Gabriela Fedor, Frank-Olaf Mähling, Christoph Wincierz, Thilo Krapfl
Evonik Operations GmbH – Specialty Additives, Darmstadt, Germany
Justin Langston
Evonik Industries, Horsham, USA
Juno Shin
Evonik Korea Ltd, Seoul, S.Korea
Summary
This paper describes important rig and laboratory tests for industrial gear oils to ensure high performance of wind turbines.
Original equipment manufacturers have defined specifications to guarantee designed performances such as anti wear, scuffing, micropitting, shear stability, elastomer compatibility, etc. Foaming behavior is an important test item of the Winergy specification. Stringent filterability tests are required by filter manufacturers Hydac and CC Jenssen.
Part one of this paper described NUFLUX™ technology, a new class of wind turbine gear oils that contain VISCOBASE® synthetic base fluids. The fluid technology has demonstrated its performance in a large number of rig tests and wind turbine field trials.
In part two, laboratory and rig test results of NUFLUX™ formulations are presented and the performance is compared to mineral and PAO based gear oils.
NUFLUX™ industrial gear oils meet the requirements of DIN 51517-3 as well as AGMA 9005-F16 and fulfill the relevant OEM requirements for wind turbine gear oils. Lab, rig and field trials have proven this technology as an alternative to fully PAO-based wind turbine gear oils.
- Performance and material compatibility testing – key parameters for bench testing and condition monitoring
To prove the performance of a lubricant formulation for the very demanding purpose of an industrial or wind turbine gear oil, a series of laboratory investigations must be complemented by a number of OEM specific bench tests.
Figure 1 shows the basic performance parameters that the fluid must pass to fulfil the industry standards, and the overlap with more complex requirements specified by individual gearbox manufacturers.
NUFLUX™ technology meets the requirements of an ISO VG 320 industrial gear oil according to DIN 51517-3 as well as AGMA 9005-F16 and relevant OEM requirements.
The most important additional OEM requirements are:
(1) foaming behaviour evaluated according to Winergy-requirements [1];
(2) filterability according to an in-house test of filter-manufacturer Hydac and CC Jenssen, with the focus on fine filtration (2,5-5 µm pore size filters);
(3) static- and dynamic seal compatibility tests, not only with standard elastomer types for small and mid-size gearboxes, but also with seal materials specified by the relevant wind turbine gearbox OEMs, typically conducted at Freudenberg [2, 3];
(4) micropitting tests according to the FVA 54/7 procedure at temperatures of 60°C and 90°C;
(5) FZG scuffing tests, conducted at single and double speed; and
(6) severe multi-stage approval tests by bearing manufacturer FAG.
1.1 Extended Flender Foam tests at different oil temperatures
The foam test that is found in many industry standards is ASTM D892, a test which determines the tendency of gears oils to form surface foam trough the use of blown air and a gas diffuser. As this test doesn´t correspond to field conditions, Flender developed a foam test that mimics the dynamic mode of moving gears by rotating them at 1405 rpm for five minutes. Strong oil agitation and air entraption result in foam formation that is closer to field conditions. For general applications, this test is typically run with the pure formulation and with 2% and 4% of impurities (usually flushing oils with detergent additive). Test conditions specify 25 °C as an ambient temperature.
For wind turbine applications, for example the test conditions specified by Winergy, excellent foam performance is expected for a wide range of operating temperatures, going from 0 °C all the way up to 60 °C.
NUFLUX™ VG 320, specifically formulated for wind turbine applications can meet Winergy specification demands (Figure 2).
1.2 Mechanical rig tests
There are a few ways of testing and evaluating the wear protection of the lubricant. Some specifications include the four-ball method (ASTM D 2783) for determining load-carrying capacity but caution must be used when correlating these results to field performance. Other specifications have adopted the FE8 wear test (DIN 518 19-3) which uses two roller bearings under axial load and which was developed by FAG Schaeffler. Further tests methods also exist such as FAG 4-step test program, FZG standard scuffing load test, or Micropitting (FVA 54/7).
Micropitting (FVA 54/7)
Micropitting is a wear phenomenon that occurs on the tooth flanks, often seen as surface damage on high rolling gears made of hardened steel.
