Tribologie und Schmierungstechnik
tus
0724-3472
2941-0908
expert verlag Tübingen
10.24053/TuS-2024-0003
51
2024
711
JungkFeasibility study on preloaded flexible conical plain bearings as wind turbine main bearings
51
2024
Jan Euler
Georg Jacobs
Timm Jokobs
Julian Röder
Wind turbines (WT) are a key technology towards a carbon neutral energy production worldwide. Currently about 30 % of the Levelized Cost of Electricity (LCoE) consist of service and maintenance. Critical components are the main roller bearings which have a failure probability between 15 and 30 % within 20 years. Main bearing replacements are expensive maintenance procedures because an external crane is needed which cost about $250.000 per day for offshore WTs. Hence, a discussed countermeasure within the wind industry is to use segmented plain bearings in future WTs. These types of bearings allow an up-tower sub-component wise replacement of faulty parts without the need to dismantle the whole drivetrain. Therefore, the Chair for Wind Power Drives (CWD) developed, tested and validated a segmented conical plain bearing concept for the use as a main bearings for WTs.
To function properly common plain bearings need a minimum clearance to allow the formation of a convergent lubrication gab. This initial clearance negatively influences run-out, as the shaft con move freely within the limits of the clearance. As WT drivetrain concepts become more integrated, main bearings for WTs need to fulfill higher requirements regarding the allowable run-out of the main shaft. Therefore, a fundamental design challenge arises for plain bearings as rotor main bearings. One approach to reduce run-out for roller main bearings is to use preloaded tapered roller bearings. For common plain bearings however preloading is not possible. However, the concept of preloading was successfully transferred to the flexible conical plain bearing concept developed at the CWD and the main shaft run-out severely reduced. In this work the feasibility of a preloaded flexible conical plain bearing as a WT rotor main bearing is evaluated and the advantages and disadvantages contrasted.
tus7110014
Science and Research 14 Tribologie + Schmierungstechnik · volume 71 · issue 1/ 2024 DOI 10.24053/ TuS-2024-0003 Feasibility study on preloaded flexible conical plain bearings as wind turbine main bearings Jan Euler, Georg Jacobs, Timm Jakobs, Julian Röder* submitted: 16.10.2023 accepted: 17.02.2024 (peer-review) Presented at the GfT Conference 2023 * Jan Euler, M. Sc. Prof. Dr.-Ing. Georg Jocobs Timm Jokobs, M. Sc. Julian Röder, M. Sc. CWD RWTH Aachen 52074 Aachen Windenergieanlagen (WEA) sind eine Schlüsseltechnologie auf dem Weg zu einer kohlenstoffneutralen Energieerzeugung. Derzeit entfallen etwa 30 % der Windenergie-Stromgestehungskosten auf Wartung und Instandhaltung. Eine kritische WEA-Komponente ist das Hauptlager, das innerhalb der üblichen Anlagenlebensdauer von 20 Jahren eine Ausfallwahrscheinlichkeit von 15 bis 30 % aufweist. Der Austausch der Hauptlager ist eine teure Wartungsmaßnahme, da ein externer Kran benötigt wird (~ 250.000 $ pro Tag bei Offshore-WEA). Eine in der Windindustrie diskutierte Maßnahme zur Reduktion der Wartungskosten ist die Verwendung von segmentierten Gleitlagern. Diese Art von Lagern ermöglicht den Austausch von beschädigten Segmenten, ohne, dass der gesamte Antriebsstrang demontiert werden muss. Deshalb hat der Chair for Wind Power Drives (CWD) ein segmentiertes konisches Gleitlagerkonzept für den Einsatz als WEA- Hauptlager entwickelt, getestet und validiert. Zur Ausbildung eines tragfähigen Schmierfilms, benötigen herkömmliche Gleitlager einen konvergenten Kurzfassung Schmierspalt und daher ein Mindestspiel. Dieses Mindestspiel wirkt sich negativ auf die Führungsgenauigkeit aus, da sich die Welle innerhalb der Grenzen des Spiels frei bewegen kann. Mit der zunehmenden Integration von WEA-Antriebskonzepten müssen WEA- Hauptlager höhere Anforderungen an die Führungsgenauigkeit der Hauptwelle erfüllen. Daraus ergibt sich für