French methodology for hot bituminous mix design

Yves BROSSEAUD Senior researcher

IFSTTAR – ex LCPC FRANCE

www.ifsttar.fr

Framework

Mix design presentation Method Tests and devices Performances – specifications Pertinence of method : comparison lab/field Conclusion

Method and overview

One material type for each need

Optimized with performance based criteria

In relation with its use on the road

Method used in France since more than 25 years old, so with a very long experience

One laboratory test for One performance

Principle

设计沥青混合料的体积方法

Summarise of French mix design – different steps

Mix design and specifications : level 2 PCG + Water + Rutting RESISTANCE (fabrication-preparation sample-checking)

Marshall test

Preparation of samples in laboratory

Good control quality of mix: composition, voids, homogeneity, ... Accurate and Relevant Tests

Relevant comparison with in situ materials

Mixer BBMAX 80

Plate compactor: 400*600*150 180*500*25 à 100

Vertical gamma Bench EN 12697-35 EN 12697-7

EN 12697-33

Design steps

Selection and identification of components Choice: gradation & binder content

Compactability test (gyratory)

Duriez test

Rutting test

Modulus test

Fatigue test

Formulation selected

Level 1

Level 2

Level 3

Level 4

Compaction

Water sensitivity

Rutting

Stiffness

Fatigue

Standard Performances Lab study Gradation & Binder content designer’s choice

Formulation tool: Gyratory

Compactor (PCG)

Mix design and Composition

0

10

20

30

40

50

60

70

80

90

100

0.1 1 10

Sieve [mm]

Paas

ing

[%]

ECF BBSG BBM BBTM BBUM BBDr

Typical grading curves

MLPC Gyratory shear compactor

Gyratory Shear compactor Standard (NF EN 12697-31) Characterisation of void %

reduction under axial force + gyratory shear

Mix design by adjustment of void content according to product standards

Estimation of site void content Vsite = V(Ne)

Ne nbr of cycle as thickness [mm]

r= 0,95 R= 1,38 (% voids 60g)

Compactability characterisation

Void content

Failed

Failed

Interpretation of gyratory compaction test

Conformity study of a mix in relation to product standard specification for each material type ( NF EN 13 108-1 /2 / 7)

Pass

% voids

Number of passes 2 26 16 8 20

10

5

4 6

8 12 cm

Void content versus layer thickness In site compaction process

Main difference with Superpave interpretation: N initial, N design 4%, N max

表2-9 空隙率规范

Water sensitivity : Duriez test

Standard EN 12697-12 exNFP 98-251-1

Two compaction processes D< 14 mm H 190 mm, 60 kN, 5 min D>14 mm H 270 mm, 180 kN, 5 min Stored at 18 °C, 7 days

in air (50 % moisture) in water

Vertical compression (1 mm/s ) Ratio r/R (and % voids) Repeatability and reproducibility

r = 0,08 R =0,13 (ratio of 0,73)

Decision to use of an adhesion agent

European standard used also indirect tensile test (EN 12697-12)

Level 2 : rutting resistance test

Wheel

Rut depth measurement

Standard (EN 12697-22 ex NFP 98-253-1)

Influence of heavy, slow, channelled traffic under high temperature

relevant correlation with site, repeatability (r = 1,2 et R =1,3)

Test conditions: Smooth tire, pressure 0.6 MPa Load 5 kN, speed 1 cycle/s Controlled temperature 60°C

LPC Wheel tracking test

Rutting

Number of cycles

Average Regression

Typical results with LPC Wheel tracking test

Deformation law = k NB

B

Influence of binder type

Influence of sand nature

Influence of void content

Comparaison Résultats d ’orniérage (EME)

