14 Tribologie + Schmierungstechnik 64. Jahrgang 5/ 2017 investigated. It has been shown that the base material has a significant influence on the adhesion between the polymer coating and the substrate. For an aluminium alloy substrate (Al390-T6), due to insufficient adhesion of the polymer coating, premature coating failure and seizure occurred, however, for a grey cast iron or sintered iron substrate, the adhesion of the polymer coating was sufficient and no coating failures could be observed. The mentioned study also showed that the lifetime of the transfer film can be extended by the addition of MoS 2 particles as a solid lubricant in the coating. According to the authors, the effect of the wear particles coming from the polymer coating, which also act as a solid lubricant, is more dominant in the overall wear behaviour of polymer coatings than the mechanical properties of the polymer coating itself. In other studies with similar operating conditions as those encountered in lubricant-free axial piston compressors using CO 2 as coolant, PAI, PEEK and fluorocarbon-based polymer coatings were evaluated [3,4]. In these studies, it was outlined that PAI-based coatings possess the best performance both in terms of friction and wear. It was also found that the tested PAI-based coatings possess a higher hardness and a lower modulus of elasticity than the PEEK-based coatings [4]. However, the improved tribological performance of PAI-based coatings was not directly related to their micro-mechanical properties, but was rather related to the structural uniformity of the coating. In other words, the load-deformation behaviour of the highest performant PAI-based coating (DuPont™ Teflon ® 958-414) was very reproducible, indicating a uniform and amorphous microstructure. However, the load-deformation behaviour of the highest performant PEEK-based coating (1704 PEEK / PTFE ® ) was very different and unpredictable, which is associated with its semi-crystalline and porous microstructure (PEEK-based coatings possess a high crystallization rate). Due to the amorphous microstructure of PAI-based coatings, very fine wear particles are formed during sliding, which lead to the formation of a cohesive transfer film. On the other hand, due to the semi-crystalline structure of PEEK-based polymer coatings, large flake-like abrasion particles are formed, which prevent from any efficient formation of a transfer film [3,4]. 3 Sample geometry In the present study, uncoated pins were tribologically tested against polymer coated discs. The contact surface of the pin had a diameter of 8 mm, resulting in a contact area of approximately 50 mm 2 . The edges were rounded with a radius of 0.3 mm in order to avoid any mechanical stress peaks at these edges. The discs had a diameter of 24 mm and a height of 7.85 ± 0.05 mm. In order to take into account a rough, textured surface structure as prevalent in real applications, the surfaces of the pins and discs were milled. The R z roughness values of the surfaces were ranging between 3 µm and 4 µm. Figure 1 shows topographic images of both surfaces. 4 Base material, polymer coatings and lubricants Nodular grey cast iron EN-GJS-700-2 was used as the base material. Compared to other nodular grey cast iron grades, this material has a high strength (in accordance with DIN EN 1563 Standard, the specified tensile strength is 700 N/ mm 2 ), has a high wear resistance and, at the same time, has a good mechanical machinability. EN-GJS-700-2 is normally used for machine parts and tools which are subjected to high dynamic and static loads. Different high-performance polymer coatings based on PAI and PAEK were selected. In addition to polyether-ether-ketone (PEEK), which is the most popular polymer of the PAEK family, a mixture of different PAEK types was also used. All of the selected coatings contain solid lubricants for the reduction of the adhesive component of friction. Table 1 lists the different coatings investigated in the present study. Polyalkylene glycol-based (PAG) oil was used as lubricant, which is suitable for use in transcritical CO 2 applications, e. g. air conditioning compressors and heat pumps. The oil contains additives to enhance wear protection under extreme operating conditions and has a kinematic viscosity of 68 mm 2 / s at 40 °C. Aus Wissenschaft und