Gleitlager als Rotorhauptlager eine grundlegende konstruktive Herausforderung. Ein Ansatz zur Erhöhung der Führungsgenauigkeit bei Wälzlagern ist die Verwendung vorgespannter Kegelrollenlager. Bei herkömmlichen Gleitlagern ist eine Vorspannung nicht möglich. In dieser Arbeit wird ein vorgespanntes flexibles konisches Gleitlager als WEA-Hauptlager vorgestellt und bewertet werden. Schlüsselwörter Windenergie, Gleitlager, Hauptlager, Führungsgenauigkeit, Vorspannung Wind turbines (WT) are a key technology towards a carbon neutral energy production worldwide. Currently about 30 % of the Levelized Cost of Electricity (LCoE) consist of service and maintenance. Critical components are the main roller bearings which have a failure probability between 15 and 30 % within 20 years. Main bearing replacements are expensive maintenance procedures because an external crane is needed which costs about $250.000 per day for offshore WTs. Hence, a discussed countermeasure within the wind industry is to use segmented plain bearings in future WTs. These Abstract types of bearings allow an up-tower sub-component wise replacement of faulty parts without the need to dismantle the whole drivetrain. Therefore, the Chair for Wind Power Drives (CWD) developed, tested and validated a segmented conical plain bearing concept for the use as a main bearing for WTs. To function properly common plain bearings need a minimum clearance to allow the formation of a convergent lubrication gab. This initial clearance negatively influences run-out, as the shaft can move freely within 1 Introduction The European Union’s goal is to produce 40 % of its energy through renewables by 2030 [1]. In order to spur on the set up of more wind turbines (WT) the costs need to fall further. The main bearing of WTs is a crucial component regarding maintenance cost. Main bearings suffer from comparatively high failure probabilities of up to 30 % [2]. Currently mostly rolling bearings are commercially available as WT main bearings. In order to exchange a failed rolling main bearing the drivetrain of the WT needs to be disassembled. The exchange of these bearings results in high costs (about $250.000 per day for offshore WTs), as expensive equipment such as cranes or special maintenance vessels are required. Segmented plain bearings as main bearings for WTs are one possible solution to replace the error prone roller bearings. Segmented plain bearings can potentially be repaired up-tower as individual segments can be exchanged by hand or with the use of on-board cranes. One such segmented plain bearing concept was developed and validated at the CWD in the course of the WEA-GLiTS research project [3, 4]. The so-called FlexPad bearing was designed and validated as a main bearing for the Vestas V52 WT (750 kW). The flexible design allows the segments to follow the movement of the shaft, while maintaining parallel surfaces between the sliding segments and the shaft. This behaviour allows large areas for pressure build-up and prevents edge wear [3, 4]. The fundamental design of the FlexPad is shown in Figure 1. The design is characterized by the following key parameters: The cone shape for both bearing halves is determined by their angle of inclination α and the respective inner D i and outer diameter D o . The shape of the sliding segments (pads) and their support structure (arms) is mostly determined by their respective thickness (s pad and s arm ). Moreover, the flexibility of the concept is also characterized by the position x Groove and depth of the groove t groove in the arm. When transferring the concept towards a market relevant scale, changing requirements to the bearing’s performance need to be considered. One such requirement for main bearings is to limit the shafts movement to a specified maximum run-out. For increasing rated power there is a trend towards more integrated drivetrain concepts with high torque density. Typically, there is a high mechanical integration of main bearing, gearbox and generator [6-9]. Due to the integration of