1 Introduction A screaming, tormenting screech - many connect the sound of the dentist’s drill to aching teeth. Root canal treatment is not a pleasant experience in itself. Additional high-frequency sounds of the supposed torture tool squeaking at the patient’s eardrum inevitably lead to a negative connotation of the dentist and the dental devices. Aus Forschung und Praxis 15 Tribologie + Schmierungstechnik · 67. Jahrgang · 2/ 2020 DOI 10.30419/ TuS-2020-0009 Innovations in the Development of Miniature Ball Bearings - Expanding the Performance Limits of Dental Turbine Bearings Thomas Röper, Christoph Bayer, René Büchner, Holger Schurz, Herbert Niedermeier, Thomas Kreis* Die Lagerung einer Dentalturbine ist hohen Beanspruchungen ausgesetzt. Unter anderem müssen die Miniaturkugellager Drehzahlen bis zu 500.000 min -1 , Mangelschmierung (mitunter Trockenlauf), periodischen Heißdampfsterilisationen und Lager-Schiefstellung standhalten. Dabei wird die Lebensdauer eines Dentalkugellagers primär durch den Kugellagerkäfig limitiert. Einen Erklärungsansatz für diese Charakteristik liefert die Betrachtung des Dentalkugellagers als tribologisches System. Jenes Tribosystem wird neben den Reibpartnern Wälzkörper und Laufbahnen vor allem durch den Einfluss des Kugellagerkäfigs bestimmt, der an den Führungsflächen im Kontakt Käfig-Kugel beziehungsweise Käfig-Ringschulter höchste tribologische Beanspruchungen erfährt. Idealerweise werden die jeweiligen Reibflächen durch einen Schmierstoff getrennt. Bleibt dieser aus, unterliegt ein Standard-Käfig hohem abrasivem Verschleiß, der in Materialabtrag, Schwingungen und Erwärmung bis hin zum Bruch resultiert. Durch die Wechselwirkung des Käfigs mit dem Kugelsatz übertragen sich die Verschleiß-Auswirkungen auf das Gesamtsystem und führen in letzter Konsequenz zum Ausfall des Kugellagers und somit zum Stillstand der Dentalturbine. Diesen Herausforderungen begegnet GRW mit den Innovationen XTRAlon ® und 3-Radien-Profil. Schlüsselwörter Dentalturbine; Hochtemperaturbeständigkeit; Kippkompensation; Kugellager; Laufbahn; Schiefstellung; Sterilisation; Trockenlauf Abstract: Bearings of a dental turbine are exposed to high stresses. Among other things, the miniature ball bearings have to withstand speeds of up to 500,000 rpm, insufficient lubrication (sometimes dry running), periodic superheated steam sterilizations and bearing misalignment. The life of a dental ball bearing is primarily limited by the ball bearing cage. An explanation for this characteristic is the consideration of the dental ball bearing as a tribological system. In addition to the friction partners rolling elements and raceways, this tribosystem is primarily determined by the influence of the ball bearing cage, which experiences the highest tribological stresses on the guide surfaces in the contact zone cage - ball or cage - ring shoulder. Ideally, the respective friction surfaces are separated by a lubricant. If the lubricant is lacking, a standard cage is subject to high abrasive wear, resulting in material removal, vibration and heating all the way to breakage. Due to the interaction of the cage with the ball set, the wear effects are transmitted to the entire system and ultimately lead to failure of the ball bearing and thus to a standstill of the dental turbine. GRW meets these challenges with the innovations XTRAlon ® and ‘three-way’ profile. Keywords ball bearing; dental turbine; dry running; high temperature strength; raceway; sterilization; tilt compensation; tilting Kurzfassung Abstract * Thomas Röper, B.Eng. Orcid-ID: https: / / orcid.org/ 0000-0001-5802-3975 Dipl.-Ing. (FH) Christoph Bayer Dipl.-Ing. (FH) René Büchner Holger Schurz, Mechanical Engineering Technician Dipl.-Ing. (FH) Herbert Niedermeier Dipl.-Ing. (FH) Thomas Kreis Gebr. Reinfurt GmbH & Co. KG; D-97222 Rimpar TuS_2_2020.qxp_T+S_2018 04.06.20 14: 11 Seite 15 the outer rings floatingly in the housing. This system results in a vibration damping X arrangement, which can compensate for axial shock loads to a certain extent. A disadvantage of this bearing system is an uneven load distribution. Due to the asymmetric external load of the dental tool, the radial deflections of the hollow shaft are mainly absorbed by the front (left) bearing. Thus, the use of the dentist’s drill is not only a challenge for patients. The high-speed ball bearings are also challenged with relatively high loads as well as changing bending moments of the hollow shaft. 