperature estimations by ultrasound have been made [5]-[11]. In our previous works [12]-[15], an effective ultrasonic methods for measuring internal temperature distributions of heated materials were developed and applied to internal temperature profiling of heated materials. Since the ultrasonic method, so-called ultrasonic thermometry, can determine surface temperature at a heated surface without using any thermal boundary conditions at the heated surface, it is quite attractive and highly expected to apply the ultrasonic thermometry to temperature measurements at a friction surface between two rubbing materials. In this work, the ultrasonic thermometry has been applied to frictional temperature measurements in a rubbing acrylic plate. The influence of the loading pressure at the rubbing surface on frictional temperature rise has been examined.
2. Method for determining internal temperatures of heated material When a material slides over another material, heat is generated at the interface due to friction and the heat is transferred from the interface to each material. The generated heat and its transfer depends on the thermal properties of the materials and sliding conditions such as contact surface geometry and sliding speed. Figure 1 shows a schematic an internal temperature gradient in a material whose single side is uniformly heated due to friction. To investigate the temperature gradient, a one-dimensional unsteady heat conduction with a constant thermal diffusivity is considered. Assuming that there is no heat source in the material, the equation of heat conduction is given by [16] T t 2 T 2 x, (1) wheret is temperature, x is the distance from the heated or cooled surface, t is the elapsed time after the heating or cooling starts, is the thermal diffusivity. It is known that the temperature distribution can be estimated by solving Eq. (1) under a certain boundary condition. In an actual heating process such as a frictional heating, however, the boundary condition at the frictional surface is quite difficult to know and often being changed transiently during heating. Such boundary condition can usually not be measured. Because of little knowledge about the boundary condition, temperature gradient is hardly determined from Eq. (1). This kind of problematic situation usually occurs in frictional heating processes. To overcome the problem mentioned above, we have applied an effective method, so-called the ultrasonic thermometry [13-15], to the evaluation of the temperature of the friction surface as well as the temperature distribution beneath the surface. This method consists of an ultrasonic pulse-echo measurement and an inverse analysis coupled with a one-dimensional finite difference calculation. The advantage of the method is that no boundary condition at the heating surface is needed. A onedimensional finite difference model consisting of a large number of small elements and grids is used for analyzing heat conduction in the heated material. Using two information, the initial temperature distribution and the temperature dependence of the ultrasonic velocity of the material to be evaluated, not only the temperature at the heating surface but also the temperature distribution in the material can be determined by the method. Temporal variations Figure 1. Schematic of a one-dimensional temperature distribution in a rubbing material whose single side is heated by friction.
of the determined temperatures can also be obtained as long as the transit time of ultrasound propagating in the direction of temperature gradient of the heated material is continuously being measured using an ultrasonic transducer as shown in Figure 1. It is noted that both the temperature at B which is the outer surface and the temperature dependence of the material should be given as known information in the inverse analysis. The detailed procedure for temperature determination by the method is described in References [13-15]. 3. Experiment An acrylic resin plate (50 mm x 50 mm x 10 mm thickness) is used as a specimen and its surface is rubbed with a felted fabric of 5 mm thickness so that the rubbed surface of the acrylic plate is heated by frictional heat. Figure 2 shows a schematic of the experimental setup used. This system provides not only ultrasonic pulse-echo measurements but also internal temperature measurements of the acrylic plate using thermocouples inserted in the plate, so that the validity of the ultrasonically determined temperatures can be verified by comparing with those measured using the thermocouples. Three thermocouples, TC1, TC2 and TC3, are inserted in the acrylic plate, at three locations, 1, 5 and 9 mm from the bottom (friction surface), respectively. Another thermocouple, TC4, is placed on the top surface of the plate to measure the surface temperature which is used as a known data in the ultrasonic thermometry. The acrylic plate is forced on the surface of the rotating felted fabric so that an appropriate rubbing between the felted fabric and acrylic plate can occur. The loading force is changed from 60 N to 90 N. The loading pressure is then calculated from the compress load measured using a load-cell. The diameter of the felted fabric is 150 mm and the speed of rotation is 0 rpm. The sliding velocity between the fabric and acrylic plate is approximately 0.5 m/s at the location of the ultrasonic measurement. A longitudinal ultrasonic transducer of 2 MHz is installed on the top surface of the acrylic plate and ultrasonic pulse-echo measurements are performed. Ultrasonic pulse-echoes from the bottom surface (friction surface) of the acrylic plate are continuously acquired every 30 ms with a PC based real-time acquisition system. The sampling rate of ultrasonic signal is 100 MHz. Signal fluctuation due to electrical noise in measurements is reduced by taking the average of fifty ultrasonic signals. Such ultrasonic signals are used for determining both the temperature at the friction surface and temperature distribution beneath the friction surface of the acrylic plate. PC A/D board Specimen: Acrylic resin plate Ultrasonic transducer Loading Ultrasonic transducer Ultrasonic pulser/receiver Amp. Thermocouples Felted fabric on rotating disc TC4 TC3 TC2 TC1 Acrylic resin plate Felted fabric Rotating disc Figure 2. Schematic of the experimental setup: acrylic plate on a rotating felted fabric disc and schematic diagram of measurement system (left); cross sectional view of the setup (right).
