protection are applied as the resulting wear is often nonlinear [4], [5]. In this case, once the protection is gone, a rapid wear increase will take place, often resulting in a sudden breakdown of the component. Hence, condition monitoring of the wear protection is a key aspect for plant maintenance and for this reason, early warning systems are highly beneficial for maintenance procedures [6]. 1.1 Tactile vs. contactless sensors In the context of industrial maintenance, there are three different kinds of wear monitoring: volume-based, massbased and distance-based methods. Measuring the volume or mass of the worn material would be impractical or even impossible in many applications of interest. Only Aus Wissenschaft und Forschung 28 Tribologie + Schmierungstechnik · 66. Jahrgang · 4/ 5/ 2019 DOI 10.30419/ TuS-2019-0021 1 Introduction Open tribosystems in material processing industry often suffer severe abrasive wear ranging up to some millimetres of wear depth [1]. For this reason, components like transport chutes, sieves, mills, etc. are optimised for increased wear resistance [2]. For efficient maintenance, detailed knowledge of the current operating conditions is necessary [3], especially when coatings or local wear Für vorrausschauende und kosteneffiziente Instandhaltung in der Schwerindustrie ist die Kenntnis des aktuellen Anlagenzustands von verschleißbelasteten Aggregaten von entscheidendem Interesse. Die Onlinemessung in abrasiver Umgebung ist durch viele Störfaktoren beeinflusst, wie z.B. Staubkontamination, Temperatur, Oberflächenstruktur, Ausrichtungsfehler, wechselnde Belastungen. Eine Bewertung der Einsetzbarkeit in abrasiver, staubverschmutzter Umgebung verschiedener Sensorprinzipien im Labormaßstab wurde in dieser Arbeit durchgeführt. Ultraschall-Abstandssensoren und induktive Sensoren zeigten die höchste Toleranz gegenüber dieser Umgebung. Induktive Sensoren wiesen bei entsprechendem Messabstand jedoch unzureichende Genauigkeit auf. Laser Triangulationssensoren zeigten die höchste Genauigkeit, sind aber empfindlich auf Staubkontamination. Schlüsselwörter Verschleißmessung, Abrasion, Verschleißsensor, Instandhaltung, Online-Messung, Staubkontamination, Condition Monitoring For predictive and cost efficient maintenance in heavy industry the actual condition of wear stressed components is of crucial interest. Online measurement in abrasive conditions is influenced by many factors like dust contamination, temperature, surface structures, misalignment or changing loads, to name a few. A lab-scale assessment of applicable wear sensor techniques for abrasive and dusty environments was the goal of this work. The best dust resistance was found for ultrasonic distance sensors and inductive sensors. However, the inductive sensor showed insufficient accuracy under the chosen conditions. Laser triangulation sensors showed the highest accuracy but are nonetheless sensitive to dust contamination. Keywords Wear measurement, abrasion, wear sensor, maintenance, online-measurement, dust contamination, condition monitoring Kurzfassung Abstract * Dr.mont. Markus Varga, MSc orcid-iD: https: / / orcid.org/ 0000-0001-8272-4122 Ing. Reinhard Grundtner, MSc orcid-iD: https: / / orcid.org/ 0000-0002-2177-4879 Alexander Maurer, BSc AC2T research GmbH, Wiener Neustadt, Austria DI. Dr. tech. Martin Kirchgaßner Castolin GmbH, Wiener Neudorf, Austria Online wear measurement in harsh environment. Part 1: Possible measurement strategies Markus Varga, Reinhard Grundtner, Alexander Maurer, Martin Kirchgaßner* Eingereicht: 10. Mai 2019 Nach Begutachtung angenommen: 17. Juni 2019 TuS_4_5_2019.qxp_T+S_2018 23.08.19 13: 15 Seite 28 distance-based sensoring methods combine the required accuracy, robustness and flexibility for using them in open tribosystems or at non-static machine parts operating under severe conditions. Distance measuring sensors can be classified in tactile and contactless sensors [7]. While tactile distance sensors need a so-called (mechanical) adjustor to relay the information from the measured object to a converter, this is not necessary for contactless sensors. Their main advantage is that the surface of the measured object remains unaffected from mechanical interactions with the measurement system. This however leads to higher sensor costs. 