See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/327531866

REVIEW OF FALLING WEIGHT DEFLECTOMETER FOR ASSESSMENT OF

FLEXIBLE PAVEMENT

Article · September 2018

CITATIONS READS

0 2,398

2 authors:

Prakash Makwana Praveen Kumar

Indian Institute of Technology Roorkee Indian Institute of Technology Roorkee
7 PUBLICATIONS   6 CITATIONS    110 PUBLICATIONS   1,209 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Effect of Ageing on the Viscoelastic Properties of Asphalt Binder View project

RAP in concrete View project

All content following this page was uploaded by Prakash Makwana on 08 September 2018.

The user has requested enhancement of the downloaded file.

 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

www.elkjournals.com

………………………………………………………………………………………………………………………

REVIEW OF FALLING WEIGHT DEFLECTOMETER FOR ASSESSMENT

OF FLEXIBLE PAVEMENT

Prakashkumar Makwana Dr. Praveen Kumar

Road and Building Department, Government of B.E (Civil), M.E. (Transportation Engineering)
Gujarat (Assistant Engineering) Ph.D. FIE, LMIRC, MIGS.
Research Scholar at Indian Institute of Professor at Transportation Engineering Group
Technology, Roorkee. Uttarakhand – 247667 Department of Civil Engineering. Indian
prakashkumar.makwana@gmail.com Institute of Technology, Roorkee.
Uttarakhand – 247667

ABSTRACT

Fast development of road networks has become a trend in India and everywhere in the world. From the past couple
of decades, it has been observed that numerous highways are in a phase of deteriorations. Identifying the reasons
for deteriorations requires a pavement evaluation study. Many performances study have been made out by exploring
flexible pavements, by the users of widely accepted falling weight deflectometer (FWD) as a non-destructive test
(NDT) and considered it as a standard for structure assessment. The primary objective of this study is to a review
of an FWD instrument and the also study of the empirically derived methods and a back calculation process for
computing layer moduli and factors influencing it. The essential need of correction factors to get reliable layer
moduli is an also discussed, in addition to the investigation of advancement of low-cost indigenous FWD models

Keywords: Falling weight deflectometer (FWD), back calculation process, correction factors, surface
deflection

1. INTRODUCTION
Rapid construction of road infrastructure has Widely accepted NDT. In NDT, in situ test is
become a trend in India and all over the world. conducted on in service pavement without
In past few decades, it has been observed that disturbing or breaking out pavement layer. NDT
many road works require early stage of tools for evaluating material layer properties of
maintenance. To identify causes of it, require a in service pavement are extensively used
structural evaluation study to assess the existing worldwide. Generally, a wave prorogation
layers properties of pavement. Many technique and deflection based approach have
performances study have been made out by gained popularity in the field of pavement
exploring flexible pavements by the users of engineering. In wave prorogation techniques,
1
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

vibration source kept on a surface of the closely simulate loading condition of actual
pavement and velocities and wavelength of moving load [1]. The FWD has been being used
surface waves are measured, which are emitted now for over numerous years for pavement
from vibration source and transmitted through assessment, including utilization on unbound
pavement layers. This approach requires highly asphalt layers. It is a trusted apparatus and
advanced computer programmer for reliable regarded by numerous researcher as a standard
results interpretations, therefore, it is not widely against another mention NDT [2]. The primary
used. From the early 1970s, the surface objective of this study is to a review of an FWD
deflection approach is extensively used for instrument and the also study of the empirically
assessing pavement material because of its derived methods and a back calculation process
reliability, speedy operation and ease of use. for computing layer moduli and factors
Surface deflection is overall responses (in terms influencing it. The essential need of correction
of deflections) of the full depth of pavements factors to get reliable layer moduli is an also
under predefined standard application of load. A discussed, in addition to the investigation of
surface deflection is measured by non- advancement of low-cost indigenous FWD
destructive deflection tests. Back calculation models.
analysis is performed to determine the structural FWD test, in which mass is allowed to
properties of distinct layers or to estimate the fall from a predefined height on pavement
moduli value of distinct layers and computed surface and surface deflections or deflections
moduli values are furthermore used for analysis basin are measured using a velocity transducer
of pavement and estimating the remaining life (geophone) or deflection sensors, which are
and overlay requirement analysis of pavements. equipped with FWD. It is observed that the
Structural evaluation studies are conducted amplitude deflection at distinct radial point
with various tools such as Benkelman beam occurs at distinct time moments, which are not
deflection (BBD), lightweight deflectometer closely simulating the actual transient deflection
(LWD) and FWD. The best capable devices for conditions of moving wheel load. Therefore, a
measuring accurate pavement response are built measured deflection is further evaluated through
on the using dynamic loading and the assessment back-calculation analysis. Moreover, a detailed
of the deflections. Amongst the various of operating principle, deflection basin is
deflectometer assembled devices, the FWD is discussed in subsequent sections.
the extensively utilized and considered as a
benchmark test for pavement evaluation due to

