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

net/publication/326355198

Use of Falling Weight Deflectometer for Airport Pavements: Proceedings of the

5th GeoChina International Conference 2018 – Civil Infrastructures Confronting
Severe Weathers and Cli...

Chapter · January 2019

DOI: 10.1007/978-3-319-95768-5_11

CITATIONS READS

0 818

1 author:

Greg White
University of the Sunshine Coast
129 PUBLICATIONS   498 CITATIONS   

SEE PROFILE

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

Understanding the mechanisms, causes and methods of evaluation for ravelling of grooved and ungrooved airport asphalt View project

Waste plastic as an asphalt binder/mixture modifier View project

All content following this page was uploaded by Greg White on 24 August 2018.

The user has requested enhancement of the downloaded file.

 This is a pre-publication formatted paper cited as:
White, G 2018, ‘Use of Falling Weight Deflectometer for Airport Pavements’, 5th
GeoChina International Conference, HangZhou, China, 23-25 July.

Use of Falling Weight Deflectometer for Airport Pavements

Greg White1, PhD, CPEng

1
Director, Airport Pavement Research Program, University of the Sunshine Coast,
Queensland, Australia, [email protected]

ABSTRACT: The Falling Weight Deflectometer (FWD) provides a rapid and cost-
effective means of measuring the response of pavements to a dynamic load. Its broad
application has allowed the development of significant guidance regarding empirical
interpretation of FWD results. Software has also been developed allowing the
estimation of layer modulus, including the subgrade, and in some cases the calculation
of a theoretical strength rating for aircraft pavement structures. However, many
practitioners question the reasonableness of the moduli calculated and the reliably of
the resulting airport pavement strength rating. This paper reviews the use of FWD in
airport pavement strength evaluations and assesses its appropriateness for particular
situations and tasks. Example projects are considered and data analysed to determine
the benefits and limitations. Further work is recommended to better understand the
impact of variability in FWD results on inferred layer modulus and airport pavement
strength rating.

INTRODUCTION
With airports getting busier and upgrade funding becoming tighter, Non-Destructive
Testing (NDT) which is rapid to perform and relatively economical to procure, has
gained popularity. Particularly for busy airport runways that can not readily be closed
for days of traditional intrusive pavement investigation (White 2017). The Falling
Weight Deflectometer (FWD) is arguably the most common NDT device used for the
structural evaluation of existing airport pavement infrastructure (Celaya & Nazarian
2014).

FWDs have been in use since the 1980s. These generally trailer-mounted NDT
devices apply a dynamic load while the trailer is stationary. The load can be varied
and deflections are estimated from surface strain accelerations measured by geophones
placed at various distances from the load application (TRB 2008). This results in a
deflection profile or bowl that is a function of the composition and condition of the
pavement (Vuong 1989).

This paper reviews the use of FWD in airport pavement strength evaluations and
assesses its appropriateness for particular situations and tasks. Example projects are
presented and data analysed to determine the benefits and limitations. Further work is
recommended to better understand the impact of variability in FWD responses on
inferred layer modulus and airport pavement strength rating.

Page 1
BACKGROUND

Airport Pavement Strength Rating System

Airport pavements are usually either flexible or rigid structures, although composite
pavements are also used in some circumstances. Rigid airfield pavements are often
plain jointed concrete on a stabilised sub-base. The joints are required to transfer
loads, either by aggregate interlock across sawn contraction joints or by dowels across
formed construction joints. Flexible pavements usually comprise significant thickness
of crushed rock base and sub-base with a thin (50-100 mm) or thick (100-250 mm)
asphalt surface, although some regional airports use a bituminous sprayed seal
wearing surface.

Both rigid and flexible airport pavements are subject to an international strength rating
system known as ACN-PCN (ICAO 1983). The system is similar to road axle load
limits but reflects the higher level of deviation between small aircraft and large
aircraft, meaning every airport has its own Pavement Classification Number (PCN),
which represents the upper limit of Aircraft Classification number (ACN) allowed for
unrestricted operations.