Microscopically small fatigue fractures continuously form over the operating time of the gear, lead to profile form deviations, increasing performance fluctuations in the system, and finally to more severe follow-up damage modes, such as pitting, wear, or even tooth fracture [4, 5]. Micropritting wear is resulting in noise and vibrations. Its formation can be hindered when the tribological conditions of the system are re-established. Therefore, the use of a robust and efficient additive package dissolved in an appropriate base oil helps to achieve effective surface protection.
To evaluate their ability to protect gears from micropitting, lubricants are usually rated by the results achieved in the FVA micropitting test [6]. This test is divided into two sections: the stepwise phase, where starting from load stage 5 (795 N/mm²), every 16 hours the load is increased up to load stage 10 (1.547 N/mm²); and the endurance phase, where for 400 hours load stage 10 is applied. Typically after each stage, the micropitted area, the wear rate, and most importantly, the profile form deviations are recorded.
After each test period, the test gears are disassembled, and the profile of the tested flanks is measured using a 3-D measurement system. The micropitted area, the wear rate, and most importantly, the profile form deviation is recorded.
In the LS test, the failure criterion has been reached once the mean profile form deviation due to micropitting exceeds the limiting value of 7.5 μm.
The LS in which the failure criterion is reached is called “failure load stage”. Lubricants with a high micropitting load-carrying capacity reach the failure criterion of a profile form deviation of 7.5 μm due to micropitting in LS ≥ LS10 of the LS test (GFT-high).
At the end of the load stage test and endurance test with the first test gears, the load stage test is repeated with new test gears to check repeatability.
Most of the OEM standards relevant for general applications, e.g. Flender, require the test being run at one temperature, commonly 90 °C; wind applications often request LS 10 at 60 °C in addition. At the higher temperature, the test could be viewed as more severe, as lower viscosities produce thinner lubricant films. On the other hand, especially at lower temperatures, chemical factors, more so than viscosimetric factors could influence the outcome of micropitting test. Thus, a mediocre result at such moderate temperatures could be interpreted as lack of additive reactivity. NUFLUX™ exhibits excellent micropitting performance at both temperatures, achieving rating GFT-high for both conditions.
Profile form deviations of the two FVA 54/7 micropitting tests at 60 °C and 90 °C for NUFLUX™ ISO VG 320 are shown in Figures 3a and 3b, respectively.
FZG-Scuffing
The FZG-rig is used to determine the ability of a lubricant to protect gears from scuffing and to reduce friction [7]. In our investigations the test has been carried out at a single-speed of 1500 rpm (= 8,3 m/s pitch line velocity) as required by the DIN 51517-3 and also at double speed [8]. Within this test, the load stages are increased stepwise up to a load stage of 14. Table 1 shows the results according to DIN 51354-2. NUFLUX™ exceeds (at double speed) the specified requirements.
As the system is not cooled during the test, the oil sump temperature gives information on friction within the system. A lower temperature correlates with less frictional losses.
Figure 4 illustrates the average oil sump temperature development of 15 individual FZG A/8.3/90 tests with three different ISO VG 320 fluid categories. The mineral based fluids lead to the highest temperatures over all load stages. The PAO-based-fluids generate lower temperatures, especially at higher loads. But NUFLUX™ results in the lowest temperatures over all loads, on average 5 °C lower than the mineral oil formulation.
The FZG-scuffing test is not intended to evaluate temperatures. However, it is an informative stepping stone from gears to FE8 (bearings) to real life testing on a gearbox.
Gear efficiency
Encouraged by these results the same fluids had been tested on an FZG efficiency test [9]. In this test, the torque losses in an FZG rig are measured at 30 °C and 60 °C. At each temperature, the torque is stepwise increased from 0 Nm (load stage 0) to 373 Nm (load stage 10). The results can be found in Figure 5.
The FZG torque loss study confirmed the results that were generated within the FZG scuffing test. At 30 °C, NUFLUX™ showed the lowest torque losses across all load stages compared to the mineral oil and even the PAO formulation. The losses of NUFLUX™ were found to be on average 11% lower than those of the mineral oil fluid and 6% lower than those of the PAO formulation.