2
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

2. COMMERCIAL AVAILABLE FWDs Numerous field investigations were made using

Different types of commercially available this equipment, and it showed good repeatability
FWDs are briefly discussed in this section. An of deflections [5]. This low-cost equipment is
international overview of FWDs are presented in quite suitable for developing
Table 1, which are not discussed here and only Countries like India. Some of the drawbacks of
indigenous FWDs are discussed in this study. this model are a require many of the laborious

(Ref Table- 1) operations such as pulling off a chain for lifting

the mass, placing the geophones on the

2.1 IITKGP FWD Model -I pavement surface and releasing the mass. Tests
The first Indian FWD model was developed [5] are performed physically and thus it has taken
by the transportation engineering section of the more time.
department of civil engineering, Indian institute Furthermore, maneuvering the equipment on in-
of technology, Kharagpur, India. It is trailer service highways in India was found to be hard
mounted, towed with the help of a jeep. This and clumsy.
model has loading capabilities ranging from 20 To defeat all mentioned drawbacks of IITKGP
kN to 65 kN and loading time between 20-30 Model- I a second model was produced in the
milliseconds, rubber pad used as buffer (spring) year 2001 by IIT, Kharagpur, India and works
system for the obtained desired load duration, were sponsored by MORT&H.
which is closely similar to a moving vehicle
speed of 50-60 kmph. Surface deflections can be 2.2 IITKGP FWD Model –II
measured at offset distances of 300 mm apart up IITKGP FWD Model –II is a fully automatic
to 1500 mm distance with the assistance of six vehicle-mounted instrument. All the processes
geophones. A string and pulley prearrangement are computerized and surface deflections data
is employed for raising and letting down the are gathered through a data-acquisition system,
weight, whereas a clamp arrangement is built up also one additional geophone is added for
for supporting the stack at any desired height. obtaining better surface deflection data. An
Single load cell and six geophones are used to impulse loading range from 20 kN to 100 kN can
quantify the magnitudes of load and deflections be obtained by varying dropping mass and
respectively. The load and deflection are read on heights ranges from of 100 kg to 225 kg and 100
the computer with the aid of a data acquisition mm to 600 mm respectively on 300 mm loading
system. plate diameter. Which allows uniforming
distribution of stresses on the pavement and by

3
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

the help of seven geophones surface deflections corresponding peak load and peak vertical
are measured with observed load duration varies surface deflections at different radial locations
from 20 to 30 milliseconds. are measured using deflection sensors as shown
in Figure 1, DO, D1, etc., are surface deflections
2.3 Geotran FWD measured at different radial distances and
GEOTRAN FWD is a fully automatic vehicle- recorded in data acquisition system. (Ref
mounted instrument for measuring surface Figure- 1)
deflection and requires only one man to operate
all its operations. All the operations are 3.1 Deflection basin
controlled from PC/laptop through the DS4000S The reliability and usefulness of FWDs are
data acquisition system. DS4000S system is a based on the capability of simulate closely to the
very accurate and high-speed controlling actual loading condition. It includes traffic
system, that is capable of captures all required loading and stresses induced due to environment
data of geophone, load cell, and temperature. and weather condition. When a moving wheel
GEOTRAN FWD has produced the impulse load passes over the pavement, it generates load
load up to 100 kN on existing pavement by pulses. Normal stresses (vertical as well as
dropping weight from predefined height and horizontal) at a specific location in the pavement
evaluate surface deflections using seven inbuilt and it will increase in magnitude from zero to a
geophones. It has also two temperature sensors peak value as the moving wheel load approaches
for air temperature and road surface temperature the specific location. The time taken for the
measurement. Loading plate has a diameter of stress pulse to vary from zero to peak value is
300 mm with reinforced rubber plate. termed as 'rise time of the pulse'. As the wheel
moves away from the location, the magnitude of
3. OPERATING PRINCIPLE OF FWDs stress reduces from the peak value to zero. The
The working principle of all FWDs model are time period during which the magnitude of stress
same. A mass is allowed to drop from a pulse varies from 'zero-to-peak-to-zero' is the
predetermined height onto a series of springs/ pulse duration. Peak load and the corresponding
buffers placed on top of a loading plate to pavement responses are of interest for pavement
produce impulse load pulse on the pavement evaluations are shown in (Ref Figure 2).
surface, buffer system of suitable stiffness (5 The size and shape of the deflection basin permit
mm minimum thickness) used to simulate the comprehensive structural investigation of the
actual load duration of moving traffic. The pavement. Fundamentally, the exterior