Traditionally an airport’s PCN was determined by reverse engineering the existing

pavement structure, including layer thicknesses and materials, as well as the subgrade
CBR, to determine the acceptability of a particular aircraft operating at a particular
mass. The PCN was generally set to the highest ACN of all the aircraft considered to
be acceptable for the particular pavement. This approach remains the normal method
for PCN assignment in many countries (FAA 2011). However, as detailed below,
software is now available to automatically calculate a PCN, primarily from FWD
deflection data.

Falling Weight Deflectometer

There are at least four manufacturers of FWD devices, although Dynatest is the most
commonly encountered brand. All FWDs apply a vertically dynamic load onto a 300-
450 mm diameter load plate. The duration of the dynamic loading is generally 20-
65 milliseconds (Ameri et al. 2009). The load generally ranges from 7 to 150 kN for
the standard FWD, although a light version (LWD) applies 1-15 kN and various heavy
versions (HWD) apply up to 250 kN (FAA 2011). Deflections are estimated from
geophone measurements of surface acceleration, that can be spaced as required,
usually anywhere from 0-2400 mm from the centre of the loading plate. Standard
distances for geophone locations are 0, 300 600, 900, 1200, 1500 and 1800 mm from
the centre of the load plate. However, different countries adopt different spacings.
For example, Australia specifies a minimum of seven geophones, with three located 0,
200 and 900 mm from the load plate centre with the location of the other geophones
optional (Austroads 2011). This approach reflects the focus on the deflections at 0,
200 and 900 mm for the characterisation of deflection bowl, using the generic term
‘Dn’ to represent the deflection at the geophone located n mm from the centre of the
load pate.

Page 2
Deflection bowls are generally characterised by three parameters (Austroads 2011):
• Maximum deflection (D0). A general indicator of pavement stiffness and
response.
• Curvature (D0-D200). An indicator of the upper base course and surface layer
stiffness.
• D900. An indicator of the subgrade support condition.

The above indicators are only general in nature. For example, a very thick pavement’s
subgrade is less likely to be adequately characterised by D900 because the measured
deflection at 900 mm from the load plate centre will be less affected by the subgrade
when the subgrade is located much deeper than typical.

Modulus Back Calculation

A number of softwares are now available for the back calculation of layer modulus
values from deflection bowl data. Example softwares include Modulus, Elmod,
Evercalc (Ameri et al. 2009), DAPS (Bandara et al. 2002), Elsedef, Illi-Back (Kang
1998) and EfromD (Vuong 1989). Generally, the software requires pavement layer
thicknesses to be nominated as well as modulus seed values. Although attractive to
the designer, absolute back calculated modulus values should be used with caution
(Gendreau and Soriano 1998) due to the high degree of variance between reported and
observed pavement performance (Zaniewski 1991). Importantly, the nominated layer
thicknesses can significantly affect the resulting layer modulus values, as detailed
later.

Due to the prevalence of the Dynatest FWD, Dynatest’s ELMOD (Dynatest 2017) is
arguably the most common back analysis software used by practitioners. The latest
version, ELMOD 6, also includes PCN determination directly from deflection bowl
data.

Following layer modulus estimation, the software uses nominated aircraft type and
frequency to determine the most damaging aircraft for each nominated pavement
layer. The ACN of the most damaging aircraft at the ‘just acceptable’ aircraft mass is
reported as the allowable pavement PCN. The allowable mass may be higher than the
published maximum mass of the aircraft in order to ‘just fail’ the pavement by the end
of the design life. The pavement damage caused by each aircraft is based on an
allowable stress or strain levels for each pavement layer. This approach does not take
into account superposition of the impact of multiple aircraft types, relies upon the
allowable stress/strain determined for various pavement materials and relies upon the
accuracy of the pavement layer thicknesses.

In ELMOD 6, every test location has a PCN value reported. The PCN values vary and
a protocol is required for determining a characteristic PCN from the hundreds of test
points.

Page 3
EXAMPLE APPLICATIONS

Homogenous Sectioning
Arguably, the most common application for FWD surveys in airport pavements is the
identification of sections of pavement of homogenous response to load. Sections of
homogenous response inform intrusive testing locations for increased efficiency.
Rather than randomly locating intrusive test points, the intrusive testing is planned to
reflect the designated sections of homogenous response to load. The approach
increases efficiency for pavements that are operationally difficult or expensive to close
for intrusive testing.

Maximum deflection (D0) is often used as the basis of homogenous sectioning.