At 60 °C, it can be seen that both the PAO formulation and NUFLUX™ technology outperform the mineral oil formulation just at higher loads. On average, NUFLUX™ technology and the PAO-fluid show a torque loss reduction of more than 6%.
Bearing tests
Bearings are the crucial connecting components between the highly-loaded mechanical gear sets of a wind turbine and the static gearbox cage.
Roller bearings are a key component of the wind turbine gearbox. The range of their operating conditions is broad and some can be very severe. Therefore, the ability of the oil and additive system to protect the bearings from damage is at least as important as it is for the gear system. Unfortunately, a given lubricant may provide good protection to the gears but not necessarily to the bearings.
To take this into account, bearing manufacturers have set up their own test procedures covering a broad range of operating conditions.
FE8 wear test (DIN 51819-3)
The FE8 test head is fitted with a shaft and two sets of tapered roller bearings. The test is conducted in two runs under identical conditions (80 kN load, 80°C and 800 rpm). These parameters represent borderline conditions in the gear system (boundary/mixed lubrication). After the first run, the bearings are examined gravimetrically and are exchanged for new ones. The weight loss is then defined as a mean Weibull roller and cage numbers, which has a limit of 30 mg on rollers in most of the specifications, including Flender and DIN 51517-3. Cage weight loss is just reported without particular limits, but usually not forgotten during discussions with OEMs.
Extensive variations have shown that the quality of wear protection of the formulation is mainly controlled by the additive package.
FAG 4-step test
The most important bearing test for wind turbine gearboxes is the 4-step test run mainly on previously mentioned FE8 test rig [10, 11]. This test was developed to simulate different critical conditions in wind turbine gear boxes that could potentially occur in the field.
The procedures has four steps.
- Short-term test performed on FE8 test rig according to DIN 51819, Parts 1 to 3 with 80 KN axial load and at 80°C for a duration of 80 hours.
- Fatigue test with moderate mixed friction performed on FE8 test rig at 75 rpm with 100 KN axial load and at 70 °C for a duration of 800 hours.
- Fatigue test under EHL conditions (10 bearings), performed in FAG test rig L11 at 9000 rpm, with 8.5 KN axial load, and at about 80 °C for a duration of 700 hours.
- Deposit test at higher temperatures in the presence of water. This modified PM paper-making machine oil test from FAG is performed on a special FAG test rig at 750 rpm, with 60 KN axial load, and at up to 140°C for a duration of 600 hours.
In Table 2, the results of the FAG 4-step bearing test program with NUFLUX™ ISO VG 320 is summarized. The formulation achived excellent overall result.
The frictional behaviours of NUFLUX™ in thrust ball bearings and cylindrical roller thrust bearings were independently evaluated in two studies by a group at the University of Porto [12, 13]. In these studies, five ISO 320 wind turbine gear oils were tested in a modified Four-Ball machine where the Four-Ball arrangement was replaced by a rolling bearing assembly which was developed to measure the friction torque as well as the operating temperature at several different points. The five oils included: one PAO-formulation, one ester-based formulation, one PAG-formulation, one mineral oil formulation and NUFLUX™ technology (“MINE”). In these two papers, it was found that NUFLUX™ showed very good frictional behavior, especially in thrust roller bearings, in which it showed by far the lowest frictional torques of all tested fluids.
- WTGO approval process (Winergy example)
Wind turbine gear oil approvals start with extensive lab and bench testing. Once this testing is completed, and confirmed to meet technical requirements, then an approval to conduct field trials is granted by the wind turbine gearbox manufacturer. The field trial is an official part of the final approval process and can take several years.
The timeline of a full wind turbine gear oil approval is decribed in Figure 6.