4
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

deflections describe the modulus characteristics on the radius (ae) of the stress bulb at the
of the sub-grade, although the bowl nearby to the subgrade-pavement interface so that r equal or
loading plate permits investigation of the greater than ae, suggestive value, r is equal to or
modulus characteristics of the nearby surface greater than 0.7ae.
layers. A wide basin with little curvature ES (psi) = 0.24 P/ (dr * r) (1)
describes that the upper strata of the pavement Garg and Thompson (1998) proposed regression
are stiffer to the sub-grade. A basin with the equations (2-3) for estimating the subgrade
equal peak deflection, but high curvature nearby modulus from FWD test using pavement
the loading plate describes that the upper layers deflection, in which, D3 in miles (0.001 inches)
are weaker to the sub-grade. measured at 1097 mm radial distance from the
4. DETERMINATION OF LAYER center of the loading plate [7].
MODULUS For AC pavements:
Reliable estimation of individual layer modulus Log ES = 1.51-0.19 D3 +0.27 log (D3) (2)
from measured deflections of the FWD test is a For full depth AC pavements:
complex procedure. By taking into account of Log ES = 24.7-5.41 D3 +0.31 (D3)2 (3)
the size and shape of radial offset deflections, Choubane and McNamara (2000) proposed the
various researchers attempt to find layer equation 4 for predicting embankment subgrade
modulus and developed empirical relations. The modulus from FWD measured deflection at a
pavement theories based back-calculation radial distance of 1097 mm [8].
procedure is also reviewed in this section. ES = 0.03764 (P/dr) 0.898 (4)
Alexander et al, (1989) proposed an equation 5

4.1 Empirical models for subgrade modulus from the deflection (mils)
measured at a radial distance of 1830 mm (D72)
Attempts were made in the past by researchers
from the center of the loading plate for an
to estimate layer modulus from the measured
applied load of 111206 N [9].
surface deflections using NDT techniques are
presented, in here and these are briefly reviewed Es (psi) = 59304.82 (D72)-098737 (5)
Roque et al, (1998) produced the equation 6 for
in this section.
the appraisal of subgrade modulus based on the
AASHTO (1993) [6] recommends the equation
deflections measured at 60 inches radial distance
1 for back calculated subgrade resilient modulus
from the middle of the dual plates using a dual
using a deflection measurement from the center
load [10].
of the load, further recommend that the
minimum sensor distance (r), be estimated based ES (ksi) = 36.334(DX /60)-1.015 (6)

5
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

Molenaar and Van Gurp (1982) developed the distance, applied load doesn’t induce any
equation 7 to predict subgrade soil modulus from deflection. Therefore, AASHTO (1993) defines
the FWD deflections (in meters) measured at a minimum radius distance based on the radius of
radial distance of 2000 mm [11]. the stress bulb induced due to the applied load
ES (MPa) = 6.614 *10-3*d2-1.00915 (7) and suggest minimum radius is equal to or
Subgrade modulus can also be determined by greater than 0.7 times of the radius of bulb stress
Harr (1966) from the average deflection value [6]. Garg and Thompson (1998) and Choubane
measured during the third, fourth and fifth drops and McNamara (2000) used radius distance of
of the load in a portable falling weight 1097 mm from center loading plate [7-8]. In
deflectometer (PFWD) using Equation 8. [12] addition, Alexander et al, (1989); Roque et al,
Es (MPa)= 2 P A (1-μ2) r a/ d (8) (1998); Molenaar and Van Gurp (1982) and
Wimsatt (1999) developed a regression Wimsatt (1999) used radius distance of 1830
Equation 9 using FWD deflection (mm) mm, 1524 mm, 2000 mm, 1828.8 mm from the
measured at a distance of 1828.8 mm [13]. center of the loading plate respectively [9-
Es (MPa) = 0.24 P/(W7*1828.8) (9) 11,13]. Moreover, Equations 1,4,8 and 9 are
developed by using deflection measured at a
Discussion radial distance from the center of loading plate,
An empirically developed relation for applied load, and radial distance, which are
determining individual layer modulus is a based based on Boussinesq solution, particularly
on the size and shape of a deflection basin. applied to the axis of symmetry. While,
Fundamentally the outer deflections describe the empirical equations 2,3,5,6 and 7 are based on,
modulus of the subgrade while deflection closer only a function of deflection measured at a radial
to the loading plate permits analysis of the near distance from the center of loading plate. These
surface layers, it is based on the typical pattern equations are employed only outer sensor
of load distribution or stress zone observed deflection values; Equations 2,3,5,6 and 7 are
under applied load in the flexible pavement. not widely used because of these are based on
Development of subgrade modulus from the only deflection, while equations 1,4,8 and 9
exterior peak deflection is not a straight forward consider important, which attribute to strength
process. It is crucial to characterize a radial characteristics, such as deflection, applied load,
distance from center of the loading plate to the and radial distance.
exterior deflection center. Although, it is well- Again, the load distribution approach is utilized
known that a from the specified interval of to determine the modulus of granular and the