main bearing and gearbox the run-out of the main shaft is directly influencing the first gear stage. If the run-out is to large, this would negatively influence the gearbox performance. Common plain bearings are designed with nominal clearances of 0.3 ‰ to 3.5 ‰ of their nominal diameter [10-15]. Within the clearance the shaft is free in its radial movement. The requirement of integrated drivetrains for low run-out results in a need for high stiffness of the main bearing. Especially for the FlexPad concept this constitutes a challenge, as the flexibility of the support structure inherently results in an increased possible shaft run-out. In this work a preloaded FlexPad bearing is investigated and its performance is discussed for different clearance values under production load conditions. Science and Research 15 Tribologie + Schmierungstechnik · volume 71 · issue 1/ 2024 DOI 10.24053/ TuS-2024-0003 the limits of the clearance. As WT drivetrain concepts become more integrated, main bearings for WTs need to fulfill higher requirements regarding the allowable run-out of the main shaft. Therefore, a fundamental design challenge arises for plain bearings as rotor main bearings. One approach to reduce run-out for roller main bearings is to use preloaded tapered roller bearings. For common plain bearings however preloading is not possible. However, the concept of preloading was successfully transferred to the flexible conical plain bearing concept developed at the CWD and the main shaft run-out severely reduced. In this work the feasibility of a preloaded flexible conical plain bearing as a WT rotor main bearing is evaluated and the advantages and disadvantages contrasted. Keywords wind power, plain bearings, main bearings, pre-loading, run-out Figure 1: Schematic of the FlexPad concept with its key design parameters [5] side (GS) of the shaft. The RS and GS evaluation points are depicted in Figure 2 (right). They represent the centre points of their respective bearing halves. Furthermore, an evaluation of the hydrodynamic performance of the bearing was conducted for each simulation. 3 Results and discussion In the following chapter the simulation results will be presented and discussed. Focus is on radial run-out, its reduction through preload and the consequences for the hydrodynamic performance of the bearing. 3.1 Shaft run-out As a reduction of shaft run-out is the main motivation for clearance reduction, firstly the bearing’s run-out performance is evaluated for various amounts of positive clearance. In this first stage the bearings performance was investigated for clearances as low as 1 ‰ relative to Science and Research 16 Tribologie + Schmierungstechnik · volume 71 · issue 1/ 2024 DOI 10.24053/ TuS-2024-0003 2 Method The FlexPad design used in this contribution was already presented by Rolink et al. [5]. The bearing model is shown in Figure 2 (left). The main design parameters are shown in Table 1. In earlier works nominal clearances between 0.56 ‰ and 2.3 ‰ were investigated for the FlexPad concept [5, 16, 17]. In this work the clearance range is extended to also include preloading. The analysis of the effects of preloading is done via multi-body (MB), elasto-hydrodynamic (EHD) simulations. The model creation was conducted using the toolchain presented in [5]. The simulation setup is identical to [5, 16, 17]. The simulations were performed using the software FIRST. The performance of the bearing was analysed for the static operating conditions described in Table 2. For each bearing simulation the run-out was evaluated. The measured run-out describes the deflection of the shaft respective to its initial position. The run-out was evaluated for the rotor side (RS) and for the generator RS GS Figure 2: FlexPad MB-EHD-model (left), FlexPad shaft with highlighted RS and GS evaluation point for run-out [°] D o [mm] D i [mm] No. Pads [-] Span width [mm] x groove [mm] b groove [mm] t groove [mm] s arm [mm] s pad [mm] 46.7 473.7 256.7 12 249.1 69.7 8.1 9.2 15.2 20 Table 1: Reference design