1.2 Contactless distance sensors for wear measurement In this group there is a great variety of different distance sensors based on different physical principles. To find the most suitable sensor for wear measurement in harsh abrasive environments a subjective evaluation in form of a FMEA (Failure Mode and Effects Analysis) has been carried out before this research, proposing inductive sensors, ultrasonic sensors and laser-triangulation sensors for wear measurement in abrasive condition with large wear volumes. Their working principles are sketched in Figure 1 and they will be introduced briefly in this section. 1.2.1 Inductive distance measurement Eddy-current sensors are part of the inductive sensor family. For this class, the complex coil impedance changes due to the influence of a permeable and/ or electrically conductive measuring object, whose position is relative to the coil (Figure 1a). This change in impedance can be achieved by generating eddy currents in electrically conductive targets and/ or by magnetic field induction. [7, p. 583] The main advantage of the eddy-current sensor is that only electrically conductive materials (e.g. metals) have an influence on the sensor signal. Most non-metallic materials like sand, cement, etc. don’t interact with the magnetic fields, which leads to the main benefit of this system: non-metallic transported material does not influence the measurement. Unfortunately, there is also a disadvantage: longer measuring distances require exceedingly larger coils. The diameter of the coil determines the size of the measuring spot. When moving components should be monitored or large wear depths are expected, the measurement range should start at > 100 mm which leads to measurement spot-diameters in the same range. Due to similar reasons capacitive sensors will not be applicable in open tribosystems, as they require a measurement distance larger than 10 mm [7, p. 584]. 1.2.2 Ultrasonic distance sensors Commercially available ultrasonic distance sensors using air as the transmission medium are based on the so-called pulse-echo principle (Figure 1b). Usually in periodic sequences pulse-shaped ultrasonic signals (“bursts”) are emitted from the ultrasonic transducer, where the ultrasonic signals usually consist of packets with a number of oscillations (typically 5-100) which are tuned to the operating frequency of the transmitter transformer. The transmitted signals propagate at the speed of sound in the transmission medium and are reflected by an object at distance d. The echo reaches the receiver converter after a delay time and is converted into electrical signals. Often the same ultrasonic transducer is used for both transmission and reception. Apart from the cost advantages, the ultrasonic transducer is aligned parallax-free on the same local area during transmission and reception. A disadvantage when using a single converter for transmitting and receiving is that the sensor is “blind” for a short time after sending the pulse. [7, p. 618 f.] Aus Wissenschaft und Forschung 29 Tribologie + Schmierungstechnik · 66. Jahrgang · 4/ 5/ 2019 DOI 10.30419/ TuS-2019-0021 Figure 1: Working principles of different distance sensors: a) eddy-current sensor; b) ultrasonic sensor; c) laser triangulation sensor TuS_4_5_2019.qxp_T+S_2018 23.08.19 13: 15 Seite 29 berg and Erdemir [14] also highlighted the importance of continuous monitoring of wear and load for adapting tribological systems in the future. Measurement of wall thickness and wear by ultrasonic reflectometry is a commonly used method, nevertheless it is often too inaccurate because of detrimental environmental influences. Brunskill et al. [15] highlighted this problem and gave a suggestion on how they can be compensated for accurate wear measurement. Unfortunately, this measurement method is not suitable for use at moving components. Eddy current sensors can be used for detection of small distance changes and are of high reliability and durability [e.g. [16]]. In tribology they are also used for oil film thickness detection in hydrodynamic conditions [17], next to capacitive sensors [e.g. [18]]. However, for larger distances both systems have substantial drawbacks. Another important issue to be taken into consideration for online wear measurement devices is the significant differences between lab-scale testing with well-controlled conditions and heavy industry applications, for which laboratory systems are often not applicable due to harsh environmental conditions. This paper discusses if online wear sensors are capable of measuring static as well as dynamic objects in harsh abrasive environments. Special focus should be laid on the dust resistance of the measurement system, as contamination with abrasive particles cannot be avoided in such applications. 