6
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

surface layer. The equations 10-13 are a function Back-calculation is a reverse analysis for finding
of surface course thickness and the combination a layer moduli from pavement response (in terms
of measured deflection at a radial distance at 0, of surface deflection) underneath the application
200, 500, 800, 1600 mm etc... from the center of of a given load. The back-calculation is a
the loading plate. Badu et al. (1989) developed numerical technique concerning the following
equations 10-11. [14] modeling mechanisms as shown in Figure 3: (a)
loading model, (b) pavement and material
For Granular layer: models, (c) a pavement response model, and (d)
Log EBASE (ksi)= 3.280-0.03326(t1)-0.1179log back-analysis model [4].
(D7) + 3.3562log (D1 –D2) -9.0167 log(D1-D4)- Loading model is a defined on the basis of a
4.8423 log (D1 –D5) (10) mode of applied load, which consist a static load,
For Bituminous layer: a moving load, a vibratory load, and an impulse
Log EAC (ksi)= 2.215-0.2481 (t1) -12.445 log load. The dynamic loading model yields more
(D1 –D2) + 17.205 log (D1 – D3) -5.87 log (D1 – precise results, but it is a creates inertia and
D4) (11) resonance as extra effects. The modulus of the
Roque et al. (1998) developed equation 12-13. subgrade could be devalued by half or more, and
[10] the base and subbase moduli were exaggerated
For Granular layer: by about the same fringe when dynamic impacts
EBASE (ksi) = 105.81136(t2)-1.0785 * (Dx /36 - Dx are precluded from the analysis [15]. Pavement
/ 60) 6.02523+2.4888/ Dx / 60 * (Dy/ 0+ Dx / 12)- model is consisting of a modeling of the
1.15(Dx / 36) 2.1609−1.6202/ (Dx / 36)-5.302/t2 * (Dx / 60) pavement compositions, layer thickness, and
3.6706−0.0498t - 0.686t (12) Poisson ratio. Material model is a defined as a
1 2−3.09/Dx /60)

For Bituminous layer: properly modeling of the nature of materials

EAC (ksi) = 78.2254 (t1) 0.5554 (Dy/0 -D under the application of load. Granular materials
y/305)

(0.7966-19.1332/t ) * (Dy/0 - Dx/200)17.4791/t1 (13) and subgrade materials are stress-dependent and
1

Several combinations of measured deflections nonlinear in nature. Subgrade modulus

have sometimes contributed to inappropriate decreases with the increasing in stress levels,
modulus results; therefore, these are not widely therefore it is a demonstrating stress-softening
practiced. type characteristic. The vital factor affecting the
subgrade modulus is the vertical deviator stress.
4.2 Back calculation A modulus of granular materials increases with
increasing in stress states (stress-hardening),

