1

OAK RIDGE NATIONAL LABORATORYU. S. DEPARTMENT OF ENERGY

Peter J. Blau Materials Science and Technology Division Oak Ridge National Laboratory Oak Ridge, Tennessee USA

Scuffing: From Basic Understanding to Engine Materials Testing*

DEER ConferenceDEER Conference Detroit, MI – August 2007

* Research sponsored by DOE/ EERE / OFCVT: “Durability of Diesel Engine Components”

2

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

Definitions for Scuffing

� Swedish skuffa – ‘to push’

� Webster’s Unabridged Dictionary, 3rd ed.: “to walk without lifting the feet; to poke or shuffle a foot in exploration or embarrassment; to become scratched, chipped, or roughened by wear.”

� ASTM Terminology standard G40: scuffing – a form of wear occurring in inadequately-lubricated tribosystems that is characterized by macroscopically observable changes in texture, with features related to the direction of motion.”

Operational definition: “I’ll know it when I see it.”

3

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

Characteristics of Scuffing

� Scuffing can roughen a surface with no net loss of material (not ‘wear’).

� Can smooth an initially rough surface.

� Need not be progressive (one cycle).

� In machinery (engines), scuffing is associated with inadequate or failed lubrication.

� Occurs non-uniformly. Can start at one place on a surface and spread to another after continued operation.

� Can lead to seizure (incipient galling) in tight-tolerance components.

� Scuffing damage is difficult to measure in a quantitative way.

Zirconia on 304 SS (600o C)

4

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

Historical Attempts to Define Critical Scuffing Criteria

� ‘Plasticity Index’ (Greenwood and Williamson, PRS, 1966).

� Modified Plasticity Index (Whitehouse and Archard, PRS, 1970).

� ‘Film thickness ratio’ L (Beerbower, ASLE Trans., 1971)

2/1*'

⎟⎟⎠

⎞

⎜⎜⎝

⎛

⎟⎠⎞

⎜⎝ ⎛=

RH

E σψ

c

h

σ =Λ

2/1

*

*' 06.0 ⎟⎟⎠

⎞

⎜⎜⎝

⎛

⎟⎠⎞

⎜⎝⎛=

β σψ

H

E

β∗ (autocorrelation function) is measure of surface randomness, β* = 0 when surface heights are random; β* =1 for a flat, smooth surface)

5

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

Some Problems with Critical Scuffing Criteria based solely on Mechanics

� Surface roughness rarely remains constant once relative motion begins (break-in, wear-in). Plasticity indices cannot easily accommodate the consequences of time-dependent changes in roughness and texture during continued contact.

� Plasticity models ignore the important effects of the lubricant chemistry (Park and Ludema, Wear, 1994)

� Testing approaches to oil additive formulation are based on lubricant properties. Tests keeps the material combination and surface conditions constant.

6

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

Current Scuffing Tests Popularly Use Stepped Loading to Determine Critical Loads

� Stepped loading does not simulate actual component conditions.

� Dwell times at each load allow subsurface damage to accumulate and may not produce the same result as if load were held constant for the same time but at the level that induced scuffing.

� Stepped loading tests do not address the issue of localized initiation and propagation of scuffing damage.

� Stepped tests often use unidirectional sliding but many key engine components reciprocate (e.g., pistons, fuel injectors, actuators, valve guides)

L

t

μ

7

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

Some ASTM Standards Used to Quantify Scuffing-Related Performance of Oils

� ASTM D5182: “Standard test method for the scuffing load capacity of oils (FZG visual method)” ( Example of step loading )

� Motor-driven gear set uses twelve 15-minute test stages with examinations after each one.

� Document shows sample images of polishing, scuffing, and scoring to estimate the extent of damage

� “Failure” occurs when total width of scuffing or scoring damage for all 16 teeth on the test gear ≥ the width of one tooth (20 mm). Metric: the ‘failure load’ stage.

� ASTM D6078 Ball-on-Cylinder Lubricant Evaluation (BOCLE)

� ASTM D6425 Optimal SRV Test – high-speed linear oscillation

8

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

ORNL’s Approach to Scuffing Measurement Case Study I: Fuel Injector Plungers

� Develop a convenient bench-scale laboratory test that does not rely on stepped loading, but characterizes the time-dependent progression of damage on a contact surface.

� Flexibility to use either experimental test coupons or production parts.

� Minimize the time per test.

� Enable studies of the effect of surface finishes, coatings, and other surface engineering approaches.

� Develop scuffing maps / models useful in material selection.

9

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

Development of the ‘Pin-on-Twin’ Test

� Geometry allows testing both simple cylinders and actual fuel injector plungers.

Load

Top pinBottom pins

� High-speed data acquisition captures friction for various locations on the stroke as a function of time or numbers of cycles.