2 Experimental 2.1 Utilised distance sensors For online measurement in abrasive environment a FMEA done by the authors pointed out that inductive, ultrasonic and laser triangulation sensors have the highest potential. In the context of moving, heavily loaded components which suffer heavy abrasive wear, the required measurement range/ object distance should be in the range of 100 mm. The necessary accuracy for maintenance actions for such components is in the range of 0.5-1 mm. One sensor of each type was selected for the required measuring range and distance: ■ Inductive sensor: Proxitron, MKN 070.19 S4, distance 0 -70 mm (largest industrially available range), resolution 70 µm ■ Ultrasonic sensor: Baumer, UNCK 09U6914/ D1, distance 3 -150 mm, resolution 0.3 mm, with beam nozzle for focusing ■ Laser sensor: Panasonic, HG- C1100, distance 65 - 135 mm, accuracy 70 µm Aus Wissenschaft und Forschung 30 Tribologie + Schmierungstechnik · 66. Jahrgang · 4/ 5/ 2019 DOI 10.30419/ TuS-2019-0021 In view of wear measurement in abrasive environment ultrasonic sensors show a big resistance against dust, but they are not able to measure through adhering material as in the case of inductive sensors. Compared to eddycurrent-sensors the measuring spot is smaller but it still reaches diameters a few millimetres at a measuring distance close to 100 mm. The spot size can be reduced by sound focusing techniques, e.g. by a beam nozzle. 1.2.3 Laser triangulation sensors Laser-triangulation sensors belong to the optical sensor family. Sensors based on this principle are suitable for absolute distance measurements with high precision and speed. The angular distances to an object point are measured from the ends of a base line of known length and its linear distance is calculated (Figure 1c). With the light point triangulation sensor, the object is illuminated by a point light source via a focusing lens. The light spot is imaged obliquely into the detector plane at an angle α with the aid of an imaging lens. An important prerequisite for the sensor to function properly is that the light is scattered from the object surface and not just reflected, so that just a part of the incident beam is reflected to the detector. The position of the spotted image on the detector is related to the distance between the object surface and the sensor. The absolute distance can be calculated from triangulation. [7, p. 662] Like all optical sensors laser triangulation sensors are sensitive to dust present in air and adhering both to the sensor’s and the measured objects’ surface. 1.3 Application of distance sensors in the field of tribology Although a wide variety of sensors is used in the field of tribology, e.g. force and torque sensors for calculation of the coefficient of friction [e.g. [8], [9]], temperature sensors, chemical sensors [e.g. [10]], etc., the usage of sensors for measuring wear loss online is still sparsely present in both real field applications and the literature. In most cases the wear loss is measured after testing either gravimetrically or by topography measurements [e.g. [11]]. Several relevant examples of online wear measurement systems described in the available literature shall be reviewed here. Eder et al. [12] used 6 laser triangulation sensors arranged in the three-dimensional space to investigate the small movements between shaft and bearing and therefrom derive the bearing wear, with the calibration of such measurements being able to be performed by means of 3D microscopes. The progress of wear over time for pneumatic seals could be obtained by Belforte et al. by scanning the seal profile [13], without the necessity of disassembling the seals from their seat. To this end, triangulation sensors were used to examine the non-uniform circumferential distribution of wear. Holm- TuS_4_5_2019.qxp_T+S_2018 23.08.19 13: 15 Seite 30 2.2 Evaluation of distance sensors in the laboratory To evaluate the applicability of the different sensor types in harsh abrasive environment while still providing accurate measurement results, laboratory evaluations were performed. Those experiments target three issues