1

10

100

1000 10000 100000Nbre de cycles

% o

rniè

re

France 20 - 30

Pologne 20- 30

Pologne D 50

Heavy compaction Void content : 5%

Surface stability under loads at 60°C, for BBTM

Test with rutting tester

BBTM 0/10 Rut after

3000 cycles : 10,8% voids 11,5%

HSV = 1,4 mm

Rutting resistance - specifications

Test @ 60 °C

0

4

8

12

16

20

BBA 1

BBA 2

BBA 3

BBME 1

BBME 2

BBME 3

GB 2 to 4

EME 1-2

Rut

dep

th [%

]

Surface Base

Number of cycles 10 000 30 000 10 000 30 000

Out of spec

On spec

Level 3 and 4

Mechanical tests: stiffness measurement (direct

tensile test or 2 points bending on trapezoidal samples)

Fatigue resistance test

Determination of bituminous mixes mechanical characteristics for pavement structural design

Need for stiffness characteristics and fatigue resistance : admissible strain for 1 million cycles

Traffic direction

Wearing course

Base Sub-base

Natural soil

Bituminous Layers Thickness t Stiffness E ε, σ (tensile)

Non treated

ε, σ

Mechanical tests

Linear domain (modulus test) Frequencies Domain - dynamic test

(complex modulus in flexion)

temporal Domain - quasi-static test (secant modulus in traction)

Damage domain (fatigue, test in flexion with impose

constant displacement)

Test conditions : Frequencies F [1,3,10,25,30,40 Hz] Temperature 15 °C linear domain low deformation 50*10-6

σ ε

solicitation

time

σ* = σο eiωt ε∗ = εο ei(ωt-ϕ)

σ* = Ε* ε∗

E* =|E*| eiϕ = E1+iE2

Fr, θ

Complex modulus test (NF EN 12697- 26)

h 120 mm Base 40 *40 mm

Complex modulus test

EN 12697-26 ex NFP 98 260-2 2 points bending on trapezoidal samples, 4 repetitions

E pavement design 15°C, 10 Hz Master curve (rheological behavior)

r= 335 MPa , R= 2750 MPa (E = 15300 MPa)

- 1 parallelepiped sample from cut of plate or PCG sample - Control temperature (climate chamber) - Frequencies imposed (vibrator system) -Solicitations in linear domain (ε<50 10-6) (adaptation with solicitation) - Modulus at fixed conditions: 15°C, 10 Hz - Machine « compact » and easy to use

Device to measure modulus with simple system

Stiffness

Modulus @ 15°C: complex (10 Hz)

0 2000 4000 6000 8000

10000 12000 14000 16000 18000 20000

BBA 1-2

BBA 3

BBME 1

BBME 2-3

GB 2- 3

GB 4 EME 1-2

Mod

ulus

[M

Pa]

Out of spec

On spec

Fatigue according to NF EN 12697- 24 - annex A

Service life N

Ln εο

time

time

Displacement resp. (ε)

εο εο

Load resp. (σ)

Fo Fo/2

Ln (duration of life)

106

ε6

Principle of fatigue test

Fatigue law = k Nb

Fatigue test

Standard EN 12697-24 ex NF P 98-261-1 2 points Bending beam on trapezoidal samples

B=56, b=25, t=2, h=250 mm 3 strain levels with 6 specimens each, 10°C and 25 Hz Strain calculated for 1 million of cycles ε6

(better behavior for high ε6)

r = 4,2 µstrain R= 8,3 µstrain

Droite de fatigue

1000

10000

100000

1000000

10000000

10 100 1000

Déformation

No

mb

re d

e cycles

ε6

Modulus and fatigue test

M 2 F apparatus

- Solicitations with imposed displacement – Test: 2 samples tested simultaneously - Control temperature (climate chamber) - Frequencies variables : 5 to 25 HZ (motor with eccentric) -Optional: measurement of complex modulus before test at 15 °C, 10 Hz, ε = 50 10-6

- Electronic data capture allows to help for analyze of test - Tester report with ε6

Fatigue equipment in flexion (M2F) for bituminous mixes

Fatigue test

Admissible strain @ 10 °C and 25 Hz [µstrain]

70

80

90

100

110

120

130

140

150

BBA 1 BBA 2 BBA 3 BBME 1

BBME 2-3

GB 2 GB 3 GB 4 EME 1

EME 2

ε 6 [µ

stra

in]

Out of spec

On spec

Type

of m

ix

Gira

tory

. (V

oids

%)

C80

(D 1

0mm

)C

100

(D 1

4mm

) C

120

(D 2

0 m

m)

Wat

er s

ensi

tivity

r/R ra

tio

Rut

dep

th(6

0°C

-100

mm

)*

10.0

00 c

ycle

s (%

)**

30.