1.3. Failure mechanisms of turbine bearings The external conditions in the dental turbine application pose a great challenge to any deep groove ball bearing. After each dental treatment, the handpiece is cleaned, oiled and sterilized in an automated purification process [3]. Despite the maintenance interval, a lack of lubrication occurs after a short time, since the exhaust air of the turbine escapes not only via the exhaust air duct of the handpiece but also via the ball bearings. As a result, the Aus Forschung und Praxis 16 Tribologie + Schmierungstechnik · 67. Jahrgang · 2/ 2020 DOI 10.30419/ TuS-2020-0009 1.1. Dental handpiece - a high-tech system The so-called dental handpiece is part of the basic equipment of every dental surgery. It incorporates the various tools necessary for dental care. Figure 1 shows a selection of commercially available variations. The grip sleeve is equipped with a grip profile, which ensures ergonomic gripping of the instrument. In addition to the ability to illuminate the mouth with the help of a light eye, it can be cleaned and cooled with spray nozzles. A push button at the top of the head allows quick replacement of the tool. Inside the head is the dental turbine, which is pneumatically accelerated to up to 500,000 revolutions per minute. Thereby the blades of the rotor are impinged by the compressed air, thus experiencing a driving force. Via the lever - between force application and shaft center - the resulting drive torque produces a rotational movement of the rotor shaft. Due to the clamping function of the shaft, which is necessary to hold and fixate the dental tool, the torque is transferred to that tool represented by a cylindrical pin in Figure 1. In practice, a variety of tools such as drills and milling cutters are clamped. On the other end of the handpiece is the connection for the compressed air supply. The necessary airflow is supplied by a conduit system in the handle sleeve and then discharged through a likewise integrated, parallel pipe. [2] 1.2. Supporting the dental turbine - high precision miniature ball bearings For the subsequent consideration of the turbine bearing, the focus is on the internal construction of the handpiece head. Figure 2 shows the relevant designations as well as standard dimensions. Two bearing assemblies enable the required positioning and stabilization of the hollow shaft in the housing head of the dental handpiece. The ball bearings are made of high quality inner and outer rings, steel or ceramic balls (Ø 1 mm) and cages made of high-performance plastics. As a rule, the bearings are completed by covers, which reduce leakage of the medically appropriate lubricant from the ball bearing and entry of potential dirt particles into the bearing. This protective function is adopted in the model in Figure 2 by a modified face side of the outer ring. Typically, the inner rings are frictionally connected to the shaft. O-rings (radial) and wave washers (axial) support Figure 1: Dental handpieces with exposed turbines [1] Figure 2: Half-section through the housing head of a dental handpiece [1] TuS_2_2020.qxp_T+S_2018 04.06.20 14: 11 Seite 16 maintenance lubricant is discharged. Another negative influence is the asymmetric external load, which is transferred to the cutting edge of the drilling tool during dental treatment, which leads to a tilt of the rotor shaft. In conjunction with a misalignment of the bearings - due to an obliquely mounted outer ring - the tilt affects the performance of the ball bearing. According to the state of the art, deep groove ball bearings running under tilt are exposed to considerable constraining forces. The balls move on elliptical orbits. Depending on the position on the ellipse, they are driven at different circulation speeds. Due to fast or slowly moving rolling elements the ball bearing cage undergoes a permanent alternating stress. The consequence is an overload of the cage and thus a rapidly progressing cage wear, resulting in early failure of the ball bearing. This effect can be partially compensated by modern high-performance plastics mixed with friction-reducing additives. Progressive wear leads to debris, which accumulate inside the bearing. A loss of speed arises from increased friction. Immediately before the failure of the higher loaded front bearing, the handpiece is perceived as loud and droning (due to significant speed fluctuations during operation). At the same time, unpleasant vibrations are transmitted to the grip sleeve of the handpiece. 2 Materials and Methods In order to meet these challenges, two approaches were pursued. One was the development of a suitable cage material. The second approach considered the geometrical influence of the raceway on the kinematics. 