However, some examples have also used CV and D900 in combination with D0. An
example is shown in Figure 1 where D0 analysis (Figure 1a) indicates clear changes
in response at approximately CH400 and CH1400. The 1000 m in between is very
consistent. Figure 1b (CV) indicates that the upper base course is generally consistent
along the full length of the runway, while the subgrade response (D900) indicates the
changes in D0 most likely reflect changes in the subgrade condition (Figure 1c).
Intrusive testing subsequently verified this finding, encountering a change in subgrade
material type at CH1400.

Identifying pavement areas of homogenous response is often performed visually,

however, statistical tools are also available. Cumulative sum analysis was developed
in the 1950s for statistical quality control of sequential processes developed. The
cumulative differences method is an adaptation recommended by the US Highways
Associations (AASHTO 1993) designed for identifying trends in various pavement
responses (Thomas 2004). Conveniently, cumulative differences graphs start and end
at 0, and changes in gradient indicate changes in pavement response trend, as shown in
the example at Figure 2 for D0 values along a runway centreline and at three different
offsets either side of the centreline. The cumulative difference chart indicates changes
in overall pavement response at chainages 700 m and 1700 m, resulting in three
homogenous sections. A similar analysis was also undertaken on CV and D900 values.
One disadvantage of the cumulative differences approach is a reduction in intuitive
interpretation of the results, reflecting the ‘relative to the average’ nature of the values.

Pavement Layer Modulus Estimation

As discussed above, various softwares are available for estimating layer modulus from
deflection bowl data. ELMOD 6 is currently the most common, reflecting the
prevalence of the Dynatest FWD equipment. Like other softwares, ELMOD 6
requires the layer thicknesses to be nominated, as well as seed values of modulus.
Depending on the software used, two main approaches are possible. The first returns
an estimated modulus value for each recorded test location. A representative value is
then determined by selecting a percentile-based representative value of modulus for
each layer, with 10 and 15 percentile values common. The second approach requires
all the deflection bowl data to be analysed, a representative deflection bowl adopted
and the representative deflective bowl converted to estimated modulus values for the

Page 4
various layers. This approach requires more interpretation of the deflection data.
Consequently, practitioners not familiar with FWD analysis often prefer the
percentile-value approach, which requires less judgment.

The data presented in Figure 1 was analysis by ELMOD 6, assuming a four-layer

system (40 mm asphalt, 160 mm base, 200 mm sub-base and subgrade). Table 1
summarises the data for the three homogenous sections. The variability is
concerningly high, with coefficients of variation (CoVs) ranging from 30% to 160%.
Similarly, the modulus of the surface layer varies significantly, despite all surface
layers at the airport being constructed and surfaced as a single project. The estimated
modulus values range from 72 MPa to 48,900 MPa and the interquartile range
(difference between 75 and 25 percentile values) divided by the median is 0.44 to 0.59
across the three homogenous sections, which is high.

Subgrade CBR Estimation

ELMOD 6 also estimates modulus values for the subgrade layer, which can be used to
estimate a subgrade CBR value. There are many equations for the estimation of
subgrade CBR from modulus, and vice-versa, with CBR = Modulus / 10 commonly
adopted for airport pavement design purposes. However, ELMOD 6 uses a power-law
conversion as shown in Figure 3, approximated as CBR = 0.0113 × Modulus1.5625.
This significantly increases the subgrade CBR over the typical airport pavement
design CBR range of 3-15, by up to 100%.

Like the modulus values, the CBR values are concerningly variable with even the
most homogenous section from Figure 1 (CH 380-1370) having a CoV of 67% for
estimated CBR with values ranging from 1 to 61. This range of CBR values is highly
unlikely within a generally uniform site and a runway of generally sound performance.

As another example, the estimated 10 percentile CBR values, by homogenous section

and centreline offset, for the airport represented in Figure 2, is shown in Figure 4.
Although some variation along the runway length is accepted, it is unlikely that the
subgrade CBR actually varies from above 20 at offsets 3 m, 6 m and 9 m either side of
the runway, to just 6 along the centreline. Further, the performance of the airfield did
not reflect this variation in performance under aircraft traffic, adding weight to the
belief that the data is not representative of the true subgrade condition. It is
hypothesized that testing on the crown may have affected the results by preventing
uniform contact between the FWD loading plate and the pavement surface. However,
further work is required to better understand the potential source of these illogical
results.