7
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

especially with confining pressure and/or bulk measures of output error utilized by analysts are:
stress, and slightly with deviator stress. The (a) the sum of the absolute differences (SAD),
response models have analyzed the pavement (b) the sum of the squared differences (SSD),
responses based on the kinds of material model and (c) the sum of the squared relative errors
and loading model employed for analysis. (SSRE). Various methods have been utilized to
Mostly, it is classed as four varieties of models land at an answer that gives a worthy match
(a) linear static analysis, (b) nonlinear static between the evaluated and measured deflection
analysis, (c) linear dynamic analysis, and (d) bowl. The most widely recognized assault is one
nonlinear dynamic analysis. that uses an iterative gradient search algorithm,
In the linear static analysis, linear material for example, the gauss–newton method.
model and static loading model are utilized, Contrasted and the alleged database techniques
layer thicknesses and Poisson's ratios are known [17-18] and the regression equation based
and only one unknown (i.e., elastic modulus) for methodology [19-20], this methodology, for the
each layer. Most widely used linear static most part, takes longer time because of the need
layered elastic programs is KENLAYER [16]. In to perform the forward structural response
the nonlinear static analysis, it is also employed model over and again.
static loading model, but the main change lies in
the material models, which is a utilizing Discussion
nonlinear material model. This gives more than Back-calculation procedure is depending on
one unknown model parameters for each layer coordinating of computed and measured
and also the trustworthiness of back calculated pavement deflections and it comprises of the
values of these parameters is a significant matter accompanying three noteworthy strides: (a)
to be considered. Linear and nonlinear dynamic determination of a trial set of qualities for the
investigations required the time history obscure pavement parameters, (b) forward
information of load and the deflection bowl calculation of pavement response taking into
defined by the amplitude values. The time account the parameter values chose, and
history of deflections might be utilized rather correlation of the computed response with the
than the peak deflection bowl for better results. measured, and (c) changing the chose parameter
The back-analysis portion is grounded values by method for a suitable search algorithm
on the minimization of the “output error,” i.e., to accomplish enhanced coordinating of the
the uniqueness between the deliberate and computed and measured responses. The
figured surface deflections. The three normal accuracy of layer modulus depends on the choice

8
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

of loading, pavement and material, pavement ET1/ ET2 = (2.6277- 1.38 log10 T1) / (2.6277 –
response and back analysis models employed for 1.38 log 10 T2) (14)
analysis. A most widely used example is a linear
material model and static loading model based Rada et al (1988) gave the expression for
layered elastic programs is KENLAYER. It modeling the variation of stiffness with
needs an only one unknown parameter (i.e. layer temperature [22].
modulus) requires to find, but it doesn’t consider
a non-linearly characteristic of materials and that ET1 / ET2 = 10 3.245 x 10−4 (T11.798- T21.798) (15)
lead to inappropriate results. To get accurate
results it is recommended that to use a non-linear Antunes (1993) proposed the equations 16-17,
dynamic pavement response model. For that a based on the analysis of back calculated moduli
requires a profoundly computational effective obtained from the FWD data collected at
PC program, for example, Finite element different temperatures [23].
method (FEM).
5. CORRECTION FACTORS For Asphalt Concrete:
Properties of bituminous mix changes with ET1 / ET2 = (1.635 - 0.0317 T1) / (1.635 -
temperature, modulus values got at distinctive 0.0317 T2) (16)
temperatures are typically set to fit a standard
temperature for the design of pavements and For Bituminous Macadam:
overlays. Attributes of a granular layer are ET1 / ET2 = (1.795 - 0.0398 T1) / (1.795 -
highly altered by moisture content, thus seasonal 0.0398 T2) (17)
moisture correction and also particular
temperature alteration factors were created by Kim et al (2000) presented the equations 18-19
different experts for confirming the modulus as for adjusting the deflection value and moduli
well as deflection are studied in this section. value for temperatures of 680F, where, t is
Ullidtz and Peattie (1982) employed the thickness of the Asphalt Concrete (AC) layer
deflection data from AASHO road test and the (inch) and T is AC layer mid-depth temperature
SHELL procedure for finding of mix stiffness (0F) at the time of FWD testing, α is 3.67 x 10-4
and developed the equation 14 for comparing the x t1.4635 for wheel paths and 3.65 x 10-4 x t 1.4241
moduli obtained at two different temperatures for lane centers [24].
[21].

9
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

For Deflection: statistical analysis of back calculated moduli and

measured AC temperatures [28].
D68 = DT * [10α (68 –T)] (18)
ETo= 100.018 (T-20) * ET (23)
For Modulus:
Ali and Slezneva (2000) acquired a relationship
E68 = ET * [100.0153 (68 –T)] (19) for estimating AC layer modulus as a function of
average AC layer temperature (0C) and
Chen et al (2001) suggested the equation 20 for temperature gradient in the AC layer (0C/m)
adjusting the layer modulus for a given [29].
temperature [25].
EAC= -934 + e (9.53−0.033*(Tp)+0.0018*(TG) (24)
ETw = ETC/ [(1.8Tw +32)2.4462 * (1.8Tc
+32)-2.4462] (20) IRC:115-(2014) developed equation 25
temperature correction factor corresponding to a
Johnson and Baus (1992) recommended the 350 C temperature; this component is valid for
equation 21 for adjusting the bituminous layer temperature ranges 250 to 400 C [30]
modulus for a standard temperature of 700F [26].
E (T10 c) = α E (T20 c) (25)
ETw = ETC/ [(1.8Tw +32)2.4462 * (1.8Tc +32)-
2.4462] (21) Where, α = [1-0.238 ln T1/ 1-0.238 ln T2]

Ullidtz (1987) built up a theoretical account for Granular layer and subgrade materials are
temperature correction based on back calculated susceptible to moisture variation, therefore
moduli values obtained from AASHO Road Test IRC:115-(2014) recommended equation 26-29
deflection data [27]. for moisture correction by considered summer
and winter seasons variation for granular layer
ETo= (1/3.177-1.673 log10 T) ET (22) and subgrade [30].