of relevance for these applications: ■ evaluation of the measuring spot diameter ■ influence of the tilting angle on the measured signal ■ influence of particle contamination in the field of view The measuring spot diameter strongly influences the lateral accuracy of the sensor. The larger the measuring spot is, the larger the surface features have to be. To detect the minimal feature size, e.g. a crack in the surface, two steel plates (S235JR) were moved into the field of view of the sensor (Figure 2a-b) where the gap width simulates a crack. For the inductive sensor the distance was 35.5 mm to the backplate and the thickness of the moving plates was 10 mm. When measuring real surfaces, a surface tilt is often inevitable. This can be caused by asymmetric load, inhomogeneous wear, or flanks in the present surface structure. To measure the maximal possible tilting angle the sensors were targeted towards a flat plane, which then was tilted around the central measurement line (Figure 2c). Abrasive material and dust has to be expected in the field of view of the sensors in abrasive applications. To evaluate this influence in the laboratory, the sensors were targeted onto a verticcally oriented flat plate and abrasive of different size was strayed between sensor and plate. The following particles were used for this experiment: ■ coarse abrasive: corundum, particle size 5 -10 mm ■ medium-size abrasive: quartz, 1.4-2.2 mm ■ fine abrasive: quartz, 212-300 μm 3 Results and Discussion 3.1 Minimal detectable feature size Uniform wear will be detected as increasing distance to the sensor. When local wear protection is applied, e.g. hardfacings, break-outs of the surface are also possible. In this case, the minimal lateral resolution of the sensor is also of interest. The measurement spot of the laser triangulation sensor is < 1 mm, which is small enough for most applications, both other sensors were tested according Figure 2a-b. Aus Wissenschaft und Forschung 31 Tribologie + Schmierungstechnik · 66. Jahrgang · 4/ 5/ 2019 DOI 10.30419/ TuS-2019-0021 Figure 2: Test-setups for the laboratory investigations: a-b) influence of feature size simulated by closing gap; c) influence of tilted surfaces For ultrasonic sensors the minimal feature size which can be measured depends on the continuous expansion of the diameter of the sonic cone when the sonic waves move towards the target. Hence the minimal structure width is highly distance-dependent, as displayed in Figure 3a. E.g. at the distance of 70 mm structures of <12 mm can be detected accurately. Due to the use of a beam nozzle the sonic cone gets “focused” and doesn’t expand linearly. The sensor has a “binary” characteristic when targeting edges: If an object, like the rising edge of a surface structure, moves into the measuring spot, there will be an instant jump of the output value to the new minimal distance when it is within the measuring spot diameter. Thus the ultrasonic sensor will just detect an outbreak if there is a flat reflecting surface spot of e.g. more than ~12 mm diameter in its middle at 70 mm distance. The behaviour of the inductive sensor for small structures is different compared to the ultrasonic sensor. Because of the working principle of inductive sensors, no sharp edges can be clearly detected within the measurement spot. When the sensor is exposed to the gap width experiment, firstly at a gap width of 10 mm, an influence on the measurement result can be seen in Figure 3b by an increase of the distance. Only gaps or cracks larger TuS_4_5_2019.qxp_T+S_2018 23.08.19 13: 15 Seite 31 surement does not fail instantly as for the ultrasonic measurement, but the value will decrease as the angle increases as seen in Figure 3c. Analysing the distance signal, there is no way to distinguish between a change in distance and a change of the tilting angle. Small angle changes have little effect on the result. For tilting up to 5° there is almost no distortion (0.28 %) of the measured value, up to 10° the measuring error is less than 2 % but at 30° the error is 24 %. The triangulation sensor shows no influence of the tilting angle until ~83°. Obviously enough light of the diffuse reflection reaches the sensor until this value. At a steeper angle there is not enough light reflected back to the sensor under these surface conditions on steel. 