Cumulative Fatigue Damage Analysis of Concrete Pavement Using APT Results

1
Cumulative Fatigue Damage Analysis of Concrete
Pavement Using APT Results

• Shreenath Rao
• Jeffery Roesler
• University of Illinois

Second International Conference on Accelerated
Pavement Testing, Minneapolis, MN, September
26-29, 2004
2
Acknowledgements

• Caltrans
• University of California, Berkeley
• Dynatest, Inc.
• CSIR

3
Presentation Outline

• Overview and Background
• Cumulative Fatigue Damage Models
• Field APT Project Background
• Data Analysis and Results
• Discussion and Conclusions

4
Fatigue Cracking

• Key failure mechanism in rigid pavements
• Result of repeated loading
• Modeled in mechanistic-empirical design
  procedures by cumulating damage at locations of
  critical stresses

5
Miners Fatigue Damage Accumulation Hypothesis

• Each load application causes damage at location
  of critical stress
• Amount of damage due to one load application is

6
Critical Stress Locations





7
Fatigue Crack Corner Break 1,000 reps.
(Section 525FD)
45 kN, 150-mm nom. thickness
8
Miners Fatigue Damage Accumulation Hypothesis

• Allowable number of load applications is a
  function of
• Stress ratio changes with each load application
  and is affected by a number of factors

9
Fatigue Models Tested

• Zero-Maintenance Design Beam Fatigue Model
• log N 17.61 17.61 SR
• Calibrated Mechanistic Design Field Fatigue Model

P Cracking Probability
10
Fatigue Models Tested

• ERES/COE Field Fatigue Model
• log N 2.13 SR-1.2
• Foxworthy Field Fatigue Model

11
Fatigue Models Tested

• PCA Beam Fatigue Model
• log N 11.737 - 12.077 SR
• 2002 Design Guide Field Fatigue Model
• log N 2.0 SR-1.22

12
Transfer Functions for Cracking

• Damage is related to cracking observed in the
  field through the use of transfer functions such
  as

a and b are field calibration coefficients
13
Project Background

• Heavy Vehicle Simulator (HVS) used to test high
  early-strength fast-setting hydraulic cement
  concrete (FSHCC)
• Two full-scale test pavement strips
• North Tangent and South Tangent
• Each approx. 210 m in length
• State Route 14
• 8 km south of Palmdale, California

14
Heavy Vehicle Simulator
15
Typical Instrumentation Layout
16
South Tangent Test Sections

• Main objective - Evaluation of the fatigue
  behavior of 100-mm, 150-mm, and 200-mm FSHCC
  slabs
• 150-mm Class 2 aggregate base
• No dowel bars, tie bars, or widened lane
• Slab widths - 3.7 m
• Joint spacing between 3.7 m and 5.8 m

17
South Tangent Test Sections

• Bi-directional creep speed (2.0 6.5 km/hr) HVS
  wheel loads
• HVS on edge of concrete slab no wander
• Loading (20 100 kN) varied from section to
  section
• Temperature control box around the tested area
  for some sections

13 test sections from 519FD to 531FD
18
North Tangent Test Sections

• Main objective - Evaluation of the fatigue
  behavior of 200-mm FSHCC slabs with various
  design considerations
• 100-mm cement-treated base
• 150-mm Class 2 aggregate subbase
• Three designs
• No dowel bars, AC shoulder
• Dowel bars, tied PCC shoulder
• Dowel bars, AC shoulder, 0.6 m widened lane
• Slab widths - 3.7 m and 4.3 m

19
North Tangent Test Sections

• Joint spacing between 3.7 m and 5.8 m
• Uni- and bi-directional creep speed (2.0 km/hr)
  and 10 km/hr HVS wheel loads
• No wander of HVS wheel load
• Loading (40 90 kN dual wheel and 70 150 kN
  aircraft) varied from section to section
• Temperature control box around the tested area
  for some sections

10 test sections from 532FD to 541FD
20
Fatigue Analysis

• Several critical locations on the test slabs were
  analyzed for cumulative fatigue damage
• Fatigue failure ? number of repetitions to first
  crack (longitudinal, corner, transverse)
• Stresses calculated for each load increment to
  failure using FE program (ISLAB2000)
• Top-down cracking critical because of high
  effective built-in gradients and temperature
  control box

21
Key Inputs for Stress Computations Using ISLAB2000

• Slab dimensions layer thicknesses, joint
  spacing, slab width
• Layer material properties elastic modulus,
  coefficient of thermal expansion, density
• Subgrade modulus of subgrade reaction
• Joint and lane-shoulder LTE
• Thermal gradients through slab
• Effective built-in curling - construction curling
  shrinkage warping
• Load magnitude, location, tire pressure, axle
  configuration

22
Unloaded Stress Distribution (Top) - Section 520FD
Stresses in Transverse Direction
23
Stress Distribution (Top) Due to 35kN Dynamic
Load - Section 520FD
Stresses in Transverse Direction
24
Unloaded Stress Distribution - Section 520FD
Stresses in Longitudinal Direction
25
Stress Distribution (Top) Due to 35kN Dynamic
Load - Section 520FD
Stresses in Longitudinal Direction
26
Stress Distribution (Top) Due to 35kN Dynamic
Load - Section 520FD
Stresses in Longitudinal Direction
27
Influence Chart for Moving Load Section 520FD
Stress, MPa
psi
A
B
35 kN
100-mm nom.
28
Influence Chart for Moving Load Section 520FD
Stress, MPa
psi
35 kN
100-mm nom.
29
Influence Chart for Moving Load Section 520FD
Stress, MPa
psi
35 kN
100-mm nom.
30
Influence Chart Analysis Summary
535FD, critical transverse joint location (1.2 m
from left slab corner) using half-axle 90kN load
31
Influence Chart Analysis Summary
535FD, critical lane-shoulder joint location (1.5
m from left slab corner) using half-axle 90kN
load
32
Cumulative Damage Peak Stresses, Linear
Curling, Beam Strength
33
Cumulative Damage Peak Stresses, Nonlinear
Curling, Beam Strength
34
Cumulative Damage Peak Stresses, Nonlinear
Curling, Slab Strength
35
Fatigue Models
36
Discussion

• Concrete Fatigue Models
• Strength Variability
• Strength Loss Early Age Restrained Cracking
• Beam Strength vs. Slab Strength
• Stress Location vs. Stress Field
• Miners Hypothesis Limiting Assumptions
• Stress history, loading rate, stress reversals,
  rest periods

37
Stress Distribution (Top) Due to 35kN Dynamic
Load - Section 520FD
Stresses in Transverse Direction
38
Stress Distribution (Top) Due to 35kN Dynamic
Load - Section 520FD
Stresses in Longitudinal Direction
39
Stress Distribution (Top) Due to 35kN Dynamic
Load - Section 520FD
Stresses in Longitudinal Direction
40
Conclusions

• Miners law for damage accumulation tested 6
  fatigue models
• Large variability in accumulated damage to
  observed cracking
• Alternative approach to Miners Hypothesis needed
  to account for
• materials fracture properties
• size effect
• various loading effects