Baltzer and Jansen (1994) built up the For Summer:

temperature correction model 23 based on Egran_Mon = - 0.0003 * (Egran_Sum)2 +
0.9584* (Egran_Sum) -32.989 (26)

10
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

Indian conditions, 35 C0 temperature and

For Winter: monsoon measured modulus are considered as
Egran_Mon = 10.5523* (Egran_Win)0.624 - a standard for pavement design.
113.857 (27)
6. CONCLUSION
For Summer:
Esub_mon = 3.351 * (Esub_win 0·7688) - 28.9 FWD is extensively used for assessing pavement
(28) material because of its reliability, speedy
For Winter: operation and ease of use, and also consider as a
Esub_mon = 0.8554* (Esub_sum) - 8.461 benchmark test for pavement evaluation due to
(29) closely simulate loading condition of actual
moving load. The degree of utilization of FWD
5.1 Discussion in developing nations like India is constrained
Bitumen material is susceptible to the due to the high cost of international
temperature variation and does change in commercially available FWD. Keeping up such
characteristics of it, also the temperature profoundly immoderate equipment is turned out
variation effect on the FWD measured to be troublesome due to the absence of skill.
deflections. Granular layer and subgrade Henceforth the improvement of an ease FWD
materials are susceptible to moisture variation. will be useful in the legitimization of the
Therefore, requires a correction factor for pavement assessment approach in India.
standard temperature and worst moisture content Indigenous low-cost GEOTRAN FWD is a fully
for pavement design. It is a basic methodology automatic vehicle-mounted instrument for
to apply a correction factor to back calculated measuring surface deflection and requires only
modulus values. Since at last focus of FWD test one man to operate all its operations. Vital
is to the determination of layer modulus, components of the equipment are: - (i) it is
although the few researchers tried to developed equipped for applying a drive load up to 100 kN
a correction factor for deflection. Correction with a pulse duration of around 20-30
factors are geographical locations, environment milliseconds and (ii) all the operations are
and material specified, and varies from place to controlled from PC/laptop through the DS4000S
place. On account of the empirical nature of it, it data acquisition system.
is not considered as standard and requires The size and shape of the deflection basin permit
separately specified correction factors. For comprehensive structural investigation of the

11
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

pavement. Fundamentally, the exterior Bitumen material is susceptible to the

deflections describe the modulus characteristics temperature variation and does change in
of the subgrade, although the bowl nearby to the characteristics of it, also the temperature
loading plate permits investigation of the variation effect on the FWD measured
modulus characteristics of the nearby surface deflections. Granular layer and subgrade
layers. A number of empirical models are materials are susceptible to moisture variation.
developed for the estimation of layer moduli Correction factors are geographical locations,
from radially measured deflection basin with environment and material specified, and varies
another parameter such as applied load, layer from place to place. On account of the empirical
thickness. However, these models are effective nature of it, it is not considered as standard and
for the sets of conditions and construction requires separately specified correction factors.
methodology for which they were developed.
Therefore, it is a necessary to validate these REFERENCES
models for a different set of conditions and also [1]. Hoffman, M.S. and Thompson, M.R. (1982).
due to empirical nature of these models, they are Comparative Study of Selected
not widely used for estimation of layer moduli. Nondestructive Testing Devices.
Extensively used method for the Transportation Research Record No. 852,
estimation of layer moduli is back-calculation Transportation Research Board,
method. Back-calculation procedure is Washington, DC, pp. 32-42.
depending on coordinating of computed and [2]. Fleming, P.R., Frost, M.W., & Lambert, J.P.
measured pavement deflections. A most widely (2007). Review of lightweight deflectometer
used example is a linear material model and for routine in situ assessment of pavement
static loading model based layered elastic material stiffness. Transportation research
programs is KENLAYER. but it doesn’t record: journal of the Transportation
consider a non-linearly characteristic of Research Board, 2004(1), 80-87.
materials and that lead to inappropriate results. [3]. R-81 Research Scheme (2003), "Structural
To get accurate results it is recommended that to Evaluation of Pavements in Eastern India
use a non-linear dynamic pavement response using Falling Weight Deflectometer", Final
model. For that a requires a profoundly Report submitted by the Indian Institute of
computational effective PC program, for Technology, Kharagpur to the Ministry of
example, Finite element method (FEM). Road Transport and Highways, Government
of India