4. Results and discussion Figure 3 shows ultrasonic pulse echoes measured from the acrylic plate. Two echoes, the first and second reflected echoes, are clearly observed. The transit time through the acrylic plate can precisely be determined from the time delay between the two echoes by taking a crosscorrelation between them. Using the transit times acquired during the rubbing process, the temperature at the friction surface and internal temperature distribution of the acrylic plate are obtained as a function of measurement time. It is noted that the temperature dependence of the longitudinal wave velocity of the acrylic resin plate used is found to be approximately v(t) = 0.0214T 2 1.9749T+2769.3 (m/s). The temperature dependence and the initial temperature of the acryl plate before heating, 22.5 o C, are used in the estimation. Figure 3. Measured ultrasonic pulse echoes from the bottom surface of the acrylic resin plate of 10 mm thickness. Figure 4 shows the estimated temperature distribution and its variation with elapsed time, where the numbers shown in Figure 4 denote the elapsed time after rubbing starts. It is noted that this result is obtained under the loading pressure of approximately 28 kpa. It can be seen in Figure 4 that the temperature distributions estimated ultrasonically almost agree with those measured using thermocouples, while the estimated value of thermocouples are slightly higher than those of ultrasonic results. It is worth noting that there occurs a steep temperature gradient in the shallow area of near-surface even at 0.5 s immediately after the rubbing started, and the temperature at the friction surface increases with the elapsed time and accordingly, the temperature distribution becomes deeper. Similar tendency to that shown in Figure 4 is observed in the results obtained under the loading pressures of 38 kpa and 42 kpa. Figure 5 shows the temperature variations at the friction surface and 1mm beneath the friction surface of the acrylic resin plate. The temporal variation in the loading pressure is also depicted in the figure. It has been found in the ultrasonic result shown in Figure 5 (a) that the temperature at the friction surface increases rapidly just after the rubbing due to the loading occurs and then drastically decreases when the loading is removed and gradually decreases with the elapsed time. It is noted that the friction surface temperature reaches almost 80 o C by the friction heating due to the loading pressure of 28 kpa for 2.2 s. On the other hand, the internal temperature at 1mm beneath the friction surface increases slowly a little later after the rubbing starts and then decreases very slowly. The temperature rise is not significant and it reaches only 36 o C at most. Although the internal temperature at the same location measured using the thermocouple shows a similar tendency to that of the ultrasonic result, there is a
certain discrepancy between them. The reason for the discrepancy is currently being investigated. Figures 5 (b) and 5 (c) which are obtained under the loading pressures of 38 kpa and 42 kpa, respectively show the similar tendency to that of Figure 5 (a). It should be noted that the friction surface temperature increases with the loading pressure and it can reach Figure 4. Temperature distribution near friction surface of the acrylic resin plate and its variation, measured by the ultrasonic method and thermocouples. Temperature ( o C) 140 100 60 Ultrasound (friction surface) Ultrasound (1 mm Thermocouple (1 mm 50 40 30 10 Presure (kpa) Temperature ( o C) 140 100 60 Ultrasound (friction surface) Ultrasound (1 mm Thermocouple (1 mm 50 40 30 10 Presure (kpa) 0 0 40 Elapased time (s) 0 0 40 Elapased time (s) (a) (b) Temperature ( o C) 140 100 60 Ultrasound (friction surface) Ultrasound (1 mm Thermocouple (1 mm 50 40 30 10 Presure (kpa) 0 0 40 Elapased time (s) (c) Figure 5. Temperature variations at the friction surface and 1mm beneath the friction surface of the acrylic resin plate, measured by the ultrasonic method and thermocouple, under the loading pressure (a) 28 kpa, (b) 38kPa, and (c) 42 kpa.
