2020/8/12

路⾯面⼯工程特論
Material Characterization
⿈黃建維

Pavement Design Procedure

u Traffic
u Materials
u Environments
u Distresses
u Performance

1
 2020/8/12

Material Characterization

u Resilient Modulus
u The deformation under each load repetition is
nearly completely recoverable and can be
considered elastic.
u After 100 to 200 repetitions, the strain is
practically all recoverable.
u A nondestructive test and it can be tested under
different loading and environmental conditions
using same sample.

Material Characterization

𝜎%
u Resilient Modulus 𝑀" =
𝜀"

2
 2020/8/12

Material Characterization

u Loading Waveform

Function of vehicle speed and depth

Material properties?

Material Characterization
u Loading Waveform
Haversine and triangular Square Wave Form

3
 2020/8/12

Material Characterization

u Resilient Modulus

The effect of rest period?

0.1 s for loading; 0.9 s for rest period

Internal measuring:
The effect of equipment deformation
End restraint
Piston friction

Material Characterization

u Resilient Modulus
u Granular Materials

Sample conditioning

Test procedure

example

𝑀" = 𝑘( 𝜃*+

4
 2020/8/12

Material Characterization

u Resilient Modulus
u Fine-Grained Materials

Sample conditioning

Test procedure

10 psi is probably largest in a subgrade

example

Material Characterization

u Resilient Modulus
u Asphalt Mixtures
Sample conditioning

Test procedure

Compression test:
4 inches in diameter and 8 in. high
0.1s for loading duration with 0.9s rest period

Indirect tensile test:

4 in. in diameter and 2.5 in. thick
0.1 s for loading with 3s rest

5
 2020/8/12

Material Characterization

u Correlation with Other Tests

u Subgrade Soils

Material Characterization

u Correlation with Other Tests

u Hot Mix Asphalt

6
 2020/8/12

Material Characterization
u Correlation with Other Tests
u Base

Material Characterization

u Correlation with Other Tests

u Subbase

7
 2020/8/12

Material Characterization

u Dynamic Modulus of Bituminous Mixtures

u Dynamic Complex Modulus

Resilient modulus test v.s. Dynamic complex modulus

Frequency is selected as the most closely simulates the actual traffic load

Under tension haversine; tension-compression and compression

Material Characterization
u Dynamic Modulus of Bituminous Mixtures
u Dynamic Complex Modulus

Apply stress under tension haversine; tension-compression and compression

Differences are insignificant among tension, tension-compression and compression

for temperatures 4 to 21 and frequencies 1 to 16 Hz.

Differences are significant among tension, tension-compression and compression

for temperatures 21 to 38 @ frequencies 1 Hz. (1/2to 2/3 of that in compression)

Phase angle are pronounced between tension and compression and less pronounced
between tension-compression and compression. Tension has the greatest value of
phase angle, least in compression.

8
 2020/8/12

Material Characterization

u Dynamic Modulus of Bituminous Mixtures

u Dynamic Stiffness Modulus
A load duration of 0.1 s and a rest period of 0.4 s are
applied at the third points.

Dynamic complex modulus v.s. Dynamic stiffness modulus

Material Characterization

u Fatigue Characterization
u Bituminous Mixtures
Beam test with third-point loading
Apply repeated tension-compression loads in the form
of haversine waves for 0.1s duration with 0.4s rest
period.

9
 2020/8/12

Material Characterization

u Fatigue Characterization
u Bituminous Mixtures

Constant stress is for thicker pavements (more than 6 in.)

Constant strain is for thin pavements (less than 2 in.)

𝑁- = 𝑐/ (𝜀1 )3-+
𝑁- = 𝑐( (𝜀1 )3-+ (𝐸5 )3-6

Material Characterization

u Permanent Deformation Characterization

u Bituminous Mixtures
Basic Principle
The permanent deformation of paving materials
depends strongly on the testing methods and the
procedures used for fabricating and testing the
specimens.

Making the prediction of rut depth extremely difficult.