12
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

[4]. Fwa, T. F. (Ed.). (2005). the handbook of Baladi, ASTM, Philadelphia, PA, pp. 502-
highway engineering. CRC Press. 524.
[5]. Kumar, R.S., Kumar, S., Das, A., Reddy, [10]. Roque, R., Ruth, B.E. and Sedwick, S.C.
K.S., Mazumdar, M. and Pandey, B.B. (1998). Limitations of Backcalculation and
(2001). Development of an Impact Improved Methods for Pavement Layer
Apparatus for Evaluation of Elastic Modulus Moduli Predictions. Proceedings of 5th
of Pavement Layers. Indian Geotechnical International Conference on Bearing
Society Journal, Vol.31, No.3, IGS, New Capacity of Roads and Airfields, BCRA,
Delhi, July 2001, pp. 273-284. Vol. 1,pp. 409-417.
[6]. American Association of State Highway and [11]. Molenaar, A.A.A., and Van Gurp
Transportation Officials (AASHTO) Guide Ch.A.P.M. (1982). A Pavement
for Design of Pavement Structures (1993) Management System for Provincial Roads in
[7]. Garg, G., and Thompson, M.R. (1998). the Netherland. Proceedings of 5th
Structural Modelling of Flexible Pavements International Conference on Structural
in the LVR Test Sections at Mn/ Road Design of Asphalt Pavements, Vol. 1, pp.
project. Proceedings of 5th International 485-499.
Conference on Bearing Capacity of Roads [12]. Harr, M.E. (1966). Foundations of
and Airfields, Trondheim, Norway, pp. 419- Theoretical Soil Mechanics. McGraw-Hill,
428. P. 81.
[8]. Choubane, R. and McNamara, R.L. (2000). [13]. Wimsatt, A.J. (1999). Direct Analysis
A Practical Approach to Predicting Flexible Methods for Falling Weight Deflectometer
Pavement Embankment Moduli using Data.
Falling Weight Deflectometer [FWD] Data. [14]. Badu, K.T., Manzione, C.W., Ruth, B. E.
Research Report, FL/DOT/SMO/00-442, and Miley, W. G. (1989). Prediction of
State of Florida. Flexible Pavement Layer Moduli from
[9]. Alexander, D.R., Kohn, S.D. and Grogan, Dynaflect and FWD Deflections.
W.P. (1989). Nondestructive Testing Nondestructive Testing of Pavements,
Techniques and Evaluation Procedures for ASTM STP 1026, Philadephia, PA 19103,
Airfield Pavements, Nondestructive Testing pp. 245-277.
of Pavements and Backcalculation of [15]. Chang, D. W., Kang, Y. V., Roesset, J.
Moduli. ASTM STP 1026, Eds. Bush and M., & Stokoe, K. (1992). Effect of depth to
bedrock on deflection basins obtained with

