International Colloquium Tribology
ict
expert verlag Tübingen
125
2022
231
Novel journal bearing materials for wind turbine gearboxes
125
2022
Taneli Rantala
Kaisu Soivio
ict2310515
23rd International Colloquium Tribology - January 2022 515 Novel journal bearing materials for wind turbine gearboxes Taneli Rantala Moventas Gears, Jyväskylä, Finland Corresponding author: Taneli.rantala@moventas.com Kaisu Soivio Moventas Gears, Jyväskylä, Finland 1. Introduction Bearings have a marked influence on levelized cost of energy (LCoE) in wind industry. Both capital and operational expenses are influenced by bearings. [1] As the size of wind turbines is growing, the rolling element bearings become more expensive, and their performance becomes challenged due to increased load transmission requirement. One recognized opportunity to solve this issue is the implementation of journal bearings. However, traditional journal bearing materials do not provide reliable solution for the wind turbine lifetime requirements due to relatively slow sliding speeds and high, constantly changing load conditions. In standards, one can find limit values for projection pressure for traditional materials, such as Al-Sn alloys, which are in range of 5 to 7 MPa, whereas in wind turbine gearbox, specific bearing loads are generally in range from 10 to 20 MPa. Before introducing new constructional and material solutions to a wind turbine gearbox, it is necessary to investigate and prove the reliability of the selected solutions. Because it is not feasible to be done on full-scale for expected lifetime of 20 to 30 years, intelligent testing methods combining different testing size and complexity scales with developed modeling practices were chosen as approach to verify novel journal bearing materials to be used in wind turbine gearbox. This novel methodology has been developed under the i-TRIBOMAT open innovation test bed initiative as one of the three use cases. 2. Approaches 2.1 Field-2-Lab down-scaling (2 columns) First, relevant parameters and operation conditions for small scale testing need to be identified and quantified. i-TRIBO- MAT focuses on study of continuous load conditions on planet wheel radial journal bearings, so transient phenomena, mainly start-up and shut-down of turbine are not considered here. Load duration distribution (LDD) data of specific wind site acts as a basis for analysis. Input load data, desired ratio for gearbox, fixed geometrical values of bearing parts and operation mode data of a turbine can be used to define operational window for journal bearings as a function of projection pressure and sliding speed. Projection pressure and sliding speed can be selected as suitable parameters when down-scaling from product scale to laboratory scale material tests. In a similar fashion, Hertzian contact pressure can be used if all required parameters are known. Figure 1: Example of operation window, used for downscaling operation conditions With identified critical operation conditions, tests can be defined and criteria for material performance set. In combination with these input load parameters, environmental parameters such as oil pressure and temperature are set to present realistic operation conditions for sliding contact. All input values can be transferred to test setups of different scales, and viability of test method can be then considered. As time spent in critical point of operation window can be very long, continuous wear due to asperity contact is not allowed, and good run-in properties are preferred. Properties like self-lubrication during oil starvation and low wear rate are also considered to be beneficial. Objective of Moventas is to select and test possible journal bearing material candidates in product scale, and provide test results and performance data for i-TRIBOMAT Single Entry Point (SEP) members. This data is then used by the SEP members to develop laboratory scale testing environment and up-scaling tools for i-TRIBOMAT. 516 23rd International Colloquium Tribology - January 2022 Novel journal bearing materials for wind turbine gearboxes 2.2 EHD simulation Due to simplicity of definition of projection pressure, it was assumed that results it yields are not realistic enough, and more sophisticated method is needed to achieve more accurate representation of loading of journal bearing. For this goal, EHD-simulation capability was developed. Multiple simulation models of test gearbox with journal bearings were built. As expected, local hydrodynamic pressure values differ greatly from simple projection pressure values. Comparability of laboratory tests and product scale tests can be improved by constructing simulation model of both test rigs, and replacing projection pressure by local hydrodynamic pressure. Laboratory scale test conditions can be updated accordingly to represent operation conditions observed in actual product. 2.3 Product scale test (2 columns) Product scale test runs were conducted with test gearbox. First to study the limits of journal bearing system and ability to simulate the performance accurately, and second to test complete operation window of a gearbox with journal bearings. During first test run, operation conditions with clear mixed lubrication in journal bearing was found. When transitioning to slower sliding speed after that, a bearing seizure occurred. Simulation model indicated high asperity contact next to oil inlet of the journal bearing, and disassembly of the gearbox confirmed this to be the root cause for seizure. After optimizing the journal bearing geometry based on this information and modelling updates, asperity contact pressure was minimized, and complete operation window, including safety margins, was tested successfully. During disassembly of the gearbox after test run, no issues were observed in any journal bearings. Figure 2: Hydrodynamic and asperity contact pressure of journal bearing after geometry optimization 3. Conclusion Laboratory scale material tests and comparison to fullscale test results are still on-going. Initial results for selected bearing material indicate good wear resistance and sliding properties even in oil starved conditions, which are in line with gearbox test results. Results indicate the importance of advanced field-to-lab down-scaling practice. Generally implemented simplifications of tribological contacts lead to difficulties in drawing conclusions of material performance in real application, and during whole lifetime. Generally known tribological tests can put material pairs in tribological contact to performance order by certain parameters, but the prediction, i.e. up-scaling capability for system level performance is still limited and under further development. Up-scaling capable tribological testing can increase the likelihood of finding optimal material solution and simultaneously decrease the time and monetary effort related to the testing required for selection, as good performance is secured, but selected solution is not overly expensive leaving capacity half-used. Acknowledgement This project has received funding from the European Union’s Horizon 2020 research and innovation programme (innovation action) under grant agreement No. 814494 (Call: H2020-NMBP-TO-IND-2018) More details: https: / / www.i-tribomat.eu/ . References [1] Dao, C, Kazemtabrizi, B, Crabtree, C. Wind turbine reliability data review and impacts on levelised cost of energy. Wind Energy. 2019; 22: 1848-1871. https: / / doi.org/ 10.1002/ we.2404