000

cycl

es (%

)

Stiff

ness

mod

ulus

(15°

C-1

0Hz)

in M

Pa

Fatig

ue –

adm

issi

ble

stra

in(@

1 m

illio

n de

cyc

les)

G.BClass 2 ≤ 11 ≥ 0.65 ≤ 10* ≥ 9,000 ≥ 80.10-6

GBClass 3 ≤ 10 ≥ 0.7 ≤ 10* ≥ 9,000 ≥ 90.10-6

GBClass 4 ≤ 9 ≥ 0.7 ≤ 10** ≥ 11,000 ≥ 100.10-6

EMEClass 1 ≤ 10 ≥ 0.7 ≤ 7.5** ≥ 14,000 ≥ 100.10-6

EMEclass 2 ≤ 6 ≥ 0.75 ≤ 7.5** ≥ 14,000 ≥ 130.10-6

Base Hot Mix Asphalt : main performances

fundamental empirical

Summarise of mechanical properties used in pavement design

Modulus E 15°C 10 Hz Fatigue ε 10°C 25 Hz

Volume composition : voids content % Performances : Surface or wearing course wear resistance in humidity conditions rutting resistance evolution of surface characteristics (texture)

Structure or base course (thickness > 5 cm) modulus properties fatigue resistance rutting behaviour

Ranking of hot bituminous mixes

Properties determined in laboratory and some controls in place

Need for the study of laboratory

Starting from components representative of the job site and in conformity with the standards of tests of mix design, It can be possible to have a good control on:

mix design, manufacture, preparation of the sample,

with the accuracy of the means of laboratory. method optimization of the mixture authorizes, with controlled

parameter, it is thus selective, method is moreover repeatable, the given properties are relevant compared to those obtained

on average on a well led job site, the specifications of the products taken as reference for the

study are realistic compared to the behaviors awaited in place.

Path lab to field?

Assessment of link between lab to field: -PCG -Rutting test -Modulus -Fatigue

Accuracy, Selectivity, Pertinence

Classification in lab / Behavior in place Comparison between different lab Round robin test : repeatability, reproducibility Tests in different labs

How can we determined specifications?

GSC : dispersion in field (N7 Base)

Preliminary study

checking

GSC (average 32)

Preliminary study no representative Bonne correspondence : checking and average of field Low dispersion of workability : range maxi 5 to 9%

GSC : prediction voids % on field

PCG 100 girations Field density

(nuclear) GDF 5 cm

Same ordre of magnitude GSC-Field

Dispersion field = 2*GSC

N 149 Bressuire Fondation 10 cm GB3 0/14 disc. 6/10

GSC: lab – field - specifications

0%

5%

10%

PCG mobile Étendue Moyenne

GB2 0/20 GB3 0/20 GB3 0/14 GB3 0/14 EME2 0/20 BBSG1 0/10

Synthèse

Pb de compactage

Chantier Moyenne

Specific studies with LPC rutting test

Correspondence lab-field (Dot Colorado -USA) Previous standardization tests studies Comparison rutting tester english-french Behaviour of different asphalt mixes in europe with

LPC rutting tester Researches : correspondences lab-field Studies on fatigue test track in Nantes (Seattle) Research experience (mix design guide)

DOT Colorado study

33 job sites : crossing traffic, ambiance temperature

Results on sites With high temperature

Results on sites At moderated temperature

DOT Colorado study

Results on sites With low temperature

DOT Colorado study

ranging : behaviour site / LPC criteria DOT Colorado study

Rutting Prediction

Excellent correlation between rutting prevision with LPC equipment and on job site, if taking into account : heavy traffic and temperature

Conclusion de l’étude

Conclusion of Dot Colorado study

Comparison AC with or without Polymers

-18-16-14-12-10

-8-6-4-20

0 5000 10000 15000 20000 25000Number of Cycles

Def

orm

atio

n, m

m

SB Modifié BB-20

BB Pur-20

Influence of nature of bitumen

GSC –plate compactor rutting : comparison field - lab

N 149 GB3 0/14 disc.

