5th INTERNATIONAL CONFERENCE

BITUMINOUS MIXTURES AND PAVEMENTS

Thessaloniki, Greece, 1-3 June 2011

INSTRUMENTED TEST SECTION FOR THE EVALUATION OF

GEOGRIDS IN ASPHALT PAVEMENTS

A. Graziani *
Assistant Professor, Università Politecnica delle Marche, IT
A. Virgili
Professor, Università Politecnica delle Marche, IT
L. Belogi
PhD student, Università Politecnica delle Marche, IT
*
Dipartimento di Idraulica, Strade, Ambiente e Chimica. 60131 Ancona, Italy,
a.graziani@univpm.it

ABSTRACT
The evaluation of geogrid reinforcement for asphalt pavement rehabilitation is a
complex task where considerable research is needed. An instrumented flexible
pavement test section was constructed along an in-service road in Ancona
(Italy) to study the effectiveness of geogrid reinforcement in terms of stress-
strain response and pavement performance. The instrumented pavement is part
of a larger project that also involves a RILEM inter-laboratory test of geogrids
in asphalt pavements. A glass fiber grid and a carbon fiber grid were installed at
the interface between two asphalt layers. An unreinforced section with a
traditional tack-coat interface was constructed for reference. Moreover, areas
with artificial cracks and partial debonding were prepared in the lower asphalt
layer to simulate challenging situations frequently encountered in practical
rehabilitation projects. The three test section were instrumented with pressure
cells, asphalt strain gauges and temperature sensors. This paper describes the
site preparation and the test sections construction phases, with particular
emphasis to the instrumentation and the data acquisition set-up. FWD and full-
scale tests were performed on the newly constructed pavement. Results show
the effects of geogrid reinforcement on the pavement response both in terms of
surface deflection and stress-strain response. They will be the benchmark for
the planned long-term monitoring of the test sections.

KEYWORDS: rehabilitation, geogrid reinforcement, instrumentation, pressure

cell, asphalt strain gauges.

1
1. INTRODUCTION

Reinforcement techniques have an important role in asphalt pavement

construction, maintenance and rehabilitation. The basic principle is to insert a
thin reinforcement layer inside the pavement to improve its performance and
extend its service life.
Geosynthetics, in particular geotextyles and biaxial geogrids, are widely
used as reinforcing materials. Geogrid reinforcement, beneath or inside asphalt
concrete layers, can lead to many positive effects including: limitation of
reflective cracking, enhancement of fatigue resistance and reduction of rutting
and permanent deformation [1, 2, 3]. An important aspect to be considered is
the geogrid influence on interface properties. The reinforcement benefits can in
fact be concealed if the geogrid causes a significant shear strength reduction at
the interface between asphalt layers [2, 4, 5].
Laboratory tests of geogrids in asphalt pavements are commonly used for the
assessment of reinforcement behavior [2, 6, 7]. However, full scale test sections
are invaluable to properly evaluate scale effects, installation methods and the
overall pavement performance under traffic and climate actions. Although a
number of important experiences can be found in the literature, these are mostly
related to geogrid applications as unbound base and subbase reinforcement [8,
9, 10, 11].
This paper describes the construction and instrumentation of a full-scale
pavement test section to study the effectiveness of geogrid reinforcement in
asphalt layers. A carbon fiber (CF) and a glass fiber reinforced polymer (FRP)
geogrid were installed inside an asphalt “double-layer” surfacing. A reference
section with a conventional tack-coat interface was also instrumented to obtain
control data. Pressure cells and asphalt strain gauges were employed to measure
the load distribution effect of the reinforcement, and its influence on the
interface behavior in terms of horizontal strains. The objective is to relate
pavement performance under real traffic conditions to the mechanical response
measured by the instruments inside the pavement and to obtain data that could
be useful for analytical modeling of reinforced pavements.

2. PROJECT DESCRIPTION

2.1 Background
The full-scale instrumented test section described in this paper is part of a larger
research project on “Advanced Interface Testing of Geogrids in Asphalt
Pavements”. The project originated inside the TG 4 “Pavement Performance
Prediction and Evaluation” of RILEM TC 206/ATB “Advanced Testing and
Characterization of Bituminous Materials” and comprises the construction of
two full-scale pavement test sections. The first section was used to prepare
double-layer reinforced asphalt samples using real scale paving equipment and
geogrid installation techniques. Specimens cut from this section will be studied

2
in an interlaboratory test organized by RILEM and focused on the comparison
of laboratory procedures and devices for the mechanical characterization of
geogrid reinforced interfaces in asphalt concrete pavements. The second
section, that will be described hereafter, was designed constructed and
instrumented to perform a real scale analysis of the same interface
configurations. The parameters (stress, strain and temperature) measured in this
section will also give feedback for the evaluation of the test results.

