Florida Top Down Cracking Pavement Design Model

Topic Description

This presentation provides an overview of a new mechanistic-empirical top down cracking design tool for Florida.

Speaker Biography

Finished B.S.C.E. in Civil Engineering and B.S. in Mathematics from the University of North Dakota, 1986. Completed M.S. in Civil Engineering from Cornell U. and a Ph.D. in Civil Engineering from the University of Minnesota in 1996. Worked as a Civil Engineering consultant for 4+ years, joined the University of Minnesota in 1997, and the University of Florida Pavement Group in 1998. Dr. Birgisson has written over 80 papers on pavements and pavement related subjects.

Bjorn Birgisson

Session 52

University of Florida

1

2006 Design ConferenceDesigning for More Than Bridges and Roads July 31 August 2, 2006

Florida Top-Down Pavement Cracking Design Model

Bjorn Bjorn BirgissonBirgisson, , JianlinJianlin Wang, Reynaldo Wang, Reynaldo RoqueRoque

Department of Civil & Coastal Engineering University of Florida

Pavement Response and Cracking

Bottom-up cracking

Predominant in Florida

Top-down cracking

Bending effects

2

FDOT Multiyear Study

Mechanisms of Top-Down Cracking

Stiffness Gradients (Temperature differential, Aging)

Thermal Stresses

Truck tire ribs induced tension, Residual viscoelastic stresses

Cracking Models for Mixtures and Pavement

Simpler Testing and Design Calculations

Core Extracted from Field

3

A damage threshold exists (DCSE limit)

Damage = Dissipated Creep Strain Energy (DCSE)

Damage > Threshold Macro-crack(DCSE) (DCSE limit)

Macro-crack is not healable

Damage under the cracking threshold is fully healable

Florida Cracking Model Key Features

Ener

gy, E

Cycles, N

Ener

gy, E

Cycles, N

Ener

gy, E

Cycles, N

FE

DCSE

creep

xelastic elastic

Potential loading conditions in the field

Ener

gy, E

FE

DCSEDamage Healing

Day 1 Day 2

Damage Healing

x

The Threshold Concept

4

Crack Propagation (Paris Law) C

rack

Len

gth,

a

Number of Load Applications, N

dNdA

a =

Threshold

Macro-crack initiationMicro-crack Threshold

Macro-crack

Micro-crack

Crack propagation in asphalt pavements occurs in steps.

Crack Growth Model

Superpave Indirect Tensile Test:

H

V

Mixture Properties

1. Resilient modulus (Cyclic loading)

2. Creep (Constant load with time)

3. Strength (Increase load until fracture)

Dissipated energy creep rate

Energy limits

5

to calculate the amount of dissipated energy per load cycle:

HMA Fracture Model

1/ (tensile stress, D &m)DCSE cycle f=

Calculate the crack growth for a given level of applied stress.

Use Material properties m, D1 (creep rate) & DCSEf (energy limit) Structural properties AVE (modulus)

For a given mixture with known DCSEf we can predict Nffor initiation or propagation of cracking.

Multiple pairs of poor and good performing sections throughout Florida

Field Test Sections

Over 18 pairs (36 sections) to date

6

Used the HMA Fracture Model to calculate Nf for crack to propagate 2

0

2000

4000

6000

8000

10000

12000

14000

SR80

-1C

SR 16

-4C

SR 16

-6C

NW 39

-1C

TPK

2CI75

-3CI75

-1C

SR 37

5-2C

I75-1U

I75-2U

TPK

1U

NW 39

-2U

SR80

-2U

SR 37

5-1U

Section

Nf t

o Pr

opag

ate

2 in

`

Cracked Uncracked

Mixtures with Nf

7

DCSEmin is the minimum energy required to produce Nf=6000

Express the DCSEmin, D1 & m-value relation in a single function:

ADm

DCSE 12.98

min =

( ) 81.3 1046.244.33

36.6 +

=t

t

S

A

Minimum Energy

Tensile Strength

Tensile Stress

The DCSEHMA has to be greater than the DCSEmin for good cracking performance:

DCSEHMA DCSEmin

MR

Stre

ss,

Strain,

DCSE

x

DCSEHMA = AREA

Log

D(t

)

Log t1

D1

m

2.98min 1DCSE m D / A=

Energy Ratio Concept

1DCSEDCSERATIO ENERGY

min

HMA >=

8

Examined all sections Performance criteria: ER>1 ; DCSEHMA>0.75

0.00

0.50

1.00

1.50

2.00

2.50

3.00

US30

1-BS

SR80

-1C

I10-D

E

I10-D

W

SR 16

-4C

I10-M

W2

SR 16

-6C

NW 39

-1C

TPK

2CI75

-3CI75

-1C

SR 37

5-2C

US30

1-BN

I95-SJ

N

I10-M

W1

I75-1U

I75-2U

TPK

1U

SR80

-2U

NW 39

-2U

SR 37

5-1U

I95-D

N

Section

Ene

rgy

Rat

io

Cracked Uncracked

DCSEHMA2.5

Energy Ratio Results

Florida Framework for Cracking Evaluation of HMA Pavements

LABORATORY TESTSuperPave IDT

SUPERPAVE IDT FRACTURE PARAMETERS

HMA FRACTURE

MODEL

PAVEMENT RESPONSE

CRACKING MODELS

PAVEMENT DESIGN THICKNESS

&

VOLUMETRIC RELATIONOR

STRUCTUREINFORMATION

9

Accounts for structure and mixture for averaged environmental conditions

Design Premise: ensure a reasonable predicted crack depth after x number of years (Design Life)