10
 2020/8/12

Material Characterization

u Permanent Deformation Characterization

u Bituminous Mixtures
Direct Method
1. Divide the pavement and subgrade into a number of
layers
2. Conduct repeated load tests on laboratory
specimens by represented actual stresses of
deviator and radial stress
3. compute the vertical deformation
4. sum the permanent deformation over all layers.

Material Characterization

u Permanent Deformation Characterization

u Bituminous Mixtures
VESYS Method

𝜀7 (𝑁) = 𝜇𝜀(𝑁)39

11
 2020/8/12

Material Characterization

u Other properties
u Poisson Ratio
assume a reasonable value for use in design

Material Characterization

u Nonlinearity
d g 2σ τ
)( )
t
ε (t ) = g 0 D0σ + g1 ∫ ΔD ψ −ψ
t
( d
t

τ
τ
dτ
0
dξ
t
ψt =∫
a aa
0 T s g

g0 is related to the nonlinear instantaneous compliance,

g1 is associated with the nonlinear transient compliance, and
g2 is related to the loading rate effect on nonlinear response

12
 2020/8/12

Material Characterization

Convert Frequency to Time Domain

u Nonlinearity phenomenological model (Prony series)

Time-Temperature Shifting Strain Vertical Shifting

Time-Temperature Master Curve Obtain Nonlinear Parameter g1g2
For Each Strain Level (Nonlinear parameter g1g2)
(Shifting Factors aT )

Time-Strain Horizontal Shifting

Time-Temperature-Strain Master Curve
And fitting by Prony Series
(Prony Series coefficients, aS)

Material Characterization

Nonlinear contribution in the viscoelastic

response due to the level of stress
u Nonlinearity ψ
d
ε = g0 D0σ + g1 ∫ D (ψ t −ψ τ )
ve
g 2σ dτ
0
dτ
Nonlinear contribution in the
elastic response, due to the level Nonlinear contribution
of stress in the transient
response

e ve Related to g 0
q Not well-defined in the
experiments
g 0 ≈ 1.0
q Has a lot of variability

13
 2020/8/12

Material Characterization
Nonlinear contribution in the viscoelastic
response due to the level of stress
u Nonlinearity ψ
d
ε ve = g0 D0σ + g1 ∫ D (ψ t −ψ τ ) g 2σ dτ
0
dτ
Nonlinear contribution in the Nonlinear contribution
elastic response, due to the level in the transient
of stress response

e ve Related to g1g 2

s
q Function of stress level
q Function of confinement
(
g1g 2 = f σ ij , σ c )
t

Material Characterization
Nonlinear contribution
in the transient
u Nonlinearity response
ψ
d
ε ve = g0 D0σ + g1 ∫ D (ψ t −ψ τ ) g 2σ dτ
0
dτ
Nonlinear contribution in the
Nonlinear contribution in the viscoelastic
elastic response, due to the level
response due to the level of stress
of stress

e ve
q Function of stress
Related to g 2 level
q Function of
confinement

(
g 2 = f σ ij , σ c )
t

14
 2020/8/12

Material Characterization
Multiple Stress Creep Recovery test
u Nonlinearity (MSCR)
11 stress levels

10 cycles

…..… …………..………………

9 sec 1 sec Time (Sec)

Strain
1.20E-05
= 0.01% 52C FEM
2.50E-05
52C FEM Strain = 0.04%
52C Experiment 52C Experiment
46C FEM 46C FEM
1.00E-05
46C Experiment 2.00E-05 46C Experiment
40C FEM 40C FEM
40C Experiment
8.00E-06 40C Experiment
27C FEM
27C FEM
27C Experiment 1.50E-05
J(t) (1/kPa)

27C Experiment
J(t) (1/kPa)

6.00E-06

1.00E-05
4.00E-06

2.00E-06
5.00E-06

0.00E+00
0.00E+00 2.00E+01 4.00E+01 6.00E+01 8.00E+01 1.00E+02 1.20E+02 1.40E+02 0.00E+00
Time (Sec) 0.00E+00 2.00E+01 4.00E+01 6.00E+01 8.00E+01 1.00E+02 1.20E+02
Time (Sec)