2.1. Approach 1: Cage material XTRAlon ® As part of a project funded by the Bavarian Ministry of Economic Affairs “High performance polymer for ball bearing cages”, a novel high performance plastic, consisting of the base polymer polyamide-imide (PAI) and the additive polytetrafluoroethylene (PTFE), was developed over a period of three years. [4] During production, PTFE is chemically coupled to the base polymer PAI via reactive extrusion. An upstream radiation modification of the PTFE micro powder leads to the generation of the functional groups required for the chemical coupling to the PAI matrix. [5] The aim of the project was to find optimal material properties with regard to the PTFE content in the PAI and with regard to the thermal follow-up treatment process (polycondensation). The PAI / PTFE mixing ratios were set between 75/ 25 and 90/ 10 in the test series. Furthermore, a total of five different PTFE micro powders from the emulsion polymer and suspension polymer groups were selected for the compound production. Finally, different radiation sources were used for the radiation modification of the PTFE powder. Gamma radiation was used in addition to electron radiation. The radiation dose varied between 250 and 1000 kGy. [6] 2.1.1. Validation The key parameter “service life”, decisive for the optimization of the material, is determined on a GRW dental test bench with test turbine handpieces (Figure 3 a). By means of a loading cylinder the test bench simulates the loads applied by a dentist on the drilling tool of the dental instrument. The idling speed, measured immediately after the load cycle, serves as assessment criterion for the point of failure of the dental turbine. In addition to service life tests, further experiments are carried out to assess material characteristics. One of them is the burst test (Figure 3 b), conducted with defined test bodies. It provides characteristic values regarding the elasticity behavior of the material in form of a force- Aus Forschung und Praxis 17 Tribologie + Schmierungstechnik · 67. Jahrgang · 2/ 2020 DOI 10.30419/ TuS-2020-0009 Figure 3: Test setup for service life (a) and burst tests (b) [1] (a) (b) TuS_2_2020.qxp_T+S_2018 04.06.20 14: 11 Seite 17 The following materials are used in the ball bearings of the dental turbine: • rings: martensitic stainless steel X65Cr13 • balls: silicon nitride (Si3N4) • cage: polyamide-imide (PAI) For the simulation, the housing and the balls are designed as rigid bodies. For the entire rotor shaft, the inner and outer ring and the cage a linear elastic material model is selected. The outer rings are connected to the housing via O-rings. This joint is based on a hyperelastic material model. In order to determine the Mooney-Rivlin material parameters, a comparison between simulation and real properties was carried out in preliminary tests. The complete model of the dental turbine including the bearings is shown in Figure 5. It consists of 545,477 nodes, 571,042 elements and 37 components. Overall, 54 main contact pairs arise out of the model. For the FE (finite elements) computation, defined coefficients of friction were chosen for the following friction partners: Aus Forschung und Praxis 18 Tribologie + Schmierungstechnik · 67. Jahrgang · 2/ 2020 DOI 10.30419/ TuS-2020-0009 deflection graph. At a material testing machine the test body is deformed in the bore by means of a metal cone until it breaks. 2.2. Approach 2: Raceway design ‘three-way’ The structure of the dental turbine (Figure 2) leaves no room for geometric adjustments in the ball bearing periphery. Due to the floating mounting of the two outer rings in the receptacle, tilting of the rings cannot be completely prevented. If this happens, the turbine will inevitably fail earlier so that the warranty period granted by the dental equipment manufacturer cannot be achieved reliably. The described problem is solved with ‘three-way’ a new raceway design of the outer ring. Figure 4 shows the shape of the new raceway design. In regular installation position - without tilting - the performances of standard deep groove ball bearings and three-way deep groove ball bearings do not differ. Since they both share the raceway radius R1 the rolling conditions of the respective ball set, moving within the limits of the operating contact angle ± α, is identical. Neglecting tilting, both ball bearing variants have equal static and dynamic load ratings. The kinematic operating principle of the three-way profile can be described as follows. As soon as a defined tilt clearance sets in between the inner ring and the outer ring, individual balls of the ball set move into the second raceway area (R2). There, the raceway radius R2 is significantly larger than R1 (about twice as large as R1), so that the operating contact angle shifts towards the center of the raceway. This minimizes the otherwise highly elliptical character of the ball set path. Ultimately, the constraining forces between balls and raceways or balls and cage are significantly reduced. The validation of the positive effects of the ‘three-way’ design is described below using advanced dynamic simulation as well as a comparative noise test. 2.2.1. Validation Based on the explicit dynamic method in LS-DYNA, a dynamic model of a turbine bearing system is modeled. The dynamic vibrations of the bearings are simulated for the standard deep groove ball bearing and the ‘three-way’ deep groove ball bearing at combined loads and misalignment of the outer ring. In the course of the simulation, the effects on the behavior of the cage and the ball / raceway contact are analyzed. The cage is considered the weakest component regarding dental ball bearings. Hence, a direct correlation between dynamic vibrations and cage life is assumed. Figure 4: ‘three-way’ profile of the outer ring [1] Figure 5: Mesh depiction of the complete model [1] TuS_2_2020.qxp_T+S_2018 04.06.20 14: 11 Seite 18 • stiction between balls and raceways • stiction between balls and cage • dynamic friction of aforementioned contacts • friction between housing, O-ring and bearing The axial adjustment of the bearings takes place on the end face of the outer rings with one-dimensional springs and a spring rate of 10 N/ mm. The rotational speed is transmitted to the rotor shaft between the two bearings. Figure 6 shows the time course for the activation of preload (spring force), rotation (rotational speed) and load (external force application). This takes place consecutively, taking into account a certain settling time. Furthermore, noise measurements are performed on a GRW noise tester (Figure 7). An accelerometer picks up the vibrations on the outer ring of the deep groove ball bearing. A hydrodynamically supported shaft drives the inner ring at 3,000 revolutions per minute. piezo crystal converts the detected acceleration values into an electrical voltage. Corresponding band filters divide the frequencies into the ranges 500 to 1,600 Hz (LOW) and 1,600 to 5,000 Hz (HIGH). Occurring signal peaks are integrated over time and displayed as “PEAK”. [1] 3 Results Both approaches seem to influence the properties of the ball bearing. The associated results as well as their significance for the application are presented below. 3.1. XTRAlon ® The outcome of the project is a tailor-made high-performance plastic (XTRAlon ® ), characterized by excellent values in terms of heat resistance, notched impact strength, friction and wear. XTRAlon ® gives the cage a service life up to five times longer compared to commercial polyamide-imide products (with physically mixed-in additives). Figure 8 shows the results of the service life test, in which the ball bearings are initially lubricated once with a polyurea-thickened grease without subsequent re-lubrication. The difference in performance is even more pronounced when the two cage materials XTRAlon ® and PAI modified are operated in an absolutely dry, oil-free ball bearing within the dental turbine. While the ball bearing with XTRAlon ® cage has a service life of up to 15 hours, the standard bearing fails after just a few minutes. The excellent impact strength of XTRAlon ® also results in the cage behaving in a “good-natured” way in situations where external strains, such as bearing tilt, Aus Forschung und Praxis 19 Tribologie + Schmierungstechnik · 67. Jahrgang · 2/ 2020 DOI 10.30419/ TuS-2020-0009 Figure 8: Service life with initially lubricated dental turbine bearing [1] Figure 7: Noise test spindle with attached ball bearing and accelerometer [1] Figure 6: Time course for the activation of preload (spring force), rotation (rotational speed) and load (external force) [1] TuS_2_2020.qxp_T+S_2018 04.06.20 14: 11 Seite 19 act on the cage. A capital bearing failure, e.g., through a cage back break on the snap cage, can be completely avoided. Figure 9 shows the strength-displacement curve of the cage materials. In comparison to PAImodified, XTRAlon ® offers a five times higher breaking load with maximum elasticity. Thanks to its homogenous structure and the dense distribution of the PTFE bonding groups, XTRAlon ® offers significantly improved resistance against the effects of hot steam. SEM images of cage surfaces after 1,000 cycles of hot steam sterilization illustrate the difference between the conventional and the new material (Figure 10). Some heavily fissured areas can be seen on the standard cage’s transition between its plane surface and bore (Figure 10 a). The test cage made of XTRAlon ® , however, emerged from the steam pressure sterilization without any surface damage (Figure 10 b). Figure 11 shows the suitability of high performance cage materials regarding different performance features. XTRAlon ® speaks for itself. 