13
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

Dynaflect and falling weight deflectometer [21]. Ullidtz, P. and Peattie, K.R. (1982).
tests. Transportation Research Record, 8-8. Programmable Calculators in the
Huang, Y. H. (1993). Pavement analysis and Assessment of Overlays and Maintenance
design. Strategies. Proceedings of 5th International
[16]. Huang, Y. H. (1993). Pavement analysis Conference on Structural Design of Asphalt
and design. Pavements, Vol.1, Delft, pp. 671-681.Under
[17]. Anderson, M. (1989). A data base Repeated Loading. Australian Road
method for backcalculation of composite Research Board Proceedings, Vol. 8, pp.1-
pavement layer moduli. In Nondestructive 12.
testing of pavements and backcalculation of [22]. Rada, G.R., Witczack, M.W. and
moduli. ASTM International. Rabinow, S.D. (1988). Comparison of
[18]. Tia, M., Eom, K. S., & Ruth, B. E. AASHTO Structural Evaluation Techniques
(1989). Development of the DBCONPAS using Nondestructive Deflection Testing.
computer program for estimation of concrete Transportation Research Record No. 1207,
pavement parameters from FWD data. TRB, Washington, DC, pp.134-
In Nondestructive Testing of Pavements and 144.Antunes, M.L. (1993). Pavement
Backcalculation of Moduli. ASTM Bearing Capacity Evaluation using Dynamic
International. Nondestructive Tests. Ph. D Thesis, IST,
[19]. Zaghloul, S. M., White, T. D., Drnevich, Lisbon.
V. P., & Coree, B. (1994). Dyanamic [23]. Antunes, M.L. (1993). Pavement
Analysis of FWD Loading and Pavement Bearing Capacity Evaluation using Dynamic
Responce Using a Three-Dimensional Nondestructive Tests. Ph. D Thesis, IST,
Dynamic Finite Element Program. Lisbon.
In Nondestructive Testing of Pavements and [24]. Kim, Y.K., Lee, Y.C. and Ranjithan,
Backcalculation of Moduli: Second Volume. S.R. (2000). Flexible Pavement Condition
ASTM International. Evaluation using Deflection Basin
[20]. Fwa, T.F. and Chandrasegaram, S., Parameters and Dynamic Finite Element
(2001), Regression models for Analysis Implemented by Artificial Neural
backcalculation of concrete pavement Networks. Nondestructive Testing of
properties, J. Transport. Eng., ASCE,127, 4, Pavements and Backcalculation of Moduli:
353–356 Third Volume, STP 1375, Eds. Tayabji, S.

14
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

D. and Lukanen, E. O. ASTM, West for FWD Measurements. In Proceedings of

Conshohocken, PA, pp. 514-530. 4th International Conference on Bearing
[25]. Chen, D.H., Bilyeu, J., Lin, H.H. and Capacity of Roads and Airfields, Vol.1,
Murphy. (2001). Temperature Correction on MinneapolisBoard, pp. 15-31.
Falling Weight Deflectometer [29]. Ali, H, and Selezneva, O. (2000).
Measurements. Transportation Research Seasonal Trends and Causes in Pavement
Record No. 1716, Transportation Research Structural Properties. Nondestructive
Board, Washington, DC, pp. 30-39. Testing of Pavements and Backcalculation
[26]. Johnson, A.M. and Baus, R.L. (1992). of Moduli, 3 rd Volume, Eds. Tayabji, S. D.
Alternative Method for Temperature and Lukanen, E. O, ASTM STP 1375,
Correction of Backcalculated Equivalent ASTM, West Conshohocken, PA, pp.144-
Pavement Moduli. Transportation Research 160.
Record No. 1355, Transportation Research [30]. IRC 115:(2014) Guidelines for
Board, Washington, DC, pp. 25-36 Structural Evaluation and Strengthening of
[27]. Ullidtz, P. (1987). Pavement Analysis. Flexible Road Pavements Using Falling
Elsevier, Amsterdam. Weight Deflectometer (FWD) Technique
[28]. Baltzer, S. and Jansen, J.M. (1994).
Temperature Correction of Asphalt Moduli

LIST OF TABLE:
Table 1 International overview of FWD
Weight
Loading
Peak and Load
Manufacturer Deflections plate
load height of durations Remarks
Model Sensors diameter
(KN) falling (ms)
(mm)
mass
50 to 300 7 velocity
Dynatest 7 to kg transducers
25 to 30 300
Model 8000 120 20 to 380 Spacing 2.25 Denmark
mm mm apart and UK
Dynatest 30 to 7 velocity
-- 25 to 30 --
Model 8081 240 transducers

15
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

Spacing 2.25
mm apart
30 to 6 velocity
Phonix FWD
10.2 to 150kg transducers Europe and
Model ML- -- --
102.3 50 to 400 Spacing 2.4 US
10000
mm mm apart
KUAB 2M- 5 velocity (Sweden)
14 to 300 and
FWD -- -- transducers 2 – Mass
150 450
Model 8333 system. a
falling
weight
KUAB 2M- 5 velocity
300 and dropped on
FWD 7 to 65 -- -- transducers
450 second
Model 8714
buffer
weight

Source: [3] ( R-81 Research Scheme (2003)) and [4] (Fwa, T. F. (Ed.). (2005)).

LIST OF FIGURES

Figure- 1

Figure 1 Working Principle of Falling Weight Deflectometer with deflection bowl

16
 ELK ASIA PACIFIC JOURNAL OF CIVIL ENGINEERING AND STRUCTURAL DEVELOPMENT

ISSN 2394-9341 (Online); Volume 4 Issue 1 (2018)

Figure- 2

Figure 2 Load-Deflection time histories

Figure- 3

Figure 3 components of back calculation

17

View publication stats