2.2 Instrumented pavement test section

The instrumented pavement test section is located along an existing secondary
road, inside an industrial area, near the city of Ancona (Italy). Next to the test
section, the road ends into the storage area of a steel frame factory. Therefore,
most of the traffic is composed by heavy trucks that can be precisely counted
and weighted. Two geogrid types were installed inside an asphalt double-layer
surfacing:
1. a Glass Fiber Reinforced Polymer geogrid (FP);
2. a Carbon Fiber geogrid (CF).
A reference sub-section with a conventional tack-coat interface was used to
obtain control data. Each sub-section is 15 m long and 5 m wide, the overall
road width (2 lanes) is 9.5 m, with law concrete walls on both sides (Figure 1).
30.0 m
15.0 m

45.0 m
0.0 m

Wall
Reference Plan view
N
5.0 m

CF UN FP Wheel Path
Geogrid No Geogrid Geogrid
Wall

Figure 1 Layout of the instrumented test section.

The existing asphalt surfacing was milled and reconstructed, while the
unbound granular base course and the subgrade were maintained (§4.1). The
cross section of the test pavement is depicted in Figure 2. The new surfacing is
composed by:
- a lower asphalt concrete layer (thickness 40 mm).;
- the interface (either reinforced or unreinforced);
- an upper asphalt concrete layer (thickness 50 mm)
The same asphalt concrete mixture was used for the two layers (§2.3).
To simulate challenging situations frequently encountered in overlaying
projects, areas with “artificial” cracks and critical bonding conditions were
prepared in the lower asphalt layer (Figure 3). The artificial cracks where
produced through full-depth saw cuts in the lower asphalt layer, while a thin
layer of fine sand was used to create de-bonding.

3
 50 mm Upper AC Layer
Interface
40 mm Lower AC Layer

200 mm Unbounded Base

Subgrade

Figure 2 Cross section of the test pavement.

7.0 m
2.0 m

8.6 m
1.0 m
0.0 m

12.8 m

15.0 m
14.0 m
1.2 m

Reference
Wheel Path

de-bonding instrumented “artificial”

5.0 m

area cracks
3.8 m

0.4 m
0.4 m

Figure 3 Debonding and artificial cracks on the lower asphalt layer surface.

2.3 Material properties

Preliminary tests were carried out in order to obtain the main physical and
mechanical properties of the existing subgrade and base course. The subgrade is
an inorganic clayey soil of medium plasticity (Liquid Limit: 42; Plasticity
Index: 18) than can be classified as CL-ML according to the USCS. In-situ
CBR values were estimated using Dynamic Cone Penetrometer (DCP) tests, and
found to vary from 5 (underneath the subbase) to 10 (at 900 mm depth).
The base course is composed by 100% crushed, limestone aggregates, with a
nominal maximum dimension of 25 mm. In-situ CBR values exceed 100%.
The asphalt concrete used for the new lower and the upper asphalt layers was
a typical Italian dense graded mix formulation, with 12 mm maximum
dimension (AC 12) and 70/100 penetration bitumen. A summary of the quality
control tests results is reported in Figure 4.
An SBS polymer modified emulsion tack coat was applied on the surface of
the lower layer, both in the reinforced and the unreinforced subsections.
Two geogrid types were installed (Figure 5). The glass Fiber Reinforced
Polymer (FRP) geogrid is obtained by weaving continuous alkaline-resistant
pre-tensioned glass fibers, covered with a thermosetting epoxy resin
(vinylester). The grid has flat transversal strands woven into longitudinal
twisted strands, with a square 33 mm mesh. The Carbon Fiber geogrid is
characterized by the same type of strands in both directions, with a square 20
mm mesh and is pre-coated with bitumen.

4
 100
Average gradation (100% crushed
limestone aggregates)
Specification limits for AC 12
75 Bitumen 5.4% by aggregate weight
Passing [%]

50

25

0
0.01 0.1 1 10 100
Sieve size [mm]

Figure 4 Quality control test results summary.