parameters of the investigated bearing rotor speed [rpm] F [kN] F [kN] F [kN] M [kNm] M [kNm] [‰] 28 29 26 2.6 -21 22 0 - 3.5 Table 2: Operating conditions during the EHD simulations the median diameter. The radial shaft run-out is shown in Figure 3. As expected the shaft run-out increases linearly with the bearing’s clearance. Greater clearance allows for greater shaft movement within the clearance before a hydrodynamically carrying lubrication film is formed. RS radial run-out is smaller than GS run-out. As the run-out decreases linearly with the initial clearance the lowest radial run-out (RS: 0.29 mm and GS: 0.46 mm) is achieved for an initial clearance of 1 ‰. The different scaling of the RS and GS radial run-out for increasing clearance stems from the conical design and the applied load conditions. For the investigated load condition with positive thrust forces the shaft experiences axial movement towards the GS. Due to the conical shape of bearing and shaft this axial movement results in a reduction of clearance on the RS and an increase in clearance on the GS. The increased clearance results in a greater run-out for the GS. Relative operational clearance gained for the GS is roughly equal to the initial clearance of the bearing. This is to be expected, as the shaft movement towards the GS is limited by the RS bearing halve. As positive axial loads are applied the shaft bridges the initial clearance on the RS, thus increasing the GS clearance by the same amount. Similar behaviour was also observed by Rolink for similar designs [18]. The clearance increase for the GS is slightly higher than the initial clearance, due to the flexibility of the bearing, which allows for additional axial shaft movement towards the GS. Preloading, which was achieved via interference fit is also possible for the FlexPad bearing. The negative clearance resulting from the interference fit is equivalent to a preloading of the bearing. The investigated preloads and their respective necessary negative clearance are shown in Figure 4. Preload increases linearly with the simulated negative clearance as the bearing is deformed elastically. For the maximum negative clearance of -1 ‰ the equivalent preload is 460 kN, whereas for -0.1 ‰ the equivalent preload is just 38 kN, which is in the order of magnitude of the applied load case. Preloading the bearing has a positive impact on the shaft run-out. In Figure 5 the radial shaft run-out is depicted for further reduced clearance and increasing preload. Investigated were clearances between the commonly employed clearance of 1 ‰ and -1 ‰, which results in preload due to the previously described interference fit. Reduction below 1 ‰ results in further linear reduced run-out up to zero clearance. Zero clearance in a basic radial plain bearing means that the surfaces of the bushing and the shaft fit together perfectly without contact. Thus, realistically zero clearance cannot be achieved for basic radial plain bearings due to e.g. tolerances and manufacturing inaccuracies. Zero initial clearance is possible for the FlexPad bearing, as during operation the flexible segments Science and Research 17 Tribologie + Schmierungstechnik · volume 71 · issue 1/ 2024 DOI 10.24053/ TuS-2024-0003 Figure 3: Radial run-out for RS and GS for varying clearances Figure 4: Corresponding preload for investigated negative clearances Figure 5: Radial run-out for increasing amounts of negative clearance same is true for the FlexPad concept for which this relationship was already shown by Rolink et al. in [16]. Figure 6 shows the maximum hydrodynamic pressure and the maximum specific pressure for varrying clearances. Maximum specific pressure is defined as the maximum of the normal forces of the most highly loaded segment devided by its surface area. The negative clearances correspond to the preloads depicted in Figure 4. As observed in previous studies, the maximum pressure is reduced for decreasing clearances (see Figure 6). The maximum hydrodynamic pressure decreases from 85 MPa for