4.00E-05
3.50E-05 52C FEM
52C FEM
52C Experiment
52C Experiment 3.50E-05 46C FEM
3.00E-05 46C FEM
46C Experiment 46C Experiment
3.00E-05 40C FEM
40C FEM
2.50E-05 40C Experiment
40C Experiment
27C FEM 2.50E-05 27C FEM
27C Experiment 27C Experiment
J(t) (1/kPa)

2.00E-05
J(t) (1/kPa)

2.00E-05
1.50E-05
1.50E-05

1.00E-05
1.00E-05

5.00E-06
5.00E-06

0.00E+00 0.00E+00
0.00E+00 2.00E+01 4.00E+01 6.00E+01 8.00E+01 1.00E+02 1.20E+02 0.00E+00 2.00E+01 4.00E+01 6.00E+01 8.00E+01 1.00E+02 1.20E+02
Time (Sec)
Time (Sec)

Strain = 0.07% Strain = 0.1%

15
 2020/8/12

Material Characterization
Convert Frequency to Time Domain Convert Frequency to Time
phenomenological model (Prony series) Domain phenomenological
For Each Temperature and Stress Level model (Prony series) For
Each Temperature and
u Aging Stress Level

Time-Temperature Shifting Stress-Time

Time-Temperature Master Curve Vertical Shifting
For Each Stress Level Obtain Nonlinear
dξ
t Parameter g1g2 Aging Shifting for each
ψ =∫ t temperature and stress level
a aa
0 T s g
Obtain Aging Shifted Factor

Stress-Time Horizontal Shifting

Time-Temperature-Stress Master Curve
And fitting by Prony Series

Material Parameters:
Prony Series coefficients, Nonlinear parameter g1g2,
Shifting Factors( aT and aS)

Material Characterization

u Aging
2.50E-06 2.50E-06
0.01 FEM
0.01 Experiment
2.00E-06 0.6 FEM
2.00E-06
0.6 Experiment
0.8 FEM
0.8 Experiment
1.50E-06 1.50E-06 1.0 FEM
J(t) (1/Pa)

J(t) (1/Pa)

1.0 Experiment
0.01 FEM
1.00E-06 0.01 Experiment 1.00E-06
0.6 FEM
0.6 Experiment
0.8 FEM
5.00E-07 5.00E-07
0.8 Experiment
1.0 FEM
1.0 Experiment
0.00E+00 0.00E+00
0.0E+00 2.0E+00 4.0E+00 6.0E+00 8.0E+00 1.0E+01 1.2E+01 1.4E+01 0.0E+00 2.0E+00 4.0E+00 6.0E+00 8.0E+00 1.0E+01 1.2E+01 1.4E+01
Time (Sec) Time (Sec)

Unaged Aged

16
 2020/8/12

Material Characterization

u Moisture Air  void

Moisture  diffusion
binder
aggregate

Cohesive Adhesive

Schematic of cohesive and adhesive moisture-induced damage.

Material Characterization

Moisture diffusion contour through the

3D model
u Moisture

Moisture diffusion
contour

17
 2020/8/12

Material Characterization
Model
Model FE
Calibration using
Development Implementation
Lab Data

FE Model for Apply Realistic

Model Validation
Pavement Tire Contact
against Lab Data
Structures Pressure

Performance Post Processing

Simulations of Data

Procedure to Run the Performance Prediction Continuum Damage

Model
Fortran PANDA-UMAT
Fortran compiler is used to compile UMAT A Fortran code includes the Continuum Damage
(i.e. translates programming commands Model
into action).

Abaqus calls UMAT to run the

Continuum Damage Model and UMAT
gives Abaqus the material response

Abaqus interface

Pavement Loaded region

Section

Simulating
Mesh Rutting
Boundary
Condition
sCreating geometry, mesh, and applying Running and viewing the simulation
loading results

18