Top-Down Cracking Design

Use Energy Ratio for M-E Top-Down Cracking Design

Level 3:

Determine thickness for ER=1 @ design life

AC ThicknessModulus, Poissons Ratio

Layered Elastic Analysis

Stress

Energy Ratio

ER 1 Design Thickness

Mixture Properties Matrix(DCSEL, FE, St, Creep Rate)

M-E Design Flowchart Level 3

no yes

Aging model w/ Design life

10

Uses a fast pavement fracture simulator to predict depth of cracking after x years (Design life), and account for the effects of:

Mixture & Structure Temperature/aging gradients Load Configuration Traffic

M-E Top Down Cracking DesignLevel 1 & 2

Level 1: Measured properties from IDT

Level 2: Estimated properties

Updated Layer Thickness

Load Configuration Load Spectra? ESALs?

Fracture Simulator

GRADIENTS: Temperature (Matrix)Aging (Matrix)

Crack Depth at Design life (CD)DL

Traffic, n

Design Thickness

DCSEL, FE, St, Creep Rate,Mr, adjusted for aging andtemperature

LEVEL 1 Layer Thickness

M-E Design Flowchart Level 2

no yes(CD)DL 2

11

New Pavement Design using Energy Ratio

Overlay Design using Energy Ratio

Design Studio

Input Menu

Structure

AC Layer Other Layers

Loads

MAAT

12

AC Layer Basic Design

Suggested gradation or input your own

Estimate an initial design thickness (will be optimized)

Input Design AV and Vb

Binder Selection

Estimate the binder viscosity at mix/laydowncondition

Predict the in-service viscosity based on the global aging model

Correction

0.76 0.78 0.8 0.82 0.84 0.860.76

0.77

0.78

0.79

0.8

0.81

0.82

0.83

0.84

0.85

0.86Viscosity of Layer A and Layer B at 60 oC

Measured log-log (cp)

Cal

cula

ted

log-

log

(c

p)

Layer A (uncorrected)Layer B (uncorrected)Layer A (corrected)Layer B (corrected)

13

Input MAAT to predict aging

MAAT Input

Mixture Properties

Layer modulus estimate from |E*| master curve

Poissons ratio estimated from EAC

IDT parameters estimated from some basic relations

-20 -15 -10 -5 0 5 10 15 202.5

3

3.5

4

4.5

5

5.5

6

6.5

7

7.5

8Master-Curve for |E*| at 10 oC (C1 mixture)

log-frequency (Hz)

log

|E*|

(psi

)

14

Pavement Structure Other Layers

Base, sub-base and subgrade layers (elastic)

User can determine the number of layers

Default properties suggested for selected materials

User can also define the material properties

Simplified load information

Loads Input

1(1 )(1 ) 1

n

n total n

p pS Sp

+=

+

The maximum numberof ESALs/year:

totalS

n

(%)p

15

ER formula

, where

ER Calculation

0.0

0.5

1.0

1.5

2.0

2.5

0 200 400 600 800 1000 1200

Traffic (ESALs/year x1000)

Min

imum

Ene

rgy

Rat

io

Minimum ER adjusted for traffic level

According to the

traffic load input,

select the minimum

(optimum) ER

corresponding to

the traffic level

stresses at the standard

or selected location

ER value

ER Output

16

Search for the thickness that gives the minimum required ER

Thickness Optimization

1

2

1

2

t2

t1

Plot ER-thickness curves at different pavement ages

ER increases as the pavement thickness increase if the pavement is not too thin

The ER values for new and aged pavements differ significantly, especially for thick pavements

0 t1

17

Plot pavement life curves for different thicknesses

ER drops down significantly in the first couple of years

Sensitivity of ER to thickness is shown in the graph

Pavement Life Curve

h2

h1

h3

h1 < h2 < h3:

h2 = optimum thickness

h1 = h2 - 2

h3 = h2 + 2

A new M-E pavement design tool for top down cracking based on Energy Ratio

Summary

Validated on more than 30 field sections

Thickness design optimized for

traffic level

mixture type

binder type

The optimization is an automated process

18

Level 1 and 2 pavement design tool being developed

Summary (Contd)

Frame work complete

fast fracture simulator

Awarded to UF on May 1, 2006

NCHRP 1-42A: Models for top-down cracking

Questions