an initial clearance of 3.5 ‰ to 11 MPa at zero initial clearance. The lowest maximum pressure however, was determined for a negative clearance of -0.12 ‰ or a preload of 48 kN. For larger amounts of preload, the maximum pressure increases. It reaches 89 MPa for a negative clearance of -1 ‰ or a preload of 460 kN. For larger preloads, the bearing becomes hydrodynamically unfeasible. The same applies for the maximum specific pad pressure (see Figure 6). The maximum specific pad pressure decreases for decreasing clearance, from 3.15 MPa at 3.5 ‰ to 1.7 MPa at zero initial clearance. The minimum of 1.66 MPa is reached for a negative clearance of -0.12 ‰ or preload of 48 kN. For increasing preload, the maximum specific pad pressure increases linearly and reaches its maximum at 3.2 MPa for a preload of 460 kN or -1 ‰ negative clearance. To understand the reason for the improved bearing performance under preload, the pressure distribution needs to be evaluated. In Figure 7 the pressure distribution for the RS and the GS of the bearing are shown for a non-preloaded bearing design with 1 ‰ clearance and a preloaded design (-0.12 ‰ negative clearance or 48 kN preload). The applied load conditions are identical. As can be seen, regardless of preload, both bearing designs have hydrodynamic pressure build-up on all RS pads. This is caused by the positive thrust force on the bearing, which presses the RS shaft cone into the bearing, creating sufficiently small lubrication gaps and allowing for hydrodynamic pressure build-up. The maximum specific pad pressure on the RS however, is higher for the preloaded design. It reaches 1.2 MPa for the non-preloaded bearing and 1.5 MPa for the preloaded bearing. However, bearing wide maximum pressures and maximum specific pad pressures are reached at the GS side of the bearing. This occurs regardless of clearance and preload and is the result of the overall bearing design and load conditions. The pressure distribution on the GS is highly influenced by the amount of clearance. For the- Science and Research 18 Tribologie + Schmierungstechnik · volume 71 · issue 1/ 2024 DOI 10.24053/ TuS-2024-0003 bent so that the surface is parallel to the counterpart, which allows for desired hydrodynamic load bearing of the FlexPad bearing (see chapter 3.2). For preload the run-out is even further reduced. The shaft run-out decreases asymptotically towards 0.024 mm for the RS and 0.043 mm for the GS at -1 ‰ relative clearance. Thus, large preload decreases the run-out by 92 % for the RS and 91 % for the GS compared to a bearing with 1 ‰ clearance. For larger negative clearance than -1 ‰, the bearing ceases to function as the preloads become too large (see chapter 3.2). Radial run-out decreases asymptotically towards its minimum at -1 ‰ negative clearance or 460 kN preload. Due to fast asymptotical decrease for low amounts of preload, comparatively large amounts of run-out reduction are already possible for small preloads. 72 % of radial-runout reduction for the GS can already be achieved with -0.2 ‰ clearance (or 79 kN preload). The positive effect preloading has on the run-out of the FlexPad bearing is in its cause identical to preloading for tapered rolling bearings. Through preloading the flexible arms of the design are all symmetrically strained. As forces are applied to the shaft, inducing movement, the strain on the arms increases for some and decreases for others. If the shaft would be moved downwards, the downward forces exerted by the upper arms would decrease and the upward forces exerted by the lower arms would increase. The overall bearing therefore generates more upwardly directed forces per increment of downward movement as it would have without the preloaded condition. 