Forschung Figure 1: Topographic images of (a) pin and (b) disc. Measuring range is 2.9 x 2.9 mm (a) (b) T+S_5_17 31.07.17 10: 58 Seite 14 Tribologie + Schmierungstechnik 64. Jahrgang 5/ 2017 5 Experimental setup and procedure The tribological tests were carried out on a linear-oscillating tribometer (SRV ® 4, Optimol Instruments Prüftechnik GmbH, Germany). The tribometer was modified in such a way that the experiments can be carried out under a gas atmosphere. The modified test setup consists of an upper and a lower specimen holder, which are connected by an elastomer (natural rubber) sleeve and thus form a sealed specimen chamber (Figure 2). A sensor system, with which it is possible to regulate gas flow and gas pressure, has been developed for regulating and monitoring the gas atmosphere. The modification of the test setup for different gas atmospheres was described in more detail elsewhere [5]. The upper specimen holder is designed as a spherical joint in order to allow a selfalignment of the pin. The lower specimen holder consists of a sample bath for discs, in which the sample can be, if necessary, completely covered with lubricant. During the tribological tests, pins were subjected to a normal force of 245 N, which corresponds to a nominal contact pressure of approximately 5 MPa. The experiments were carried out at a contact temperature of 160 °C, an oscillation frequency of 25 Hz and a stroke of 4 mm and for a total duration of 120 minutes. For all tribological investigations, 2 mg of lubricant was distributed evenly over the disc surfaces (condition of starved lubrication). During the experiments, the gas chamber was continuously supplied with CO 2 having a purity higher than 99.5 % (Linde AG, Linde Gases Division, Germany). A gas overpressure of 0.02 bar was applied to the sealed specimen chamber at a constant gas flow of approximately 1.3 l/ minute. Each tribological test was repeated 3 times in order to obtain statistically reliable results. During the tests, the coefficient of friction was measured and after the tests, the wear scars of the worn polymer-coated discs were evaluated using an optical stereo-microscope (Leica M205 C, Leica Microsystems GmbH, Switzerland). The wear scar depths were measured using a 3D confocal microscope (μsurf, Nanofocus AG, Germany). The mean depth of the 2D profile measured in the centre of the wear scar and perpendicular to the direction of movement was used as a reference value for the wear scar depth. 6 Results Representative friction curves of the different tested polymer coatings are shown in Figure 3. In tests with PAI 1, coefficient of friction had a starting value of approximately 0.07, but after about 18 minutes, it rose above a value of 1 (not seen in Figure 3 because the y-axis is scaled up to 0.3 only), which was the reason of a premature test termination. This increase in friction was the result of a coating failure and the corresponding adhesive wear which occurred afterwards on the exposed grey cast iron surfaces. Tests with PAI 2 and PAI 3 showed a pronounced running-in phase during which the coefficient of friction increased. After this phase, the coefficient of friction for both polymer coatings stabilised at a value of about 0.12. It should be noted that for PAI 2 the coefficient of friction has sometimes stabilised at a higher 15 Aus Wissenschaft und Forschung Table 1: Properties of the polymer coatings investigated in the present study Denomination PAI 1 PAI 2 PAI 3 PEEK PAEK Coating thickness (µm) 15-20 15-20 15-20 15-20 15-20 Base Polymer PAI PAI PAI PEEK PAEK* Solid Lubricant PTFE / SiO 2 PTFE / MoS 2 / PTFE / MoS 2 PTFE PTFE Graphite *Mixture of different PAEK polymers, including PEEK and PEK Figure 2: Schematic representation of the tribological experimental setup T+S_5_17 31.07.17 10: 58 Seite 15 16 Tribologie + Schmierungstechnik 64. Jahrgang 5/ 2017 value of about 0.2 (not shown in Figure 3), which indicates a less stable sliding behaviour for PAI 2 when compared to PAI 3. The starting coefficient of friction for PEEK was about 0.07 and it stabilised at a value of about 0.13. Thereafter, the coefficient of friction remained stable until the end of the test. The lowest coefficients of friction (around 0.06) were measured for PAEK where the friction curve was almost constant over the entire test duration. For