Figure 5 Glass fiber polymer geogrid (left); Carbon fiber geogrid (right);

3. INSTRUMENTATION

3.1 Instrumentation layout

The test sub-sections were instrumented with pressure cells, asphalt strain
gauges and temperature sensors. The position of the instrumented area inside
each sub-section is highlighted in Figure 3, while Figure 6 shows the plain view
and cross section of each instrumented area.
One Earth Pressure Cell (EPC) was installed in each section to measure the
vertical stress at the top of the unbound granular base [12]. The EPC lies on a
horizontal plane, 50 mm below the asphalt interface, along the reference wheel
path. Four “H-shaped” Asphalt Strain Gauges (ASG) were installed in each
section, inside the lower asphalt layer, underneath the double-layer interface,
(60 mm below the pavement surface). Two ASG measure the horizontal strain
in the longitudinal (traffic) direction, and two the horizontal strain in the
transverse direction [13]. Six additional strain gauges (traditional bonded foil
and bonded wire type) were installed close to the ASG, three in the longitudinal
direction, and three in the transverse direction.

5
 Plain view
1.0 m 0.6 m 0.6 m

Earth Pressure Cell

Asphalt Strain Gauge Reference

0.6 m
Strain Gauge Wheel Path

Cross section
Upper AC layer
60 mm
Interface 140 mm
Lower AC layer
Unbounded base
Figure 6 Instrumented area plain view and cross section.

3.2 Sensors characteristics

Earth Pressure Cells, Geokon® type 3500, with a measuring range of 1.0 MPa
were installed. The EPCs have a diameter of 230 mm and are equipped with a
semiconductor strain gauge transducer. The sensors require an unregulated
input of 7–35 VDC and produce a voltage output in the 0–5 V range. Moreover,
each EPC is equipped with a resistance thermistor for temperature readings. The
thermistor has a measuring range from -80°C to 150°C, with an accuracy of
0.5°C. At normal operating temperatures, 0-35°C, the thermistor output is in
the 2–10 kOhm range.
Asphalt strain gauges, CTL ASG-152, with a measuring base of 200 mm and
a range of ±1500 με were installed. The ASG have a full-bridge configuration
that requires an excitation up to 10 VDC and produces a full scale output of 2
mV/V. In each subsection two ASG were also equipped with a resistance
thermistor for temperature readings. The thermistors have a measuring range
from -80°C to 150°C, with an accuracy of 0.5°C. At normal operating
temperatures, 0-35°C, the thermistors output is in the 16–33 kOhm range.
Traditional strain gauges, HBM LY41-100 (bonded foil, gauge length: 100
mm) and TML PL-60 (bonded wire, gauge length: 60 mm), where also used.
Temperature compensation strain gauges, applied on a dummy asphalt concrete
sample, were used to set-up half-bridge measuring configurations. A two-
component polyester adhesive was used for bonding (TML RP-2) and two types
of coating compounds were applied, for mechanical and thermal protection
(HBM ABM 75, TML AV138).

3.3 Signal conditioning and data acquisition system

For all instruments, signal conditioning, amplification and A/D conversion was
carried out using two portable HBM Spider8 units, connected in series to a

6
laptop computer. The HBM Catman Express software was used for data
acquisition. Each Spider8 unit is equipped with four digital amplifiers modules
(SR55) in 4.8 kHz carrier-frequency technology, suitable for passive
transducers, like the ASG, and the DC voltage output of the EPC. The
capability of one unit was extended with three DC modules (SR01) suitable for
resistance measurements up to 4.0 kOhm.
The data acquisition system was designed to carry out independent, full-
scale, field tests in each subsection. Therefore, the Spider8 units are connected
to the sensors installed in each subsection in a flexible and efficient manner. For
this task, a custom connection board was realized and installed at each
subsection, along with custom connection cables. The main system structure in
described in Figure 7:
- the EPC vertical pressure readings were carried out with a SR55
amplifier module, while its excitation was provided by a separate 12
VDC power supply circuit (Spider8 units provide only 1 V excitation
voltage);
- the four ASG were connected to the SR55 modules (4-wires full-bridge
circuits);
- the three thermistors (one from EPC, two from ASG) were connected to
the three SR01 modules; a 4.8 kOhm shunt resistance circuit was
necessary because the resistance of the thermistors at normal operating
temperatures (1–50 kOhm) falls outside the measuring range of the SR01
amplifier module (0-4 kOhm).