3.2 Hydrodynamic performance Naturally the clearance has a significant impact on the hydrodynamic performance of the bearing. Depending on the bearing and on bearing type and design it was shown, that a reduced clearance results in a reduction of the maximum hydrodynamic pressures [19-21]. The Figure 6: Maximum hydrodynamic pressure and maximum specific pad pressure for varying clearances non preloaded bearing design only three pads experience significant hydrodynamic pressure build-up. These three pads subsequently need to carry nearly all the applied loads for the GS. For the preloaded bearing design, seven pads experience significant hydrodynamic pressure build-up. As the GS loads are spread more evenly this results in lower maximum pressures. The poor pressure distribution for non-preloaded bearing designs with clearance has two causes. The first cause is general in nature. An increase in clearance leads to a reduction of pressured area and an increase in maximum pressures. The second cause is specific to the FlexPad concept. Due to the axial shaft movement and subsequent clearance increase described in chapter 3.1 one bearing halve of the conical bearing design always has to operate effectively with double the amount of initial clearance. This is only further exacerbated by the flexible nature of the FlexPad concept, which allows for further axial movement and therefore clearance increase on the GS for the given load case. This phenomenon also explains the improved hydrodynamic behaviour of slightly preloaded bearing designs compared to a design with initial clearance of zero and no preload. The bearing’s flexibility allows for axial shaft movement even for a zero clearance designs. This axial movement leads to an effective clearance increase on the GS of the bearing, which in turn worsens the pressure distribution for this bearing halve. Through preloading the bearing becomes stiffer. This limits the axial shaft movement and maintains minimal amounts of clearance for both bearing halves. This results in an optimum of pressure distribution and maximum pressure. For the investigated design and load conditions this optimum is at 48 kN preload or -0.12 ‰ negative clearance. Although a further increase in preload would further stiffen the bearing and limit its axial movement, this would not lead to a further improvement of maximal pressures since the added loads that need to be hydrodynamically carried by the bearing - stemming from the preload - negate the positive effect regarding the pressure distribution. 3.3 Challenges As the presented study was conducted simulative, no design for a preloading mechanism exist at this stage. Initial preloading of the FlexPad bearing could be achieved via procedures typical for rolling bearings i.e. springs or nuts etc. Furthermore, it could be achieved via an intentional interference fit between conical shaft and bearing. This would require very accurate manufacturing of shaft and bearing to achieve the desired optimal performance. Alternatively, the thermal expansion of shaft and bearing during operations could be used. Differing thermal expansion coefficients or temperature can facilitate a clearance reduction for plain bearings under operating conditions [14, 15]. Therefore, operational preload via thermal expansion could be explored for the FlexPad bearing in future studies. As the FlexPad bearing is envisioned as a WT main bearing the whole bearing life span and its respective operating conditions need to be considered. Most of the operating life of the bearing would be within rated ope- Science and Research 19 Tribologie + Schmierungstechnik · volume 71 · issue 1/ 2024 DOI 10.24053/ TuS-2024-0003 RS GS Ψ=1 ‰, p max =25 MPa Ψ=-0.12 ‰, p max =10 MPa Preload No Preload R G Ψ Figure 7: Hydrodynamic pressure distribution for a non-preloaded (left) with clearance of 1 ‰ and preloaded bearing (right, -0.12 ‰ corresponds to 48 kN preload) ing reaches its lowest maximum pressure of 10.4 MPa and lowest maximum specific pressure of 1.7 MPa. In total it was shown, that the FlexPad concept can operate hydrodynamically for zero initial clearance and preload. Further studies need to investigate the optimal amount of preload for the whole range of operation conditions for the FlexPad bearing and the best method to apply preload. Furthermore, the