Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Gearbox Typical Failure Modes, Detection and Mitigation Methods

Download as pdf or txt
Download as pdf or txt
You are on page 1of 24

Gearbox Typical Failure

Modes, Detection and


Mitigation Methods

Shawn Sheng
National Renewable Energy Laboratory/National Wind
Technology Center

AWEA Operations & Maintenance and Safety Seminar

January 1516, 2014

NREL/PR-5000-60982

NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.
Outline
Background
Gearbox Typical Failure Modes
Bearings
Gears
Lubricant
Detection Techniques
Real-time
Periodic
Mitigation Strategies
Bearings
Gears
Lubricant

2
Background
Gearbox Reliability Challenges
Gearboxes do not always
achieve their 20-year design
life
Premature gearbox failures:
Are widespread
Affect most original equipment
manufacturers
(Most) are not caused by
manufacturing practices, except
grind temper and nonmetallic
inclusions
Gearbox largest contributor
to turbine downtime and
costliest to repair*
* Sheng, S. (2013). Report on Wind Turbine Subsystem Reliability A Survey of Various Databases. June 2013. NREL/PR-5000-59111.
4
Gearbox Damage
Gearboxes can fail in drastically different ways
Data: 257 gearbox damaged records
Bearings: 70%; gears: 26%; and others: 4%
Observations:
Both bearing and gear failures are concentrated in the parallel section
Top failure mode is high-speed shaft (HSS) or intermediate-speed shaft (IMS)
bearing axial cracks

Damage
Bearings Gears Others
Records
Planetary 44 23 21

IMS N/A 34 9
47
HSS N/A 123

Total 257 180 68 9

5
Typical Failure Modes
A Representative Configuration
High-Speed High-Speed
Stage Shaft

Pinion

Intermediate-Speed
Annulus Shaft

Gear Pinion
Planet
Planet
Carrier
Low-Speed
Shaft

Sun
Gear

Low-Speed
Stage
Intermediate-Speed
Stage

7
Typical Failure Modes: Bearings

Macropitting
Axial Cracks
Recent failure mode of significant proportions
Most appear initiated with irregular white
etching areas
Manifests as axial cracks within the roller path
Progresses to macropitting, split inner ring, or Denting/Point-surface-
both origin (PSO) macropitting

Fretting corrosion Scuffing False brinelling


Photo Credits: Bob Errichello, GEARTECH, Johan Luyckx, Hansen Transmissions, Andy Milburn, Milburn Engineering, and Ryan Evans,
Timken
8
Typical Failure Modes: Gears

Subsurface initiated bending fatigue Scuffing (high-speed stage pinion)


(due to nonmetallic inclusion,
intermediate-speed stage pinion)

Fretting corrosion (sun pinion) Micropitting/macropitting


Photo Credit: Bob Errichello, GEARTECH
9
Typical Failure Modes: Lubricant
% Additives vs. Viscosity @ 40C cSt D-445

R&O Mineral Oil


Percent of New Oil Additive

100.0 510

Viscosity @ 40C cSt


90.0 470
80.0 430
70.0 390
350
60.0
% Electrooxidizable Additives 310
50.0 270
Viscosity @ 40C
40.0 230
30.0 190
New Oil Vis. = 46.70 48.27 cSt 150
20.0
110
10.0 70
0.0 30
0 36 72108 138
Days Aged in Turbine Oil Stability Test D-943
Plot Credit: Bill Herguth, Herguth Laboratories
10
Typical Failure Modes: Lubricant (Cont.)
Contaminations include:
Wear debris
Dust

Foam
Breathing of
housing
Air trapped in oil
Not enough
settling time

Photo Credits: Bill Herguth, Herguth Laboratories, and Art Miller, EDFR
11
Detection Techniques
Real-Time Continuous
Critical SCADA data trending:
Temperature of high-speed stage
bearing, targeting surface damage
modes: pitting, scuffing, and so
on
Vibration analysis:
Dedicated condition monitoring
system with sensor bandwidth up
to more than 10 kHz
Effective for high-speed shaft,
intermediate-speed shaft bearing,
and gear surface damage modes:
axial cracks, macropitting,
scuffing, and so on
Earlier detection than oil debris
analysis, but challenging to detect
damage to planetary stage
bearings and gears
13
Periodic
Oil sample analysis
Six months or annual oil samples sent to dedicated laboratories for typical physical
and chemical property analysis: viscosity, additives depletion, and so on
Debris element analysis may pinpoint failed components: bearings, gears, and so on
Good for oil condition, not very good for conditions of bearings and gears
Filter element analysis
Old filter cartridge can be sent to dedicated laboratories for analysis when it is
replaced with a new one
Reveals what is typically missed in conventional oil sample analysis
Identifies composition of wear debris and pinpoints damaged components: bearings,
gears, and so on