10

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

First experiments evaluated 52100 steel in #2 Diesel fuel and a low sulfur fuel (‘Jet A’)

0.00

0.05

0.10

0.15

0.20

0 20 40 60 Sliding Distance (m)

Fric

tion

Coe

ffici

ent

Annealed E52100 Steel Hardened E52100 Steel

Jet A fuel-lubrication, 10 N load, 5 Hz frequency, 10 mm stroke

11

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

Friction Trace Comparison Method Enables Initiation to be Detected

Hardened E52100 Steel (Jet A, 50 N, 5 Hz)

-0.20

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

Sliding Distance (m)

Fric

tion

Coe

ffici

ent

Initial 12 18 30 60

12 m

18 m

Scuffing Map Portrays Initiation and Spread of Surface Damage on Cylindrical Plungers

Initial surface roughness affects0.065 the time to initiate scuffing and

0.052-0.065 0.052 the time to transition to a fully0.039-0.052

0.026-0.039 scuffed surface0.013-0.026 0.000-0.013

0.039

Δμ 0.026

600

540

600 480

0.013

Scuffed 0.000 480

Slid

ing

Tim

e (s

ec)

420

Jet Fuel0 3601 360 2 240 3 Transition 4 (low S) Time (s) 300 5 120 7 8 09

240 Stroke Location (mm) 10 180

120 No scuffing Global Scuffing

Local Scuffing Sequence of friction traces on a simulated 60

fuel injector plunger 0

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45

Composite Roughness R q ' (μm)

Zirconia on Steel OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

12

13

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

New Multi-Stage Scuffing Model Includes both Lubricant and Materials Effects

Lubricant flow and viscosity

Boundary film properties

Material properties

14

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

Case Study II: High-Temperature Scuffing of Wastegate Bushing Materials for EGR Systems

� Developed a “bow-tie test” for cylinder on flat geometry up to 650o C

� Measures changes in torque due to surface damage and debris accumulation

OSCILLATION θ ~ 50o

25.4 mm 32 mm

15

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

Multiple Criteria were used to create Scuffing Severity Maps

1) Torque as a function of time

2) Changes in surface roughness

3) Examination of test specimens

J. Wu and M. Yao, Deloro-Stellite

200

400

600

800

1000

1200

1400

1600

1800

0 0.5 1 1.5 2

Tile roughness vs. specific load

SPEC

IFIC

LO

AD

ROUGHNESS, Ra (μm)

440C/PL33

Stellite 6B/Stellite 6B

316 (Nit) / IDM5399

440C/IDM5399

T-400C/T-400C

Stellite 3 ctg/Stellite 3 ctg

T-400C ctg/Stellite 3

316 (Nit) / T-400C

Worse than the others Better than the others

WORST RESULTS

(Corrected for the measured scuffed area)

(After testing)

Biodiesel Additive Issues: How much BD can improve lubricity? Fundamentally, what effect does BD have on scuffing initiation and propagation?

� The National Biodiesel Board – www.biodiesel.org – reports lubricity benefits when adding > 1% to #2 Diesel fuel (BOCLE). Soybean oil-based additives claimed to meet or exceed lubricity of current diesel fuels.

� Degree to which BD additives improve lubricity of low-S fuels depends on which test method is used (e.g., BOCLE, SRV, etc. -- report by L. Schumacher, U Idaho)

� Materials-based research is needed to understand these discrepancies and to identify BD effects on the initiation and propagation of scuffing damage in different materials.

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

16

17

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

DOE’s HTML User Program (Tribology Research User Center) Provides Access to Specialized Test Methods

High Temperature Sliding Friction / Wear

Repeated Impact

Ring and Liner

High-temperature valve tests

Hot scuff testing

Advanced materials

18

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

Summary

� Scuffing is context-dependent. Tribosystems should be individually analyzed in order to define ‘failure of function’.

� Scuffing damage can occur quickly or over time: depends on operating conditions, materials, and lubrication.

� Approaches: redesign, alter operating conditions, alternative materials or surface treatments, surface finish optimization, or change lubricant type and means of supply.

� Simulative experiments and analytical models should be applied to investigate effects of biodiesel fuels on the scuffing mechanisms in materials/coatings.

� Methods described here are available to U.S. industry and universities in the HTML Tribology Research User Center.

19

OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY

Special thanks to …

� Jun Qu, ORNL

� John J. Truhan, Jr. (UT, Caterpillar)

� Brian C. Jolly (ETSU, ORNL)

� Ray Johnson, ORNL

� Yuri Kalish, George Hansen (DDC)

� James Wu and Matthew Yao (Deloro-Stellite)

� Sid Diamond (DOE, dec.)

� Jerry Gibbs (DOE)

� OFCVT (HVPM program and the HTML User Program)