feasibility of lubrication film build-up under preloaded conditions needs to be experimentally validated. References [1] EUROPEAN PARLIAMENT: Amendments adopted by the European Parliament on the Renewable Energy Directive (in Kraft getr. am 2022) (2022) [2] HART, Edward; TURNBULL, Alan; FEUCHTWANG, Julian; MCMILLAN, David; GOLYSHEVA, Evgenia; ELLIOTT, Robin: Wind turbine main-bearing loading and wind field characteristics. In: Wind Energy 22 (2019), Nr. 11, S. 1534-1547 [3] ABSCHLUSSBERICHT: Thermisch gespritzte Gleitlagerbeschichtungen für Hauptlager von Windenergieanlagen (WEA) - WEA Triebstrang und Oberflächentechnik, eng.: “Final Report, WEA-GLiTS”: Förderkennzeichen: 03EK3036A [4] SCHRÖDER, Tim Niklas: Konisches Gleitlager für die Rotorlagerung einer Windenergieanlage, eng: “Conical Sliding Bearing for the Rotor Main Bearing of a Wind Turbine” (2021) [5] ROLINK, Amadeus; JACOBS, Georg; PÉREZ, Alex; BOSSE, Dennis; JAKOBS, Timm: Sensitivity analysis of geometrical design parameters on the performance of conical plain bearings for use as main bearings in wind turbines. In: Journal of Physics: Conference Series 2265 (2022), Nr. 3, S. 32010 [6] NEJAD, Amir R.; KELLER, Jonathan; GUO, Yi; SHENG, Shawn; POLINDER, Henk; WATSON, Simon; DONG, Jianning; QIN, Zian; EBRAHIMI, Amir; SCHELENZ, Ralf; GUTIÉRREZ GUZMÁN, Francisco; CORNEL, Daniel; GOLAFSHAN, Reza; JACOBS, Georg; BLOCK- MANS, Bart; BOSMANS, Jelle; PLUYMERS, Bert; CARROLL, James; KOUKOURA, Sofia; HART, Edward; MCDONALD, Alasdair; NATARAJAN, Anand; TORSVIK, Jone; MOGHADAM, Farid K.; DAEMS, Pieter-Jan; VERSTRAETEN, Timothy; PEETERS, Cédric; HELSEN, Jan: Wind turbine drivetrains: state-ofthe-art technologies and future development trends. In: Wind Energy Science 7 (2022), Nr. 1, S. 387-411 [7] NEJAD, Amir R.; TORSVIK, Jone: Drivetrains on floating offshore wind turbines: lessons learned over the last 10 years. In: Forschung im Ingenieurwesen 85 (2021), Nr. 2, S. 335-343 [8] EVOLUTION ONLINE: New challenges for rotor bearings in the 8-MW offshore category | Evolution. URL https: / / evolution.skf.com/ new-challenges-for-rotor-bear ings-in-the-8-mw-offshore-category/ . - Aktualisierungsdatum: 2020-04-01 - Überprüfungsdatum 2023-04-26 [9] SKF. URL https: / / www.skf.com/ dk/ news-and-events/ news/ 2019/ 2019-10-08-wind-turbine-main-shaft-bearing -design-considerations. - Aktualisierungsdatum: 2023- 04-27 - Überprüfungsdatum 2023-05-03 Science and Research 20 Tribologie + Schmierungstechnik · volume 71 · issue 1/ 2024 DOI 10.24053/ TuS-2024-0003 ration of the WT. The applied loads and rotational speed would therefore be in the range of the investigated load condition. However, turbine start up and extreme loads also need to be considered in the future. The applied preload would further increase the breakaway toque during turbine start-up. This could hamper the turbines ability to idle or start at low wind speeds. Furthermore, the increased load at the start of the turbine could increase time spent in the mixed friction regime before full hydrodynamic load bearing is reached and therefore increase wear. On the other hand, transition speed according to Vogelpohl would theoretically be lowered through the near zero clearance produced by the preload [10, 22]. The general introduction of lubricant within the zero or near zero clearance is as of yet untested through experiments or simulations and needs to be further investigated. 4 Conclusion The FlexPad concept is a promising plain bearing design for future WT main bearings. However, large scale WTs mostly favour highly integrated drivetrain concepts which necessitate strict limitations on maximum main shaft run-out. Plain bearings usually operate with a clearance, which allows free shaft movement within its limits. This creates a conflict between the requirements for main bearings of integrated WTs and the design possibilities with common plain bearings. For rolling bearings, preloading is a common practice to increase stiffness and limit shaft run-out. In the course of this work the