Standard testing of oil parameters according to Winergy specification has estimated an optimistic timeline of six to eight months. Spare part OEM testing such as FAG 4-step test might take up to two years to be completed. Entering the first field trial phase means that the scope of the testing (number of selected turbines, specified gearbox output, frequency and depth of oil monitoring, etc.) needs to be defined in close coordination with the OEM. Usually, the first phase of the field trial is conducted on smaller number of turbines (two to five) and takes one year. After one year of succesful operation with promising endoscopy and oil monitoring results, the dialogue with the OEM continues and the field trial is typically extended to more turbines and runs for another two years. Towards the end of the second field trial phase, the testing extends from standard oil monitoring to more demanding performance tests, such as micropitting, FE 8 or filterability. This adds on significant amount of time to the overall approval timeline and after the successful completion, general approval for using new lubricant is considered. Still this can be restricted to a special customer and special gearbox type depending on the conditions of the overall approval process.
NUFLUX™ ISO VG 320 has successfully completed standard and spare part OEM testing and is approved by Winergy for the field trial phase.
- Conclusions
After completing a wide range of laboratory and rig tests, NUFLUX™ technology fulfills all technical requirements for wind turbine gear oils. Good compatibility and high solvency of the oil can protect the equipment, and prevent deposits and varnish. Reduced gear operating temperatures allow for extended oil drain intervals.
NUFLUX™ ISO VG 320 has successfully completed the OEM testing and was approved by Winergy to enter a field trial.
As described in part one, NUFLUX™ ISO VG 320 served more than 40 windturbines worldwide. 20 windturbines were operated in Europe over more than six years without any oil change in gearboxes of Winergy, Moventas and ZF Wind.
NUFLUX™ technology has shown equivalent performance to widely adopted PAO-based wind turbine gear oils and received acceptance as a proven alternative for this demanding application.
References
[1] K. Tschauder; Lubricants in Wind turbines – highest requirements on development, testing, and application; GETLUB Conference 2010.
[2] J. Braun, Elastomerverträglichkeits-untersuchungen von Schmierstoffen – Reicht die bestehende Normung aus?; Tribologie und Schmierungstechnik; 56th year; 6/2009.
[3] Freudenberg Formblatt 73 11 008; Statische und dynamische Ölverträglichkeitstests mit
Freudenberg Simmerringen zur Freigabe für den
Einsatz in FLENDER-Getrieben (Tabelle T 7300)
[4] Lubricants and Lubrication; 2nd Ed.; Th. Mang, W. Dresel; 2007; pages 242 – 243
[5] G. Schönnenbeck; Einfluss der Zahnflankenermüdung (Graufleckigkeit und Grübchenbildung) hauptsächlich im Umfangsgeschwindigkeitsberich 1-9 m/s; Diss.; TU Munich; 1994.
[6] FVA Informationsblatt 54/I-IV, Testverfahren zur Untersuchung des Schmierstoffes auf die Entstehung von Grauflecken bei Zahnrädern, 1993.
[7] Lubricants and Lubrication; 2nd Ed.; Th. Mang, W. Dresel; 2007; pages 750 – 751.
[8] FVA Informationsblatt 243/5; Scuffing Test EP Oils; 2001.
[9] Wienecke, D. Einfluss der Art und Zusammensetzung von Schmierölen auf die Verlustleistung in PKW- und Schaltgetrieben.
[10] DIN 51819 part 1; Prüfung von Schmierstoffen – Mechanisch-dynamische Prüfung auf dem Wälzlagerschmierstoff-Prüfgerät FE8 – Teil 1: Allgemeine Arbeitsgrundlagen; 1999-12.
[11] DIN 51819 part 1; Prüfung von Schmierstoffen – Mechanisch-dynamische Prüfung auf dem Wälzlagerschmierstoff-Prüfgerät FE8 – Teil 3: Verfahren für Schmieröl, einzusetzende Prüflager; 1999-12..
[12] Fernandes, C. M. C. G.; Amaro, P. M. P.; Martins, R. C.; Seabra, J. H. O.; Torque loss in thrust ball bearings lubricated with wind turbine gear oils at constant temperature; Tribology International 66, 2013, pages 194 – 202.
[13] Fernandes, C. M. C. G.; Amaro, P. M. P.; Martins, R. C.; Seabra, J. H. O.; Torque loss in cylindrical roller thrust bearings lubricated with wind turbine gear oils at constant temperature; Tribology International 67, 2013, pages 67 – 80.
– End of the report –