Pavement Strength Rating

Once the subgrade CBR and the layer modulus values are determined, ELMOD 6 can
also determine a pavement strength rating, expressed as PCN. To complete the
process, a traffic spectrum, consisting of one or more design aircraft, must be
nominated. As summarised in Table 2, the PCN values are almost as variable as the
modulus values. The calculated PCN values range from 2 to 27 and the 25 and 75

Page 5
percentile values are 4 to 10. These are significant deviations in PCN considering the
damaging effect of aircraft is exponentially related to the PCN (Figure 5). For the
runway in question, the published PCN is 14, near the maximum value calculated by
ELMOD 6, which is actually the 93 percentile value. This compares to common
practice which would recommend the 15 percentile value, in this case 4. Therefore,
given the generally good performance of the runway, adopting the ELMOD 6
recommended PCN would significantly and unjustifiably limit to use of the runway.

ELMOD 6 required a nominated number of layers and a thickness for each.

Consequently, for a given aircraft, the estimated layer modulus values and subgrade
CBR are the only other factors that will impact the PCN value calculated by
ELMOD 6. Because aircraft pavement design is highly sensitive to subgrade CBR
(White 2005) the PCN values are expected to be highly correlated to the estimated
subgrade CBR values. However, this is not the case, as shown in Figure 6. This can
only reflect the unrealistically high influence of the estimated asphalt, base and sub-
base modulus values of the resulting ELMOD 6 PCN, because these are the only other
factors affecting the calculated PCN values.

Concrete Joint Load Transfer

Rigid airfield pavement thickness design determines a centre-slab thickness of
concrete. Stress concentrations at slab joints require an additional 25% of the centre-
slab thickness to withstand the increased magnitude of stress. However, load transfer
from the loaded slab to the adjacent slab is relied upon to avoid the increased concrete
thickness. In practice, longitudinal construction joints are usually dowelled to affect
load transfer, while transverse sawn contraction joints rely on aggregate interlock. In
older pavements, load transfer efficiency is often assessed by FWD.

FWD analysis of load transfer efficiency is usually expressed by the ratio of FWD
deflection measured 300 mm either side of the FWD load plate, with the FWD
orientated so that the joint is located between the load plate and the D300 geophone. In
practice, the transverse and longitudinal joints, the corner of the slab and the centre of
the slab are all assessed, as illustrated by Figure 7.

Example rigid airfield pavement joint load transfer survey results are summarised in
Table 3 and illustrated as box-and-whisker plots in Figure 8. The near-perfect
symmetry of responses measured 300 mm either side of the load plate is demonstrated
by the average ratio of 98% and low standard deviation of 3%. The longitudinal joints
exhibit excellent load transfer, with an average deflection ratio of 98%, while the
transverse joints are significantly less efficient with an average deflection ratio of 45%
and a standard deviation of 28%. This represents the less reliable load transfer by
aggregate interlock across old sawn joints, compared to that of dowels. The slab
corner results are similar to those for the transverse joints.

Page 6
LIMITATIONS

Repeatability of Response
Repeatability of the response is believed to account for some of the variability in
pavement response, as detailed below. The repeatability of a single FWD machine
and the reproducibility across a fleet of six FWDs in the USA indicated the
importance of standardised calibration to improve reproducibility (Rocha et al. 2004).
The study also identified reduced repeatability as the distance of the geophone from
the load plate increased. A simple comparison of repeated test configuration and
location within the data presented in Figure 1 supports these findings, with good
agreement between the first and repeated drop, with a linear correlation of R=96%,
although some significant outliers were identified Figure 9. Further work is
recommended to understand the typical reproducibility of FWD results.

Variability of Pavement Response

Generally homogenous sections of pavement, of nominally consistent construction,
indicate significant variability in pavement responses. For example, the homogenous
section of pavement represented in Figure 1 (CH 380-1370) exhibits high variability
in the measured D0, curvature and D900, with CoV values of 19%, 31% and 17%,
respectively (Table 4). Further work is required to understand if this level of
variability is typical and its implications for airport pavement evaluation.