3.3 Influence of particle contamination Particle and dust contamination will be inevitable in the abrasive applications. Hence the sensitivity of the sensors to contamination was investigated with free falling particles in the field of view. Figure 3d shows a measurement of the ultrasonic sensors with a heavy flow of abrasives particles. Even under this strong contamination by abrasive material the true distance can be measured effectively. However, if the particles are large enough or densely packed, the ultrasonic sensor detects the distance to the particles instead. Aus Wissenschaft und Forschung 32 Tribologie + Schmierungstechnik · 66. Jahrgang · 4/ 5/ 2019 DOI 10.30419/ TuS-2019-0021 than 30 mm can be reliably detected. For gap widths of >120 mm the correct distance is measured, which is slightly larger than the sensor’s diameter (110 mm). It has to be noted, that the value of the distance signal at a structured surface, e.g. by weld seams, represents a weighted average of all the heights within the measuring spot. Hence it is difficult to gather reliable information about structures smaller than the sensor width on the surface. 3.2 Influence of tilting angle on the signal As the tilting of the surface cannot be excluded during operation (e.g. misalignment, asymmetric loading, inhomogeneous wear) the influence of tilting was investigated according to Figure 2c. The three sensors show very different behaviour when exposed to tilted surfaces as given in Figure 3c (For better visualisation the signals were offset). The output signal of the ultrasonic and inductive sensor in dependence of the tilting angle is depicted in Figure 3c. The ultrasonic sensor delivers accurate results up to a tilting angle of ~ 8.3°. Above that angle the reflection of the ultrasonic beam cannot reach the sensor and the measurement fails. With the inductive sensor the measured distance decreases non-linearly as the tilting angle increases. The mea- Figure 3: Results of the laboratory experiments: a) minimal feature size in dependence of measuring distance for the ultrasonic system; b) nonlinear dependency of gap width detection with inductive sensor; c) influence of tilting angle on distance signal; d) behaviour of ultrasonic output signal when particles are in the field of view TuS_4_5_2019.qxp_T+S_2018 23.08.19 13: 15 Seite 32 By an intelligent signal analysis the true surface can be reconstructed when the short-term signal failures caused by the abrasive are removed. The laser triangulation sensor shows a similar behaviour as the ultrasonic sensor. As the measurement spot is much smaller it detects more abrasive particles and shows a slight disadvantage compared to the ultrasonic sensor. When thinking of a conveyor line, measurements of chutes or wear plates will always be possible at times when no good is transported. The inductive distance sensor shows no influence of the passing abrasive. The sensor just measures the distance to metallic objects so non-metallic deposits on the roll surface or abrasive particles in the air are not affecting the measuring signal. Lab experiments in abrasive and dusty environments showed the highest dust resistance for inductive and ultrasonic distance sensors. With some minor disadvantages also optical sensors were usable with acceptable failure rates. The inductive sensor did not deliver accurate enough surface information. The applicability of the sensors in real field applications will be studied in the second part of this study [19]. 4 Conclusions In this work potential sensors for heavy wear in abrasive applications are investigated on lab-scale. The minimal feature size, the influence of tilting angle and abrasive contamination were studied. The following main findings can be drawn: ■ In heavy wear regime measurement ranges of >100 mm are required to allow for large wear volumes and component movement, and also to avoid sensor damage. ■ Although inductive sensors are tolerant against dust and abrasive contamination, their measurement range is constrained and does not enable accurate wear pattern depiction for large measurement ranges. ■ Focused ultrasonic sensors (with beam nozzle) reach a measurement spot diameter of <10 mm for an appropriate measurement distance. They are suitable for detecting homogenous wear loss. Small break-outs may not be detected. ■ Laser-triangulation sensors were able to measure most of the details of the surface, but they are sensitive to contamination of the field of