Wind Turbine Blade Testing at NREL

2008 Sandia Blade WorkshopScott Hughes

National Renewable Energy Laboratory

National Wind Technology Center

Golden, CO USA

scott_hughes@nrel.gov

Outline

Background of the NWTC and wind turbine blade testing

Testing Capabilities at the NWTC Aggregate Blade Test Results Current Test Activities Test methodology development

About the NWTC Testing Group

Perform Full-scale and subcomponent tests of wind turbine blades Perform research and prototype tests on DOE sponsored

programs Perform Certification tests for industry clients through

CRADAs and other work-for-other agreements NREL also has dedicated dynamometer test facility and

field testing capabilities Development of test methodologies

Improve accuracy and efficiency of testing Develop methods for testing advanced blade designs

Bend-twist Swept Soft structures

Demonstrate advanced NDE systems and structural health monitoring systems

Why We Test Blades

Testing is a Certification Requirement Demonstrate blade can withstand the

design/test loads Minimize risk of field failures

Validate Blade Design Demonstrate as-built blade properties

consistent with design Stress and Strain Stiffness / Deflection Ultimate strength testing, buckling stability or

other failure mode

Structural Testing Facilities

Three Testing Facilities IUF Blades to 50-m A60 and 251 Blades to 19-m

Typical Tests Static Testing Fatigue Testing Property Testing (modal, mass

distribution) ISO/IEC 17025, A2LA Accredited for

full-scale blade testing Subcomponent Testing

22-kip, 100-kip load frames

Structural Testing Data Acquisition

Data Acquisition Common Sensors

Load Displacement Strain gages lots of them Accelerometers

Acquisition Systems BSTRAIN

NI based Chassis VDAS

Scalable Ethernet based system Distributed A/D layout minimizes

instrument cable lengths

NDE Capabilities Acoustic Emission Modal Testing Thermography Surface Profiling

Static Testing Tests the ability of the blade to

withstand design load cases or ultimate strength as required

Typically applied in 4-6 load vectors representing worst case loading for standard design load cases

Load application through quasi- static methods

Cranes Ballast Weights Winches Hydraulic actuators

Fatigue Testing Tests the ability of the blade to withstand operating-life

loads 20-year blade life on the order of 1x109 cycles Laboratory testing accelerates loading through increasing load

magnitude to manageable number of cycles, typically 1-million to as high as 10-million

Methods Single-axis Two-axis Forced Displacement Resonant Methodology development more on this later

smgershTypewritten Text(click image to play video)

Structural Testing History

0

2

4

6

8

10

12

14

16

Num

ber o

f Bla

des

Test

ed

AOC

15/5

0

AWT-

26/2

7

Berg

ey X

L50

GE

Win

d 1.

5-M

W

GE

Win

d 75

0-kW

GE

Win

d 55

0-kW

Foam

Mat

r ix 1

5-ft

KENE

TECH

16-

m

NedW

ind

25-m

NPS

NW-1

00-k

W

NPS

NW-2

50-k

W

Phoe

nix

7.9-

m

PSE-

85-k

W

TPI -

56-

100

TPI 3

0-m

m s

tuds

TPI/M

HI 1

000-

A

USW

56-

100

Win

dlite

SW W

indp

ower

MHI

/TPI

92

Clip

per C

93/C

96

WTC

-POC

Full Blade Fatigue - 56

Full Blade Static - 65

NDE - 15

Substructures - 39

Updated May ,07

NREL/NWTC USA

WMC / TU Delft Netherland

Blaest/Risoe DTU Denmark

NaREC United Kingdom

CENER Spain

Fraunhofer Institute Germany

Private Labs LM, Siemens, MHI, Tecsis

Blades Tested at NRELBlade Testing Labs

May 2008 125 full-scale blade tests

Aggregate Test Blade Data

Megawatt-scale blades Designs representative of current in-

field designs Blade testing in accordance with IEC

61400 -23 Standard Loading above IEC envelop not included Fatigue testing above target damage

equivalent loading not included

Catastrophic Failure Breaking of primary blade structure Complete failure of structural elements,

internal or external bond lines, skins, shear webs, root fasteners

Major parts become separated from the main structure

Functional Failure Reduction in Stiffness (5 to 10%) Permanent deformation Substantial permanent change of cross-

sectional shape After unloading the blade, a mechanism is

no longer capable of performing its design objective

Superficial Failure Small cracks not causing significant strength

degradation or bond line weakening Gel coat cracking Paint flaking Surface bubbles Minor elastic panel buckling Small Delaminations