Figure 7 Data acquisition system layout.

7
4. TEST SECTION CONSTRUCTION

4.1 Construction process outline

The instrumented test section was constructed along a secondary road where the
existing pavement was composed by a 200 mm unbound base course and a 90
mm asphalt surfacing. Construction began milling the asphalt layer. Static and
dynamic plate load tests were than carried out on the base, to measure its
bearing capacity, previously estimated using DCP tests (§ 2.3).
After EPC installation (§4.2), the lower asphalt layer of the new pavement
was constructed. On its surface de-bonding and cracked areas were prepared
(Figure 3), and strain gauges were installed and wired (Figure 6). Afterwards a
tack-coat was applied, and the geogrid reinforcement was installed in test
sections CF and FP. Finally the upper asphalt layer was constructed. The entire
construction process took six working days.

4.2 Instruments installation

In each subsection, a shallow cavity and cable trench were dug on the surface of
the aggregate base to install the earth pressure cells [12, 13]. The cavity,
approximately 100 mm deep, was leveled using a 50 mm layer of clean sand
(passing to the 2 mm sieve an retained to the 0.4 mm sieve) to prevent EPC
damage from large, sharp base aggregate particles.
The cell and its stem were carefully placed and leveled, avoiding the
formation of voids, and a second layer of protection sand was used to fill the
cavity. The cabling was also protected using a flexible steel conduit and laid
inside the trench that was backfilled with the existing base material. Before
paving operations, a small amount of asphalt concrete was placed over the cell
and compacted using a portable Marshall hammer, to provide additional
protection.
To install the ASG and the traditional strain gauges, shallow cavities were
created on the surface of lower asphalt layer. Wood plates were placed at the
installation positions and pressed by hand on the surface of the hot asphalt
concrete lift, behind the paving machine. A similar procedure was used to
obtain a small trench used to bring the cabling from the instrumented area to the
road side. For this operation a steel bar was employed. After compaction, the
plates and the bar were easily removed from the warm layer. A thin layer of
sand-asphalt mortar was used to tack the ASG in the correct position and
alignment, inside the prepared cavities. The sensors were protected filling and
leveling the cavities with a small amount of asphalt concrete, obtained
removing larger aggregate particles from the paving mix, with a 8 mm sieve.
Compaction was achieved only by static pressure. The cabling was protected by
a flexible steel conduit laid inside the trenches and backfilled with the tacking
mortar.

8
 Figure 8 shows the connection board and its roadside housing. At the end of
construction, a sensor survey revealed the following survival rate:
- pressure cells: 3 out of 3;
- ASG: 11 out of 12;
- traditional strain gauges: 16 out of 24.

Figure 8 Cables connection board and roadside housing.

5. OUTLINE OF TESTING ACTIVITIES

Two types of full-scale tests have been scheduled on the instrumented test
section: Falling Weight Deflectometer and full scale truck tests.
FWD tests will be performed at selected points inside each subsection,
including the EPC and ASG positions. Figure 9 shows: (a) the superimposition
of the 3 EPC outputs obtained for 3 consecutive FWD drops in the same spot
and (b) the ASG response to a single FWD drop.
Full scale tests will be performed driving a loaded truck over the
instrumented area of each subsection. Figure 10 shows the time histories
recorded from the EPC and the longitudinal ASG during a pass of a 2 axle truck
on the unreinforced section. Stress and strains responses to each axle load is
visibly captured.

9
 0.3 100

ASG L2
Pressure [MPa]

50
ASG T1

Strain [me]
0.2

0.1
ASG L1 -50
ASG T2
0 -100
0.12 0.17 0.22 0.1 0.15 0.2 0.25
Time [s] Time [s]

Figure 9 Sample stress and strain measurements from real scale test.

0.4 200

EPC
ASG L1
0.3 100
ASG L2
Pressure [MPa]

Strain [me]
0.2 0

0.1 -100

0 -200

FRONT AXLE REAR AXLE

-0.1 -300
2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4
Time [s]

Figure 10 Sample stress and strain measurements from real scale test.