effects of clearance reduction and preload were investigated for the FlexPad concept under static operational conditions. Preload was realised via an interference fit resulting in a negative clearance between shaft and bearing. The results of this study show that the stiffness of the FlexPad bearing can indeed be increased through preload while maintaining its hydrodynamic load bearing capability. This also leads to a reduction in shaft run-out for the investigated load conditions. GS run-out reductions of up to 91 % relative to the reference with 1 ‰ relative clearance could be achieved via preload. Too large preloads lead to a breakdown of the hydrodynamic capability of the bearing. It was further discovered, that the bearings hydrodynamic performance could be increased via preloading. This is due to the flexible nature of the FlexPad and its conical design. Preload reduces the clearance increase for the GS through axial shaft movement. This allows for a greater load carrying area on the GS and reduces the overall maximum pressure and maximum specific pressure. For the presented bearing design and load condition an optimal amount of preload was identified with regards to the hydrodynamic performance. The bearing shows optimal hydrodynamic performance for a preload of 48 kN which corresponds to -0.12 ‰ negative clearance. For this design the bear- [10] LANG STEINHILPER: Gleitlager, 1978 [11] WITTEL, Herbert; MUHS, Dieter; JANNASCH, Dieter; VOßIEK, Joachim: Roloff/ Matek Maschinenelemente: Normung, Berechnung, Gestaltung. 22., überarbeitete und erweiterte Auflage. Wiesbaden: Springer Vieweg, 2015 [12] JACOBS, Georg (Hrsg.): Maschinengestaltung. Ausgabe 10/ 2016. Aachen: Mainz, 2016 [13] DIN 31652-3. Januar 2017. Gleitlager - Hydrodynamische Radial-Gleitlager im stationären Betrieb - Teil 3: Betriebsrichtwerte für die Berechnung von Kreiszylinderlagern [14] GROTE, Karl-Heinrich (Hrsg.): Dubbel: Taschenbuch für den Maschinenbau, mit …Tabellen. 23., neu bearb. u. erw. Aufl. Berlin, Heidelberg: Springer, 2011 [15] NIEMANN, Gustav; WINTER, Hans; HÖHN, Bernd- Robert; STAHL, Karsten: Maschinenelemente 1: Konstruktion und Berechnung von Verbindungen, Lagern, Wellen. 5., vollständig überarbeitete Auflage. Berlin, Heidelberg: Springer Vieweg, 2019 (Springer eBooks Computer Science and Engineering) [16] ROLINK, Amadeus; JACOBS, Georg; MÜLLER, Matthias; JAKOBS, Timm; BOSSE, Dennis: Investigation of manufacturing-related deviations of the bearing clearance on the performance of a conical plain bearing for the application as main bearing in a wind turbine. In: Journal of Physics: Conference Series 2257 (2022), Nr. 1, S. 12006 [17] ROLINK, Amadeus; JACOBS, Georg; SCHRÖDER, Tim; KELLER, Dennis; JAKOBS, Timm; BOSSE, Dennis; LANG, Jochen; KNOLL, Gunter: Methodology for the systematic design of conical plain bearings for use as main bearings in wind turbines. In: Forschung im Ingenieurwesen 85 (2021), Nr. 2, S. 629-637 [18] ROLINK, Amadeus; SCHRÖDER, Tim; JACOBS, Georg; BOSSE, Dennis; HÖLZL, Johannes; BERG- MANN, Philipp: Feasibility study for the use of hydrodynamic plain bearings with balancing support characteristics as main bearing in wind turbines. In: Journal of Physics: Conference Series 1618 (2020), Nr. 5, S. 52002 [19] HAGEMANN, Thomas; DING, Huanhuan; RADTKE, Esther; SCHWARZE, Hubert: Operating Behavior of Sliding Planet Gear Bearings for Wind Turbine Gearbox Applications—Part I: Basic Relations. In: Lubricants 9 (2021), Nr. 10, S. 97 [20] KUZNETSOV, Evgeny; GLAVATSKIH, Sergei; FIL- LON, Michel: THD analysis of compliant journal bearings considering liner deformation. In: Tribology International 44 (2011), Nr. 12, S. 1629-1641 [21] SCHILLING, Gregor; LIEBICH, Robert: The Influence of Bearing Clearance on the Load Capacity of Gas Polymer Bearings. In: Applied Sciences 13 (2023), Nr. 7, S. 4555 [22] G. VOGELPOHL: Geringste zulässige Schmierschichtdicke und Übergangsdrehzahl (1962) Science and Research 21 Tribologie + Schmierungstechnik · volume 71 · issue 1/ 2024 DOI 10.24053/ TuS-2024-0003