Reference Limits Analysis Results


14
Periodic (Cont.)
Portable vibration sweep
Sensors are similar to what are used by permanently installed
vibration analysis systems
Typically conducted during end-of-warranty (EOW) inspections
targeting the same damage modes as permanently installed vibration
systems
Difficult to track down failure history
Borescope inspection
Dedicated scope to inspect all accessible stages of gearboxes
Normally triggered by abnormal symptoms on the gearboxes
detected by other methods and used to confirm the damage,
evaluate its severity level, and inform maintenance planning: damage
modes observable on component surfaces, pitting, scuffing, and
problems with the lubricant such as foaming
Another scenario for conducting borescope inspections is during an
EOW sweep
Depending on gearbox configuration, the inspection may not be
thorough because of the difficulty in accessing and photographing
certain components
15
Mitigation Strategies
Bearing Axial Cracks
Carburized instead of through hardened bearings1
For carburized bearings, the level of retained
austenite is recommended to be at least around
20%1
Assembly temperature of bearing inner rings
around 120 degrees Celsius2
Reduce transient event frequencies and their
magnitudes
Keep lubricant cool, clean, and dry

1. Errichello, R.; Budny, R.; Eckert, R. (2013) Investigations of Bearings Failures Associated with White Etching Areas (WEAs) in Wind
Turbine Gearboxes. Tribology Transactions, Vol. 56, No. 6, pp. 1069-1076, DOI:10.1080/10402004.2013.823531
2. Luyckx, J. (2011) Hammering Wear Impact Fatigue Hypothesis WEC/irWEA Failure Mode on Roller Bearings. Wind Turbine
Tribology Seminar, Broomfield, CO.
17
Bearing/Gear Micropitting
Grind/hone/polish gear teeth
Superfinish, coat, or smooth bearing rings or elements
Avoid shot-peened flanks
Make hardest gear/bearing element as smooth as
possible
Make pinion 2 HRC points harder than gear
Use oil with high micropitting resistance
Keep lubricant cool, clean, and dry
Use high-viscosity lubricant
Operate at high speeds
Coat teeth with phosphate, Cu, or Ag
Run-in with special lubricant [w/o ZDDP (Zinc Dialkyl
Dithio Phosphate)] at controlled loads
Errichello, R. (2011) Wind Turbine Gearbox Failures. Wind Turbine Tribology Seminar, Broomfield, CO.

18
Gear Scuffing
Optimize gear geometry/accuracy
Use nitrided steel
Grind/hone/superfinish gear teeth
Coat gear teeth with phosphate, Cu, Ag
Use high-viscosity antiscuff lubricant
Cool gear teeth with lubricant
Run-in new gear-sets at reduced loads

Errichello, R. (2011) Wind Turbine Gearbox Failures. Wind Turbine Tribology Seminar, Broomfield, CO.

19
Lubricant Contamination
Pre-filter oil
Run-in oil flushing
Main loop filtration
Offline loop filtration
Breather
Heat exchanger
Periodic seal examinations

Photo Credit: Don Roberts, DA Roberts LLC.


20
Summary
Summary
Wind turbine gearboxes can fail in dramatically different ways.
Improvements in reliability and availability have to take a
holistic approach involving design, manufacturing, testing,
packaging/shipping/handling, installation, operation, and
maintenance. It also involves almost all stakeholders, gearbox
OEMs, turbine OEMs, owners/operators, research institutes,
governmental agencies, and so on.
Each detection technique has its own advantages and
limitations. The same is true for oil sample/filter element
analysis, and end users need to come up with a solution that is
the most economical and effective for their assets. Note that
one solution for one plant may not apply to a different site.
Tracking and knowing your fleet condition through various
instrumentation and data mining are critical.

22
Gearbox Reliability Improved
Wind Stats data from Germany: about 5,000 turbines, majorities megawatt-
scale

Both 2007 and 2012 data indicate the gearbox as the highest downtime driver,
but this was reduced by 67.7% in 2012 with respect to 2007
Most subsystems show improved reliability and total downtime caused by all
subsystems per turbine; in 2012 shows 47.1% reduction with respect to 2007
23
Thank You!
Valuable reviews and comments from Bob Errichello are greatly appreciated!

NRELs contributions to
this presentation were
funded by the Wind and
Water Power Program,
Office of Energy Efficiency
and Renewable Energy,
U.S. Department of
Energy under contract No.
DE-AC02-05CH11231.
The authors are solely Photo from HC Sorensen, Middelgrunden Wind Turbine Cooperative, NREL 17855

responsible for any


omissions or errors
shuangwen.sheng@nrel.gov
contained herein.
303-384-7106

24

You might also like