both PEEK and PAEK coatings, no local friction fluctuations were observed, which strongly differentiates them from the behaviour of PAI coatings. Mean values of the measured wear scar depths of the coated samples (left yaxis) along with the stationary coefficient of friction values (right y-axis) are shown in Figure 4. No results are shown for PAI 1 because the coating was completely removed before the end of the test and scuffing (adhesive wear) of the grey cast iron surfaces occurred. However, for comparative purposes only, it should be noted that the wear scar depth on the PAI 1-coated disc was approximately 70 µm. From Figure 4, it can be seen that the highest wear scar depths were measured for PEEK and PAI 2, with values of 13 µm and 12 µm respectively. For PAI 3, the wear scar depth was slightly lower with a value of about 11 µm, and the smallest wear scar depth (approximately 3 µm) was measured for PAEK. It is interesting to note that the wear scar depths correlate well with the steady-state coefficient of friction: they both show a similar tendency. The observed correlation between wear and friction values shows that the friction is strongly dependent on the corresponding wear processes. Obviously, friction largely results from the energy loss associated with material removal (abrasive wear) and less from adhesive processes. Micrographs of the wear scars after the tribological tests are shown in Figure 5. It can be clearly seen in Figure 5 that PAI 1 is completely worn out. Cold welding was observed on both pin and disc. This confirms that after the coating was worn out, strong adhesion and scuffing occurred between the grey cast iron surfaces. After the tribological test, PAI 2 is heavily worn out: the original surface treatment under the coating can already be observed at the surface. The disc surface has been greatly smoothened and dark and bright areas are visible in the disc wear track. The bright areas are more heavily worn out regions; the dark areas are most likely to be attributed to transfer film formation (wear particles pressed onto the surface). Similar dark areas were also identified on the pin, particularly at its edges. A few wear traces oriented in the sliding direction can also be observed on the pin. The positions of these traces coincide with the dark transfer film areas observed on the disc surface, indicating that the regions related to the transfer film formation are topographically higher than the distinctly worn out bright regions. Wear scars of PAI 3 and PEEK look similar. For both coatings, it can be observed that surface smoothening mainly took place, whereby some abrasive traces and coating delamination could also be observed. The observed coating delaminations are more pronounced for PAI 3 than for PEEK. Furthermore for both PAI 3 and Aus Wissenschaft und Forschung Figure 3: Typical friction curves obtained from tested polymer coatings Figure 4: Mean values of the wear scar depth of the coated samples (left y-axis) and stationary coefficient of friction (right y-axis) of different polymer coatings T+S_5_17 31.07.17 10: 58 Seite 16 Tribologie + Schmierungstechnik 64. Jahrgang 5/ 2017 PEEK, less transfer film was formed on the pin in comparison to PAI 2. Nevertheless, dark horizontal lines oriented in the sliding direction can be observed on both pins, and their positions coincide with the coating delaminations observed on the disc surface. For PAEK, almost no wear occurred: only light smoothening of the surface topography is mainly present, while unworn coating topography is still visible in the surface valleys. The pin shows a negligible wear and almost no transfer film was observed. 7 Discussion A widely used method for describing friction mechanisms of polymer contacts differentiates between the two main mechanisms of polymer friction: adhesive and deformative friction [6- 8]. Figure 6 shows a schematic representation of the adhesive and deformative component of friction as a function of surface roughness [8]. From Figure 6, it can be seen that adhesive friction decreases with increasing surface roughness due to the reduced real contact area and the resulting lower amount of interaction sites. On the other hand, deformative friction increases due to an increasing contact pressure, which acts on a smaller real contact area. Deformative