Mechanical Engineering Tribology Laboratory (METL) November 14, 2013 Yi Shen Research Assistant Effect of Retained Austenite and Residual Stress on Rolling Contact Fatigue
Dec 15, 2015
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
Yi ShenResearch Assistant
Effect of Retained Austenite and Residual Stress on Rolling Contact
Fatigue
2
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
Outline• Background• Motivation• Objective• Analytical Work
– Introduction to 2-D FEM rolling contact fatigue model– Voronoi tessellations– 2-D crack initiation and total life of fatigue incorporating residual stress
• Experimental Work– Three-ball-on-rod rolling contact fatigue test– Torsion fatigue test
• Summary and Future Work
3
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
Background of Rolling Contact Fatigue (RCF)
Over-rolling componentsRCF in ball bearing (Rosado et al., 2009)
• Fatigue: Failure of a component subject to repeated loads that are often well below the ultimate strength or even yield strength of the material
Subsurface originated spalling Surface originated pitting– micro-cracks originate below the surface– propagation is towards the surface to form a
surface spall– leads to the formation of deep cavities
– cracks initiate at a surface irregularity such as a scratch or dent
– propagation is at a shallow angle until some critical length or depth and branching towards the surface, removing a piece of material
– leads to the formation of shallow craters
RCF in tribo-components occurs by surface and subsurface initiated spalling
Subsurface originated spalling is dominant when the bearing is operating under lubricated conditions and free of any surface irregularities such as scratch or dents or any defects
4
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
Motivation• Retained austenite (RA) does not transform to martensite upon quenching.
The amount of retained austenite has a significant influence on the rolling contact fatigue (RCF) life of steel (SAE 8620)
• In addition to any direct effect on life, retained austenite influences the residual stress (RS) profile, which also affects the RCF life of steel
• There is no general agreement about the effect of the retained austenite on component durability
Should it be at high (>35%) or low (<5%) levels? Is there any optimum choice?
Retained Austenite (light-colored areas) present in a case carburized component (Daniel, 2005)
5
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
Objectives
• Determine the optimum amount or range of retained austenite in SAE 8620 steel for rolling contact fatigue (RCF)
• Investigate how residual stresses profile influence RCF life• Explore the relationship between retained austenite and
residual stresses
6
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
Modeling of Rolling Contact
6b
7b
10b
• All physical materials are discontinuous at some level and failure in bearing contacts originates at a micron scale (comparable to the scale of discontinuities)
• Rolling contact is modeled by moving a Hertzian Pressure (2GPa - width 2b) across the surface in 21 analytical steps
• Induce residual stress (RS) field into the RCF model
b=100μm
Microstructure of steel
7
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
2-D Voronoi Element
• A set of points (seeds, sites, or generators) is specified and for each seed there will be a corresponding region consisting of all points closer to that seed than to any othero The region is thus referred to as a Voronoi cell[1]
Voronoi is a good representation of material microstructure
1b
2b
[1] B. Jalalahmadi, F. Sadeghi, 2009, A Voronoi Finite Element Study of Fatigue Life Scatter in Rolling Contacts, ASME J. Trib., 131(2) (2009).
• 33 domains with different Voronoi mesh are generated to statistically investigate the effects of residual stresses on RCF
8
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
Damage Mechanics
Where 0<D<1
m
R DdN
dD
)1(
• N is number of cycles• Δτ is shear stress reversal along
the grain (Voronoi) boundary• τR and m are material dependent
parameters• τR = 6113MPa• m = 10.0
[2] Robotnov, Y.N., 1969, Creep Problems in Structural Mechanics, North-Holland
[3] Xiao, Y.C., Li, S., Gao, Z., 1998, “A Continuum Damage Mechanics Model for High Cycle Fatigue,” Int J Fatigue, 20(7)
[2]
Elastic Damage Law[3]
• Apply damage law to RCF model
9
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
2-D Weibull Life Plot without RS
Slope of current Weibull plot: Initiation: 7.8 Total: 4.4 (within 0.51 – 5.7 by Harris and Barnsby, 2001) Portion of propagation: 64% (within 60%-80%)
Slope of Jalalahmadi’s result: Initiation: 5.11 Total: 4.08 Slope of Anurag’s result: Initiation: 4.81 Total: 5.13
10
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
2-D Initiation and Total Life Plot
RS type Without RS Linear RS V-shape RS Constant RS
Slope of initiation life 7.8 9.3 9.03 9.29
Slope of propagation life 4.4 3.3 3.7 4.2
Residual stresses have very limited influence on crack initiation life Different kinds of residual stresses have different level of influence on total life Generally, compressive residual stress will increase the total life of RCF
constant residual stress
linear residual stress
V-shape residual stress
Weibull plot for cases with different residual stresses
11
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
Effect of residual stress on life
The 2-parameter Weibull cumulative distribution function, has the explicit equation:
F(t) = Probability of failure at time t;t = time, cycles, miles, or any appropriate parameter;η= characteristic life or scale parameter; also it is the life at which 63.2% failure probabilityβ= slope or shape parameter.
LX means the life at probability of failure X% (0<X<100) Besides L63.2, we also investigate L10 and L50, which are important parameters for RCF life
L10, L50, L63.2 and L90 under residual stresses
Increase of life
RS type Linear RS V-shape RS Constant RS
Max. increase of life 92.3% 99.1% 94.8%
Min. increase of life 1.6% 2.8% 2.8%
Average increase of life 20.1% 35.6% 43.6%
RS type Without RS Linear RS V-shape RS Constant RS
Portion of propagation life 64.0% 69.8% 73.7% 75.4%
12
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
Three-ball-on-rod Test Rig• Federal Mogul three-ball-on-rod RCF machine
LLUBLBBR
LUBLB
L
SSFFF
SFF
WS
28.425tan/2
25tan/
3/
ab)/(1.5FP BR
Loading Principle:
Where a and b are the semi-axes of the contact area[3]
Parameters• Rod (8620 steel)
○ Diameter: 9.5mm (0.374in)
• Roughened Steel Ball○ Diameter: 12.7mm (0.5in)
• Oil○ Turbine oil (MIL-PRF-
23699F)
• Rotation velocity○ 3600 rpm
• Hertzian Pressure○ 3.5 GPa
13
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
Three-ball-on-rod Rolling Contact Test Low RA (RA<5%) Specimen Test Results
hourscycles Lhour
rev
rev
cyclesL
min60
min3600389.2
Currently 16 data points have been recorded Slope of Weibull plot of three-ball-on-rod test: 1.95 (within range 0.51-5.7)
14
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
Torsion TestingExperiment Setup
Custom mechanical interface between MTS rig and specimen
Bearing Steel Torsion Specimen
Custom gripsRotary Actuator Torque cell
MTS Torsion test rig
Objective of this study : To obtain static and fatigue data in shear for modern bearing steels with different amounts of retained austenite
15
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
Torsion Fatigue Test Results
• In torsion fatigue test, 8620 steel with high level of retained austenite has greater life than the one with low level of retained austenite
16
Mechanical Engineering Tribology Laboratory (METL)November 14, 2013
Summary and Future Work
Summary:• Developed and used damage model for 33 domains to research on
the effect of different residual stresses on RCF life• Finished torsion fatigue test for 8620 steels under high and low
RA level • Continued three-ball-on-rod test on rods with low-level retained
austenite
Future Work:• Get more data in three-ball-on-rod test to form the final Weibull
plot• Develop the code to model the crack propagation in RVE• Investigate and initiate the model on effect of retained austenite
on RCF