Session 40 Ebrahim Parhamifar

Development of permanent deformations inunbound base course materials- Analysis and modeling

Ebrahim Parhamifar, LTHSigurdur Erlingsson, VTI

Transportforum 2010

Objectives

The aim of the work is to introduce a new test method tostudy response and permanent deformation of unboundmaterials.

The test results are compared with some establishedmodels.

The comparison is used to build up a data base for materialparameters.

Testing box

• 800*800*800 mm testing box• Deformation gauges at 10 points and two levels

A

B

A: 80 or 150 mm crushed 0-40 mm base material

B: 200 mm of natural 0-100 mm sub-base material

Testing Box

Instrumentation of the specimen

A pavement structure

Testing box Triaxial

Applied stresses on body of a specimen

V

H

VH

H

V

Experiments in the testing box

• Six 0-40 mm unbound granular materials were used in experiments

• Variation in vertical applied stresses to simulate different traffic loadings on real roads and under pavement construction

• Two series of measurements- Short series to study material response- Long series to study permanent deformations• Different frequencies to simulate different vehicle

speeds

Typical loading cycle in the testing

box

Load (kN)

Load (kN)

Time of loadingOne loading cycle

Two typical crushed unbound granular

materials used in experiments

Material / Properties

d1 d2 d3 d4 d5 d6

Dry density (g/cm3 ) 2.08 2.22 2.20 2.29 2.30 2.25Optimum moisture content ( % weight)

7.7 5.3 7.0 6.2 5.7 6.0

Particle density>(g/cm3 )

2.73 2.60 2.73 2.78 2.73 2.57

Flakiness ratio 1.51 1.58 1.52 1.56 1.55 1.37Aggregate impact value

52 62 57 56 57 55

Nordic Ball Mill Value

23.5 25.8 20.6 15.3 15.7 25.1

% natural aggregate

0 0 0 0 0 39 /20*

Materials used for measurements in the testing box

Simulated distribution of the vertical stress in the texting box

Symmetry lines

Response measurements.Elastic strain with different vertical

stresses

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0 5 10 15 20 25 30

number of load applications

axia

l ela

stic

str

ain (m

m/m

)

212 kPa 424 kPa 637 kPa 849 kPa

Material characterization K- model

y = 21.717Q0.3865

0

50

100

150

200

250

300

350

400

0 200 400 600 800 1000 1200 1400 1600

Q (kPa)

Mr (M

Pa)

calculated values measured values

21

kR kM

Comparison between measurements and calculations in the middle of the base course

0

0.4

0.8

1.2

0 10 20 30Distance (cm)

Ela

sti

c d

efo

rma

tio

n (

mm

)

measurements KenPave

p = 477 kPa

0

0.4

0.8

1.2


Ela

sti

c d

efo

rma

tio

n (

mm

)


p = 955 kPa

0

0.4

0.8

1.2

0 10 20 30

Distance (cm)

Ela

sti

c d

efo

rma

tio

n (

mm

)


p = 1432 kPa

0

0.4

0.8

1.2


Ela

sti

c d

efo

rma

tio

n (

mm

)measurements KenPave

p = 1910 kPa

40 mm depth

Comparison between measurements and calculations at layers boundaries

0

0.4

0.8

1.2

0 10 20 30

Distance (cm)

Ela

sti

c d

efo

rma

tio

n (

mm

)


0

0.4

0.8

1.2


Ela

sti

c d

efo

rma

tio

n (

mm

)


p = 955 kPa

0

0.4

0.8

1.2


Ela

sti

c d

efo

rma

tio

n (

mm

)


p = 1432 kPa

0

0.4

0.8

1.2


Ela

sti

c d

efo

rma

tio

n (

mm

)measurements KenPave

p = 1910 kPa

p = 477 kPa

80 mm depth

Permanent deformation models

kPa cohesion, isc

angle friction is

stress deviatoric q is

ratio failure isR

parametersmodel are

36

36

q

0

0f

bandC

sincos..c

q

sinsin.

M

p.Mqqq

R

N.RA

R.C

r

bp

Leena K. T. 2009

parametersmodel are S and B m, n,

11

01

22

01

,

qpLpq

pS

m

.pL

).N.(

p

maxmaxmax

max

max

max

n

a

maxBpp

Gidel et. al 2001

parametersmodel are and ,0

0

Np e.N

Tseng & Lytton1989

Comparison between measurements and calculations in the middle of the base course

0

0.5

1

1.5

2

0 50000 100000 150000 200000number of load applications

per

man

ent

def

orm

atio

n (m

m)

measured Leena Gidel Tseng&Lytton

0

0.5

1

1.5

2

0 20000 40000 60000 80000 100000 120000


per

man

ent

def

orm

atio

n (m

m)

measured Leena Gedel Tseng&Lytton

P=396 kPa

P=566 kPa

0

1

2

3

4

5

6

7

0 20000 40000 60000 80000


per

man

ent

def

orm

atio

n (m

m)


0

1

2

3

4

5

6

7

0 20000 40000 60000 80000


per

man

ent

def

orm

atio

n (m

m)


P=1431kPa

P=955 kPa

Comparison between measurements and calculations at layers boundaries

0

0.5

1

1.5

2

0 50000 100000 150000 200000number of load applications

per

man

ent

def

orm

atio

n (m

m)

measured Leena Gidel Tseng & Lytton

0

0.5

1

1.5

2

0 20000 40000 60000 80000 100000 120000


per

man

ent

def

orm

atio

n (m

m)


P= 396 kPa

P= 566 kPa

0

1

2

3

4

5

6

7

0 20000 40000 60000 80000


per

man

ent

def

orm

atio

n (m

m)


0

1

2

3

4

5

6

7

0 20000 40000 60000 80000


per

man

ent

def

orm

atio

n (m

m)


P=955 kPa

p=1431 kPa

Material parameters Material Parameters (Lenna)

c b C*10-5 d1 40 60 0.3 35 Material Parameters (Tseng&Lytton)

*103

0 3 d labbel *10-3 labbel

0 d1

5 0.4 0.012 100 500 52 2.27 Material Parameters (Gidel)

m S pp1

*10-3 B n d1

2.43 97.39 15 0.3 0.6

Conclusions

• The texting box used to study material properties of the unbound materials show some interesting results compared to established methods

• The response behaviour observed in the testing box shows very similar results to numerical calculations (KenPave) using the K- model (stress dependent).

• The observed development of accumulated permanent deformation in the testing box is very similar to observed deformations in pavements.

• All models used to predict the permanent deformation show acceptable agreements with the measured values. Leena's model show the best correlation with measured values at both low and high stress levels

Session 40 Ebrahim Parhamifar

Technology

testing box instrumentation

testing box deformation

base course p

subbase material

material parameters

comparison betweenmeasurements

layers boundaries p

texting box symmetry