3.2. ‘three-way’ profile Figure 12 shows the energy density of the ball set’s surface totaled over a specific time. Without tilting, a low-energy state is set for both the standard bearing and the ‘three-way’ bearing (3 % of the maximum value). Tilting however leads to a significant increase in the energy density in both cases, with the difference that the Aus Forschung und Praxis 20 Tribologie + Schmierungstechnik · 67. Jahrgang · 2/ 2020 DOI 10.30419/ TuS-2020-0009 Figure 9: Impact strength and displacement range for PAImodified and XTRAlon ® [1] Figure 11: Performance features: XTRAlon ® , PAI mod., PEEK mod., PPS mod. and phenolic resin [1] Figure 10: SEM surface of PAImodified (a) and XTRAlon ® (b) after hot steam sterilization [1] (a) (b) TuS_2_2020.qxp_T+S_2018 04.06.20 14: 11 Seite 20 ‘three-way’ bearing with 78 % energy density performs much better compared to the standard ball bearing (100 %). The more favorable energy balance suggests that the vibrations transmitted to the cage are lower. For verification, the radial cage movements for both raceway designs were traced in the tilted state (Figure 13). Figure 13 a (standard bearing) demonstrates a cage movement induced by the ball set that follows no concrete pattern. This irregular run results in high cage wear the ‘three-way’ ball bearing with its special raceway. The GRW noise test and the advanced dynamic simulation prove that a tilted mounting can be compensated. XTRAlon ® gives the cage a service life up to five times longer compared to commercial polyamide-imide products (with physically mixed-in additives). In combination with the ‘three-way’ profile, a ball bearing is realized that promises safe and smooth running even in adverse environmental conditions. 5 Patents ‘three-way’ profile • EP000003316815B1 XTRAlon ® • DE102006030836B4 • EP000002035721B1 • CN000101484715B • BR000PI0714201B1 • CN000101484715B • US000009702408B2 Aus Forschung und Praxis 21 Tribologie + Schmierungstechnik · 67. Jahrgang · 2/ 2020 DOI 10.30419/ TuS-2020-0009 Figure 12: Surface Energy Density of the ball set [1] Figure 13: Radial cage deflections of the standard (a) and the ‘three-way’ (b) ball bearing [1] and unpleasant noise. The radial movement pattern of the cage in the ‘three-way’ bearing (Figure 13 b) follows a sinusoidal course. This generates significantly less cage wear, positively affecting bearing life and running noise. The observations of the simulation are confirmed in practice. During the noise test, the standard and ‘threeway’ bearings demonstrate an equivalent behavior when not tilted. However, as soon as the respective outer ring is brought into a tilted state, the two ball bearing versions differ significantly. A much smaller increase in running noise is recorded while testing the ‘three-way’ bearing. 4 Discussion The new ball bearing design ‘three-way’ can be used successfully in applications in which the installation situation offers room for misalignment or tilting of the bearings. Standard deep groove ball bearings are very sensitive to the described influences. Strongly scattering contact forces that are transmitted from the ball set to the cage lead to chaotic radial cage movements, with the consequence that the cage is caused to oscillate leading to high wear and tear. Tilted running standard bearings are perceived as noisy and fail much earlier. A remedy is TuS_2_2020.qxp_T+S_2018 04.06.20 14: 11 Seite 21 [3] Kommission für Krankenhaushygiene und Infektionsprävention beim Robert Koch Institut, “Infektionsprävention in der Zahnheilkunde - Anforderungen an die Hygiene. Mitteilung der Kommission für Krankenhaushygiene und Infektionsprävention beim Robert Koch Institut,” Bundesgesundheitsblatt - Gesundheitsforschung - Gesundheitsschutz, no. 4, p. 381, 2006. [4] D. Lehmann, Dokumentation zum Lizenzvertrag vom 12.06.2006/ 15.06.2006, Dresden, Germany: Leibniz-Institut für Polymerforschung Dresden e.V., 2006. [5] M. Heller, “hdk-dresden.de,” Leibniz-Insititut für Polymerforschung Dresden e.V., Hohe Str. 6, 01069 Dresden, www.ipfdd.de, 10 May 2011. [Online]. Available: http: / / www.hdk-dresden.de/ dokumente/ kunststoffrecycling_2011/ 07_Vortrag_Heller.pdf. [Accessed 12 8 2019]. [6] D. L. Herbert Niedermeier, “Wälzlagerkäfig”. Germany Patent WO2008003466A1, 10 January 2008. Aus Forschung und Praxis 22 Tribologie + Schmierungstechnik · 67. Jahrgang · 2/ 2020 DOI 10.30419/ TuS-2020-0009 Acknowledgments Many thanks to CADFEM GmbH for supporting us with the advanced dynamic simulation. Conflicts of Interest The authors declare no conflict of interest. References [1] Gebr. Reinfurt GmbH & Co. KG, own research, Rimpar, Germany. [2] Deutsche Gesellschaft für Sterilgutversorgung (DGSV ® ) e. V., “Übertragungsinstrumente - Hand- und Winkelstücke und Turbinen - Teil 1,” Zentralsterilisation, pp. 69- 70, January 2012. TuS_2_2020.qxp_T+S_2018 04.06.20 14: 11 Seite 22