Definition of Failures (IEC 61400-23)Test Data Population

Representative Failures

Bondline

Failures

Material Failures

Stability

Aggregate Catastrophic or Functional Failure Modes

One Test Article may have multiple failure modes

Design / Material Inadequate Design Ply Drops Panel Stability

Laminate Defect Waves Voids Dry spots

Bondlines Shear Web Leading Edge Trailing Edge

Root Fixing T-bolts Barrel Nuts Flanges

0

5

10

15

20

25

30

35

40

45

Design /Material

Laminate Defect Bondline Root Fixturing

Perc

ent o

f occ

uran

ce

0

10

20

30

40

50

60

70

80

90

100

Static Fatigue

Perc

ent o

f occ

uran

ce

Observations and recommendations from a blade testers perspective

Majority of laboratory failure events are during fatigue testing Anecdotally this is the reverse of what European laboratories and manufacturers have observed

The results of lab tests and how they relate to in-field service of blades is difficult to determine due to IP.

Encourage open forum of manufacturers, certifying bodies, operators, and testing laboratories to enable a more accurate assessment of testing methods

Sometimes it is not possible to test every detail on blade Add expanded component testing requirements Full-scale tests necessary subcomponent testing will not capture design or manufacturing details of

as-built blades Normative NDE requirements should go beyond current requirements

Test articles are typically from the initial lot of production blades if not the first article produced Benefit of attention to detail during construction Test Article can be subject to defects due to lack of production experience

Laminate imperfections detected prior to testing could be used to gauge adequacy of test article

Damage Tolerant Design based testing could however incorporate or require using worst-case test blades

Effect of multiple assembly plants and lines should be evaluated this topic not rigorously defined in current IEC standards

Blade Design Standard?

Recent and Current Tests TX-100 9-m blade

New resonant hardware (UREX) system development

Bend-twist test technology BSDS Testing

Fatigue test In progress Verdant Power

Marine Kinetic Hydropower System Blade and rotor system static testing

Other Commercial Projects 40-50-m blade static and fatigue

testing

Knight and Carver STAR Blade Testing

26-m blade fatigue test Swept planform blade

testing using resonant technology

Moments and torque (twist) of the test loads ideally optimized to properly simulate in- field conditions

NWTC Testing Capacity Current laboratory can test 50-m blades,

industry just keeps making bigger blades Need for new US test facilities

Accommodate blades larger than 50-m Accommodate demand of commercial

clients Detailed presentations to follow

Testing methodology development Forced Displacement Testing

Dual-axis capability Accurate load replication High equipment and operating cost Longer test duration Prohibitive to test entire blade length Large test loads introduced to blade

Linear Resonant Testing Low(er) Equipment costs Relatively short test duration Can test entire blade span Method scales favorably for larger turbines Ability to ramp load at resonant frequency Currently for large blades, resonant/forced hydraulic dual-axis

testing is unavailable

0

200

400

600

800

1000

1200

1400

1600

0 10 20 30 40 50 60 70

Blade Length (m)

Hyd

raul

ic A

ctua

tor F

low

Rat

e (lp

m)

Flow Rate for Forced Displacement is 0.25-Hz Cycle Frequency

Resonant Testing Flow Rate

Single-axis testing

Dual Axis Resonant Testing

Employ linear actuators in both flap and edge directions

System operated at two distinct frequencies

Potential for single-frequency operation with tuning elements

0

0.2

0.4

0.6

0.8

1

1.2

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

Frequency [Hz]

FFT

Mag

Fundamental Flap Frequency

Fundamental Edge Frequency

From Dual AxisTime-series Data

Edge Actuator

Flap Actuators

Dual-axis Resonance Movie

smgershTypewritten Text(click image to play video)

Thank You

Slide Number 1OutlineAbout the NWTC Testing GroupWhy We Test BladesStructural Testing FacilitiesStructural Testing Data Acquisition Static TestingFatigue TestingStructural Testing HistoryAggregate Test Blade DataRepresentative FailuresAggregate Catastrophic or Functional Failure ModesObservations and recommendations from a blade testers perspectiveRecent and Current TestsKnight and Carver STAR Blade TestingNWTC Testing CapacityTesting methodology developmentDual Axis Resonant TestingDual-axis Resonance MovieThank You