view by abrasive or dust. The final target of this work is an online wear measurement system suitable for prolonged operation under abrasive conditions. A field test of such sensors is the goal of the second part of this study [19]. Acknowledgements This work was funded by the Austrian COMET Programme (Project K2 XTribology, Grant No. 849109) and supported by the Province of Niederösterreich (Project “Digi-Pro”, WST3-F-5030642/ 004-2018) and has been carried out within the “Austrian Center of Competence for Tribology” (AC2T research GmbH). References [1] J. A. Hawk, R. D. Wilson, “Tribology of earthmoving, mining and minerals processing,” in Modern tribology handbook, Boca Raton, London, New York, Washington DC, CRC Press, 2001, pp. 1331-1370. [2] L. Widder, S. Leroch, M. Kirchgaßner, M. Varga, „Finite Elemente-Simulation als Werkzeug für ein spannungsgünstiges Design von Hochdruck-Rollenpressen in der Zementindustrie,“ Berg- und Hüttenmännische Monatshefte 163/ 5, p. 181-187, 2018. [3] M. Varga, M. Haas, C. Schneidhofer, K. Adam, “Wear intensity evaluation in conveying systems-An acoustic emission and vibration measurement approach,” Tribology International, https: / / doi.org/ 10.1016/ j.triboint. 2019.01.008, p. in press, 2019. [4] M. Varga, L. Widder, M. Griesinger, K. Adam, E. Badisch, “Wear progress and mechanisms in high temperature sieves,” Engineering Failure Analysis 61, pp. 46-53, 2016. [5] H. Winkelmann, M. Varga, E. Badisch, “Influence of secondary precipitations in Fe-based MMCs on high temperature wear behaviour,” Tribology Letters 43, pp. 229- 234, 2011. [6] M. Varga, K. Adam, R. Wimberger, E. Badisch, “Cost efficient tribological systems in steel production based on life cycle optimisation,” in 5th World tribology Congress, Turin, IT, 2013. [7] M. Sellen et al., „Weg-, Winkelsensoren,“ in Sensortechnik, Berlin Heidelberg, Springer, 2014, pp. 573-724. [8] C. G. He, Y. B. Huang, L. Ma, J. Guo, W. J. Wang, Q. Y. Liu, M. H. Zhu, “Experimental investigation on the effect of tangential force on wear and rolling contact fatigue behaviors of wheel material,” Tribology International 92, pp. 307-316, 2015. [9] T. Neupert, E. Benke, D. Bartel, “Parameter study on the influence of a radial groove design on the drag torque of wet clutch discs in comparison with analytical models,” Tribology International 119, pp. 809-821, 2018. [10] C. Schneidhofer, A. Grafl, K. Adam, „Online Zustandsüberwachung von Hydraulikölen in der Stahlindustrie,“ Berg- und Hüttenmännische Monatshefte 163/ 5, pp. 193- 198, 2018. [11] M. Varga, “High temperature abrasive wear of metallic materials,” Wear 376-377, pp. 443-451, 2017. [12] S. J. Eder, C. Ielchici, S. Krenn, D. Brandtner, “An experimental framework for determining wear in porous journal bearings operated in the mixed lubrication regime,” Tribology International 123, pp. 1-9, 2018. [13] G. Belforte, L. Mazza, C. Visconte, “Non contact wear measurement on pneumatic seals,” Tribology International 48, pp. 73-77, 2012. Aus Wissenschaft und Forschung 33 Tribologie + Schmierungstechnik · 66. Jahrgang · 4/ 5/ 2019 DOI 10.30419/ TuS-2019-0021 TuS_4_5_2019.qxp_T+S_2018 23.08.19 13: 15 Seite 33 film bearings,” Tribology International 34, pp. 853-857, 2001. [18] G. Garcia-Atance Fatjo, E. H. Smith, I. Sherrington, “Mapping lubricating film thickness, film extent and ring twist for the compression-ring in a firing internal combustion engine,” Tribology International 70, pp. 112-118, 2014. [19] M. Varga, R. Grundtner, A. Maurer, M. Kirchgaßner, “Online wear measurement in harsh environment. Part 2: Application roller press,” Triblologie und Schmierungstechnik, 66. Jahrgang, pp. 43-51, 4/ 5/ 2019. Aus Wissenschaft und Forschung 34 Tribologie + Schmierungstechnik · 66. Jahrgang · 4/ 5/ 2019 DOI 10.30419/ TuS-2019-0021 [14] K. Holmberg, A. Erdemir, “Influence of tribology on global energy consumption, costs and emissions,” Friction 5 (3), pp. 263-284, 2017. [15] H. Brunskill, P. Harper, R. Lewis, “The real-time measurement of wear using ultrasonic reflectometry,” Wear 332- 333, pp. 1129-1133. [16] K. Poulios, N. Drago, P. Klit, L. De Chiffre, “A reciprocating pin-on-plate test-rig for studying friction materials for holding brakes,” Wear 311, pp. 40-46, 2014. [17] S. B. Glavatskih, Ö. Uusitalo, D. J. Spohn, “Simultaneous monitoring of oil film thickness and temperature in fluid TuS_4_5_2019.qxp_T+S_2018 23.08.19 13: 15 Seite 34