Noubleau

Prélèvements chantiers

Sample from field low relation G SC–PC (high) Rut : field < Lab no significative

% de vides % d’ornière

Rut : prediction with BB with bad rutting resistance

difference one decade Field more

stable than in lab

Range does

not change

Dispersion of complex modulus : example of results in place on motorway

Low variability on each sample range : Maximum 1300 MPa Minimum 80 MPa Significative difference between each point

80001000012000140001600018000

0 5 10 15 20 25N° de prélèvement

|E*|

(MPa

)

Eprouvette 1Eprouvette 2Eprouvette 3

A 39 lot Monnières, |E*| 15°C, 10 Hz

Conclusions modulus lab/field

Average in field same order of magnitude as in lab.

high dispersions of mechanical performances on field (20 à 30%) many determination o field (20) to validate study.

Lab study ESSENTIAL.

Factors of influence : % voids and properties of bitumen reclaimed, explain the field dispersions.

Modulus

Angle

Properties of bitumen guide properties of AC

Effect of aggregate skeleton

Characterisation of AC

master curve at 10°C : modulus and phase angle

Grade bitumen influence: 20/30 – 50/70

BITUMEN

AC

courbe maîtresse à 10°C BBSG avec-sans EVA

100

1000

10000

100000

1E-08 0,000001 0,0001 0,01 1 100 10000 1000000

Fréquence équivalente (Hz)

|E*|

(MPa

)

-10°C

0°C

10°C

20°C

30°C

40°C

Arrhénius(deltah=214kJ/ l )

Characterisation of AC with polymers

Master curve at 10°C : modulus and phase angle

Variability of fatigue properties synthesis of results

Note : when modulus high, fatigue is a little low, inverse too

Déformation constante

LRPC Bordeaux LCPC

BB B Pur

BB M BmP

EME Dur

BB B Pur

BB M BmP

EME Dur

V % 5,2 2,9 3,8 5,1 4,1 Module (MPa) 8300 6800 13500 8200 6500 12900

continu ε6

pente σx/y

147 ±7

-5,3

237 ±13

-5,5

133 ±4

-5,3

130 ±6

-5,3 0,615

217 ±6

-7,5 0,499

123 ±4

-5,8 0,443

discontinu 1:10

ε6 pente

227 ± 18 -6,2

187±10 -7,4

0

50

100

150

200

250

Pur BmP Dur

LR BordeauxLCPC

Characterisation of AC polymers

Behaviour in fatigue with continue or discontinue constant

deformation

comparison Pure, PmB, Hard

Discontinue Continue

Field properties dispersion

Range of dispersion: GSC

Base mixes ± 2 à 2,5 points Surface mixes ± 1 à 1,5 points

Modulus ± 20 à 30 % Fatigue ± 10 à 15 % Rut : around 2 points with BB presented a good

rutting resistance (< 5% 30000 cycles)

On job sites where the rules of fabrication and placement were corrected apply.

conclusion

Well-known well used by road technicians Free method to improve the quality in term of

performances requirements, even for Road materials (innovative efforts, quality system,..) Road structures (allowable for private free variant, but same rules

for all,…)

A good feed back with the job site Specifications : realistic, relevant, Statistical method : take into account variation due to

construction, and dispersion of material, adjustment coefficient Method admitting by European standardization :

road material mix design, laboratory characterization tests, in future : pavement design procedure (harmonization to do)

Thanks a lot for your attention