friction also increases with decreasing hardness of the polymer material because the counter-body surface asperities can penetrate deeper into the material. For polymer coatings in contact with rough grey cast iron counterparts, as those investigated in the present study, the deformative component of friction is dominant. Therefore, the coefficient of friction can be greatly reduced by decreasing the deformative component of the friction and this may be achieved by increasing the strength or hardness of the coating. Due to the semi-crystalline structure of the PAEKbased coatings, higher strength and cohesion of the polymer coating can be achieved than for the amorphous PAI, and therefore, PAEK-based coatings may be more suitable for demanding abrasive operating conditions as those considered in the present study. However, if a PAEK-based coating does not have a sufficient cohesion strength, as was the case for the PEEK coating evaluated in the present study, it performs less efficiently than PAI-based coatings, because it is worn out in discrete portions (due to its semi-crystalline structure, large flakes of material are torn out of the surface), which increases wear and at the same time friction due to a higher energy dissipation (higher deformative friction). 8 Conclusions Contrary to studies dealing with polymer coatings in contact with smooth counter-body surfaces in which PAI-based polymer coatings showed the best results due to their low adhesion and amorphous structure, it was shown in the present study that for polymer coatings sliding against rougher surfaces, the best results are obtained for higher strength PAEK-based coatings. Due to their semi-crystalline structure, PAEK-based coatings may reach higher strength values than the amorphous PAI-based coatings, making them more suitable for applications operating under more severe abrasive conditions. 17 Aus Wissenschaft und Forschung Figure 5: Micrographs of wear scars on the pins and discs Figure 6: Schematic representation of the adhesive and deformative component of friction as a function of surface roughness [8] T+S_5_17 31.07.17 10: 58 Seite 17 18 Tribologie + Schmierungstechnik 64. Jahrgang 5/ 2017 9 Acknowledgements The work presented was funded by the Austrian COMET Programme (Project XTribology, no. 849109) and was carried out at the “Excellence Centre of Tribology” (AC2T research GmbH) in cooperation with V-Research GmbH, Obrist Engineering GmbH and Linde AG. References [1] N.G. D EMAS , A.A. P OLYCARPOU : Tribological performance of PTFE-based coatings for air-conditioning compressors. Surface & Coatings Technology, 203: 307 - 316, 2008. [2] D. D ASCALESCU , K. P OLYCHRONOPOULOU , A.A. Polycarpou: The significance of tribochemistry on the performance of PTFE-based coatings in CO 2 refrigerant environment. Surface & Coatings Technology, 204: 319 - 329, 2009. [3] S.M. Y EO , A.A. P OLYCARPOU : Tribological performance of PTFEand PEEK-based coatings under oil-less compressor conditions. Wear, 296: 638 - 647, 2012. [4] S.M. Y EO , A.A. P OLYCARPOU : Micromechanical properties of polymeric coatings. Tribology International, 60: 198 - 208, 2013. [5] F. A USSERER , S. K LIEN , I. V ELKAVRH , P. F ORÊT , A. D IEM : Investigations of the sliding and wear behaviour in various gaseous atmospheres using a SRV testing apparatus. Tribologie und Schmierungstechnik, Vol. 63, No. 1: 22 - 28, 2016. [6] B.J. B RISCOE , D. T ABOR : Friction and wear of polymers, in: Polymer Surfaces, Ed. D.T. Clark, W.J. Feast, pp. 1 - 23. John Wiley, Chichester, UK, 1978. [7] N.K. M YSHKIN , M.I. P ETROKEVETS , A.V. K OVALEV : Tribology of polymers: Adhesion, friction, wear, and masstransfer. Tribology International, 38: 910 - 921, 2005. [8] G. E RHARD : Designing with Plastics (in German: Konstruieren mit Kunststoffen), 4. Edition. Carl-Hanser Verlag, Munich, Germany, 2008. Aus Wissenschaft und Forschung Themenverzeichnisse Tribologie · Schmierungstechnik Konstruktion · Maschinenbau · Tribologie · Verbindungstechnik · Oberflächentechnik · Werkstoffe · Materialbearbeitung · Produktion · Verfahrenstechnik · Qualität Fahrzeug- und Verkehrstechnik Elektrotechnik · Elektronik · Kommunikationstechnik · Sensorik · Mess-, Prüf-, Steuerungs- und Regelungstechnik · EDV-Praxis Im expert verlag erscheinen Fachbücher zu den Gebieten Weiterbildung - Wirtschaftspraxis - EDV-Praxis - Elektrotechnik - Maschinenwesen - Praxis Bau / Umwelt/ Energie sowie berufs- und persönlichkeitsbildende Audio-Cassetten und -CDs (expert audio ) und Software (expert soft ) Bitte fordern Sie unser Verlagsverzeichnis auf CD-ROM an! expert verlag Fachverlag für Wirtschaft & Technik Wankelstraße 13 · D-71272 Renningen Postfach 20 20 · D-71268 Renningen Baupraxis · Gebäudeausrüstung · Bautenschutz · Bauwirtschaft/ Baurecht Umwelt-, Energie- Wassertechnik · Hygiene / Medizintechnik Sicherheitstechnik Wirtschaftspraxis Anzeige Telefon (0 71 59) 92 65-0 Telefax (0 71 59) 92 65-20 E-Mail expert@expertverlag.de Internet www.expertverlag.de T+S_5_17 31.07.17 10: 58 Seite 18 Tribologie + Schmierungstechnik 64. Jahrgang 5/ 2017 1 Einleitung Das maschinelle Oberflächenhämmern (MOH, engl. Machine Hammer Peening - MHP) ist ein umformendes Fertigungsverfahren zur inkrementellen Oberflächenbehandlung von metallischen Werkstücken [BLEI12]. Durch hochfrequente Schläge eines sphärischen Hammerkopfes werden Rauheitsspitzen auf der Werkstückoberfläche umformtechnisch eingeglättet [STEI13a] und gleichzeitig eine Kaltverfestigung [SCHE13] und Druckeigenspannungen [BLEI13] in die Werkstückoberfläche eingebracht. Darüber hinaus können durch das MOH definierte Oberflächenstrukturen erzeugt werden, die aufgrund von elastohydrodynamischen Effekten reibungsmindernde Eigenschaften aufweisen [KLOC14]. Mithilfe des MOH kann die Oberfläche von Tiefziehwerkzeugen bearbeitet und gezielt tribologische Systeme eingestellt werden [TRAU16]. Die Arbeitsschritte Polieren und Schlichten können durch das MOH teilweise substituiert und die Prozesskette verkürzt werden [STEI13b]. Durch Adaption des Hammersystems auf konventionelle Fräsmaschinen können beliebige Freiformflächen bearbeitet werden [GROC12]. Aufgrund von elektrischen und pneumatischen Versorgungsleitungen am Hammerkopf kann die Motorspindel der Fräsmaschine nicht rotieren. Dadurch tritt eine Punktbelastung zwischen Spindellagerkugeln und Spindellagerring beim Hämmern auf. Als Resultat dieser Punktbelastung kann es zum Ermüdungsverschleiß der Lagerringe kommen, im Speziellen False-Brinelling [UPAD13]. False-Brinelling wird als Stillstandsmarkierung bezeichnet und ist auf Schwingungen zurückzuführen, welche Mikrobewegungen des Werkstoffs verursachen [SCHA00]. Im Bereich der Gleitbzw. Schlupfzone treten die größten Schädigungen auf, siehe Bild 1 links [GREB10]. Die Schädigungsarten sind triboche- 19 Aus Wissenschaft und Forschung * Robby Mannens, M.Sc. Dr.-Ing. Dipl.-Wirt.Ing. Daniel Trauth Dipl.-Ing. Dipl.-Wirt.-Ing. Jens Falker Dr.-Ing. Patrick Mattfeld, MBA Prof. Dr.-Ing. Christian Brecher Prof. Dr.-Ing. Dr.-Ing. E. h. Dr. h. c. Dr. h. c. Fritz Klocke Werkzeugmaschinenlabor WZL der RWTH Aachen 52074 Aachen Analyse der Spindellagerbelastung hinsichtlich des False-Brinelling-Effekts beim maschinellen Oberflächenhämmern R. Mannens, D. Trauth, J. Falker, P. Mattfeld, C. Brecher, F. Klocke* Eingereicht: 17. 11. 2016 Nach Begutachtung angenommen: 20. 12. 2016 Beim maschinellen Oberflächenhämmern auf Fräsmaschinen kann die Motorspindel aufgrund von elektrischen und pneumatischen Versorgungsleitungen am Hämmerkopf nicht rotieren. Dadurch kann es infolge der Punktbelastung zwischen Spindellagerkugeln und Spindellagerring zum Verschleiß der Lagerringe kommen. Gegenstand dieser Arbeit ist die Untersuchung des Einflusses der beim maschinellen Oberflächenhämmern wirkenden Reaktionskräfte auf den Spindeladapter und die Spindellager und den damit verbundenen False-Brinelling-Verschleiß der Spindellagerringe. Schlüsselwörter Maschinelles Oberflächenhämmern, Spindellager, False-Brinelling-Verschleiß, Reaktionskraft Because of electrical and pneumatical connections to the hammer head the motor spindle of a milling machine cannot rotate during machine hammer peening. Thus, wear on the bearing rings can occur because of point loads between the spindle bearing rolling elements and rings. Therefore, the topic of this work is the analysis of the resulting reaction forces during machine hammer peening on the spindle adapter and the spindle bearings and the associated false brinelling wear of spindle bearing rings. Keywords Machine Hammer Peening, spindle bearings, false brinelling wear, reaction force Kurzfassung Abstract T+S_5_17 31.07.17 10: 58 Seite 19