LIFE EXTENSION OF WINDFARMS
FEATURE SPONSOR
Using data to extend main bearing life through predictive maintenance
Recognising the issue that main bearings, unlike the gearbox, do not have great attention paid to the lubrication, with no filter for removal of metallic particulates, no dryers or desiccant breathers labour and no lubricant change at five year intervals (typical), Romax InSight developed a process for extending bearing life.
Now flushing around 50 bearings annually, Romax continues to optimise the predictive maintenance methodology. Thorough flushing of the bearing to remove all the grease and repacking with fresh grease resets the lubrication conditions. Data gathered from flushed bearings has consistently shown significant reduction in metallic debris, water content, temperature and vibration; extending bearing life significantly.
UNDERSTANDING THE FAILURE TO CATCH IT EARLIER
Spherical roller bearings are selected primarily due to cost and their ability to handle a large amount of radial load and misalignment. However, they have been found to have limited life in many turbine designs. The failure mode is complex, but a key factor is that the thrust load is supported principally by the downwind bearing, the downwind rollers often skew during rotation as well as sliding.
This raises surface stresses and leads to surface initiated micropitting. The debris generated causes bearing and cage wear, debris denting and early life failure as the bearing spalls out.
Figure 1. Romax InSight Fleet MonitorTM hardware-independent Cloud-based predictive maintenance software tracking main bearing damage with over a year’s lead time on the failure using advanced algorithms applied to the raw vibration data feed. Temperature provides only late stage warning.
been limited to a time when a crane can be arranged. If the bearing is in poor condition, it is common to pump in additional grease to buy some time (purging).
Combined with de-rating, the bearing can be limped along without tripping the turbine on temperature as often, whilst repairs are arranged. When the bearing is overheating, typically with cracking on the raceway shoulder at bottom dead centre, it is prudent to shut down. Both examples, inefficient purging and waiting time for repairs on inactive turbines, can cost production revenue significantly.
EARLY DETECTION TRIGGERS PREVENTATIVE MAINTENANCE Preventative maintenance by grease flushing extends the bearing life. To get the maximum benefit, the early detection of poor lubrication conditions or early stage wear and spalling
42
www.windenergynetwork.co.uk
useful data into decision making: SCADA, vibration, particle counters, lubricant analysis, inspection and maintenance data. Point B is a good time to use this data to schedule grease flushing.
COMBINING TECHNOLOGIES FOR BEST PRACTICE
Combining technology for early detection and bearing life extension allows for best practice management of main bearing health and reduces the cost of failure. A predictive maintenance programme incorporates assessing principally the vibration and grease data to develop a priority list of bearings that require flushing. Then batches of bearings can be flushed during convenient service periods to extend life.
EXISTING PRACTICES
There are a number of commonly used industry practices for managing main bearing health, which include alarms that are triggered when certain factors arise. Once an alarm is sounded, by vibration or temperature, the bearing is inspected to confirm the damage highlighted in the data. Historically the options for life extension at this stage have
is required. A periodic grease sampling and analysis programme provides insight into the lubrication quality. For vibration data, Romax has developed special algorithms that provide 12-18 months on the initial spalling on rollers and raceways – this is a critical point where the damage accumulation in the bearing accelerates. Figure 1 provides a case study with one year’s warning on bearing failure, when the first debris dents where appearing on the inner race.
The same Cloud-based software shows that temperature warnings are very late stage, even using an advanced algorithm that corrects for environmental fluctuations and tracks deviations from the fleet mean. The software allows users to pull all forms of
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84