almost 110 o C due to the loading pressures of 42 kpa. Thus, it seems reasonable to think that the ultrasonic thermometry properly detect and estimate the temporal variation in the temperature at friction surface during the rubbing process. 5. Conclusion An ultrasonic thermometry has been applied to temperature measurements during the friction process between an acrylic resin and felted fabric plates. Temporal variations in the temperatures at and near the friction surface of the acrylic plate has been examined. Rapid temperature rise due to frictional heating has clearly been observed by the ultrasonic method. It has also been found through the experiments that the friction surface temperature increases with the loading pressure and it can reach almost 110 o C due to the loading pressures of 42 kpa. Thus, in-situ temperature measurements of frictional heating surface by the ultrasonic thermometry has been demonstrated. Although further study is necessary to improve the accuracy in the measurement, it is highly expected that the ultrasonic method could be a useful tool for on-line monitoring of temperatures at friction surface and beneath the surface during friction process. Acknowledgment Financial supports by the Grant-In-Aid for Scientific Research (B25289238) from the JSPS and Toyota Motor Co. are greatly appreciated. References 1. L Mu, Y Shi, X Feng, J Zhu and X Lu, 'The effect of thermal conductivity and friction coefficient on the contact temperature of polyimide composites: Experimental and finite element simulation', Tribology International, 53, pp. 45-52, 12. 2. T F J Quinn, W O Winner, 'The thermal aspects of oxidational wear', Wear, 102, pp. 67-80, 1995. 3. S Suzuki, and F E Kennedy, Friction and temperature at head-disk interface in contact start/stop tests, Society of Tribologists and Lubrication Engineers, Park Ridge, pp. 30-36, 1988. 4. X Tian, F E Kennedy, J J Deacutis and A K Henning, The development and use of thin film thermocouples for contact temperature measurement, Tribology Transactions, 35, pp. 491-499, 1992. 5. T-F Chen, K T Nguyen, S-S Wen and C-K Jen., Temperature measurement of polymer extrusion by ultrasonic techniques, Meas. Sci. Technol., 10, pp. 139-145, 1999. 6. K Balasubramainiam, V V Shah, R D Costley, G Boudreaux and J P Singh, High temperature ultrasonic sensor for the simultaneous measurement of viscosity and temperature of melts, Rev. Sci. Instrum., 70-12, pp. 4618-4623, 1999. 7. W-Y Tsai, H-C Chen and T L Liao, An ultrasonic air temperature measurement system with self-correction function for humidity, Meas. Sci. Technol., 16, pp. 548-555, 05. 8. K Mizutani, S Kawabe, I Saito and H Masuyama, Measurement of temperature distribution using acoustic reflector Jpn. J. Appl. Phys., 45-5B, pp. 4516-45, 06. 9. F L Degertekin, J Pei, B T Khuri-Yakub and K C Saraswat, In-situ acoustic temperature tomography of semiconductor wafers, Appl. Phys. Lett., 64, pp. 1338-1040, 1994.
10. K-N Huang, C-F Huang, Y-C Li and M-S Young, High precision fast ultrasonic thermometer based on measurement of the speed of sound in air, Rev. Sci. Instrum., 73-11, pp. 4022-4027, 02. 11. C Simon, P VanBaren and E Ebbini, Two-dimensional temperature estimation using diagnostic ultrasound, IEEE Trans. Ultrason. Ferroelectr. Freq. Control, 45-.4, pp. 1088-1099, 1998. 12. M Takahashi and I Ihara, Ultrasonic Monitoring of Internal Temperature Distribution in a Heated Material, Jpn J. App. Phys., 47, pp. 3894-3898, 08. 13. I Ihara and M Takahashi, Non-invasive Monitoring of Temperature Distribution inside Materials with Ultrasound Inversion Method, Int. J. Intelligent Systems Technologies and Applications, 7-1, pp. 80-91, 09. 14. I Ihara and T Tomomatsu, In-Situ Measurement of Internal Temperature Distribution of Sintered Materials Using Ultrasonic Technique, IOP Conf. Series: Materials Science and Engineering, 18, 0208, 11. 15. M Takahashi and I Ihara, Quantitative evaluation of one-dimensional temperature distribution on material surface using surface acoustic wave, Jpn J. App. Phys., 48, GB04-1-5, 09. 16. G.E. Meyers, Analytical Methods in Conduction Heat Transfer, New York: McGraw- Hill, pp. 10, 1971.