CONCLUSIONS
A full-scale instrumented pavement test section was constructed along an in-
service road to study the effectiveness of geogrid reinforcement in asphalt
layers. A carbon fiber (CF) and a glass fiber reinforced polymer (FRP) geogrid
were installed inside an asphalt double-layer surfacing. In each subsection, one
pressure cell was installed near the top of the existing aggregate base, and four
asphalt strain gauges beneath the double-layer interface. Areas with artificial
cracks and de-bonding were also prepared. The data acquisition system was
realized to carry out independent, full-scale, field tests in each subsection. A
portable data acquisition unit can be easily connected to the installed sensors
using a custom connection board that was installed in each subsection. The
entire construction process took six working days and more than 90% of the
major sensors (EPC and ASG) survived this critical phase. The first full-scale
tests performed on the section with a FWD and driving a truck over the

10
instruments position confirm that vertical stresses inside the base course and
horizontal strains beneath the reinforced interface can be measured with
confidence.

ACKNOWLEDGMENT: The research described in this paper was funded by

the Italian Ministry of Instruction, University and Research (MIUR).

REFERENCES:

[1] S. F. Brown, J. M. Brunton, D. A. B. Hughes and B. V. Brodrick

Polymer grid reinforcement of asphalt, Asphalt Paving Technology 1985 –
San Antonio, Texas, Febraury 1985.
[2] S. F. Brown, N. H. Thom and P. J. Sanders, A study of grid reinforced
asphalt to combat reflection cracking, Association of Asphalt Paving
Technologists, Annual meeting 2001.
[3] M.M. Elseifi and I.L. Al-Qadi, A simplified overlay design model against
reflective cracking utilizing service life prediction, 82th Transport
Research Board annual meeting, Washington, D.C., 2003.
[4] M. Bocci, A. Grilli, F.A. Santagata and A. Virgili, Influence of
reinforcement geosynthetics on flexion behaviour of double-layer
bituminous systems, Proc. Int. Conf.: Advanced Characterization of
Pavement and Soil Engineering Materials, Athens, Greece, June 2007.
[5] G. Ferrotti, F. Canestrari, A. Virgili and A. Grilli, A strategic laboratory
approach for the performance investigation of geogrids in flexible
pavements, Construction and Building Materials, in Press,
doi:10.1016/j.conbuildmat.2010.11.032, 2010.
[6] A. Virgili, F. Canestrari, A. Grilli and F.A. Santagata, Repeated load test
on bituminous systems reinforced by geosynthetics. Geotextiles and
Geomembranes 27, 187–195, 2009.
[7] H. Kim, K. Sokolov, L. D. Poulikakos and M. N. Partl, Fatigue Evaluation
of Carbon FRP-Reinforced Porous Asphalt Composite System Using a
Model Mobile Load Simulator, Transportation Research Record: Journal
of the Transportation Research Board, Vol. 2116. 2009.
[8] S.W. Perkins and E. Cortez, Evaluation of base-reinforced pavements
using a heavy vehicle simulator, Geosynthetics International, 12(2), 87–
98, 2005.
[9] S.W. Perkins, M. Ismeik, and M.L. Fogelsong, Mechanical Response of a
Geosynthetic-Reinforced Pavement System to Cyclic Loading, Proc. of the
Fifth Int. Conf. on the Bearing Capacity of Roads and Airfields,
Trondheim, Norway, Vol. 3, pp. 1503–1512, 1998.
[10] B.R. Cox, B. Curry, C.M. Wood, C. Young and J.S. McCartney, In-situ
strain measurement during dynamic shear loading of an unbound geogrid
reinforced pavement section, Proc. of the Fifth Int. Conf. on the Bearing
Capacity of Roads and Airfields, Vol. 2, pp. 1134–1157, 2009.

11
[11] K. S. Henry, E. R. Cortez, L. S. Danyluk, G. Brentrup, N. Lamie and T. W.
Arnold, Construction and Instrumentation of Full-Scale Geogrid
Reinforced Pavement Test Sections, ERDC/CRREL TR-08-6, 2008.
[12] D.H. Timm, A.L. Priest, T.V. McEwen, Design and Instrumentation of the
Structural Pavement Experiment at the NCAT Test Track, NCAT Report
04-01, 2004.
[13] D. H. Timm, Design, Construction and Instrumentation of the 2006 Test
Track Structural Study, NCAT Report 09-01, 2009.

12