Reliance on Input Assumptions

As detailed above, the existing pavement layer thicknesses must be input into
ELMOD 6 for modulus and PCN calculation. The assumed layer thickness has a
significant effect on the estimated modulus value. For example, a light aircraft
runway was surveyed by FWD but historical records could not determine whether the
existing base course was 150 mm or 250 mm thick. Layer modulus values were
determined by ELMOD 6 using both thicknesses (Figure 10). When the greater base
course thickness was assumed, both the base course and subgrade modulus reduced by
an average of approximately 30% (Table 5). However, the reduction in estimated
modulus was not consistent. Further work is required to understand if this level of
variability is typical for airport pavement FWD surveys. Further, the sensitivity of
pavement life to pavement thickness (White 2005) results in the layer thickness
assumption having a significant effect on the ELMOD 6 calculated PCN values. For
the previous example, the increased base course thickness assumption resulted in an
average PCN increase of 74%, despite the decrease in base and subgrade modulus
(Table 5). Further work is required to better understand the impact of input
assumptions on the estimated modulus and PCN values.

CONCLUSIONS

FWD devices provide a rapid and cost-effective means of assessing airport pavement
response to load. Although reported to be reasonably repeatable, reproducibility
between devices is less certain. The significant response variability observed in
generally homogenous sections of pavement is concerning and further work is

Page 7
recommended to understand whether the variability of the presented data is typical.
Additional work is also recommended to determine the impact of FWD response
variability and input assumptions on back-analysed layer moduli and airport pavement
PCN values. Understanding these issues will further increase the usefulness of FWD
surveys in airport pavement assessment and management.

Acknowledgments
The provision of FWD data and advice regarding testing protocols provided by Jim
Johnson-Clarke or ARRB (Melbourne, Australia) is gratefully appreciated.

References

AASHTO 1993, Guide for Design of Pavement Structure, Appendix J: Analysis Unit
Delineation by Cumulative Differences, American Association of State Highway
and Transportation Official, Washington, District of Columbia, USA.
Ameri, M, Yavari, N & Scullion, T 2009, ‘Comparison of static and dynamic
backcalculation of flexible pavement layer moduli, using four software programs’,
Asian Journal of Applied Sciences, vol. 2, no. 3, pp. 197-210.
Austroads 2011, Pavement Deflection Measurement with a Falling Weight
Deflectometer, Austroads Test Method AG:AM/T006, 31 March.
Bandara, N, Rowe, GM, Sharrock, MJ & Nickerson, CR 2002, ‘Seasonal variation of
subgrade modulus in different subgrade soils for pavement rehabilitation for non
freeze-thaw cycles’, Applications of Advanced Technology in Transportation, pp.
473-480.
Celaya, M & Nazarian, S 2014, ‘Field evaluation of NDT devices for delamination
detection of HMA airport pavements’, FAA Worldwide Airport Technology
Transfer Conference, Galloway, New Jersey, USA, 5-7 August.
Dynatest 2017, ELMOD 6 Quick Start Manual, Dynatest International.
FAA 2011, Use of Nondestructive Testing in Pavement Evaluations, Advisory
Circular 150/5370-11B, Federal Aviation Administration, Washington, District of
Columbia, USA, 30 September.
Gendreau, M & Soriano, P 1998, ‘Airport pavement management systems: an
appraisal of existing methodologies’, Transportation Research, vol. 32, no. 3, pp.
197-214.
ICAO 1983, Aerodrome Design Manager: Part 3, ICAO 9157, 2nd edn., International
Civil Aviation Organization, Montreal, Quebec, Canada, January.
Kang, YV 1998, ‘Multifrequency back-calculation of pavement layer moduli’, Journal
of Transportation Engineering, vol. 124, no. 1, pp. 73-81.
Rocha, S, Tandon, V & Nazarian, S 2004, ‘Falling weight deflectometer fleet’, Road
Material and Pavement Design, vol. 5, no. 2., pp. 215-238.
Thomas, F 2004, ‘Generating homogenous road sections based on surface
measurements: available methods’, 2nd European Pavement and Asset Management
Conference, Berlin, Germany, 21-23 March.
TRB 2008, Falling Weight Deflectometer Usage, NCHRP Synthesis 381,
Transportation Research Board, Washington, District of Columbia, USA.
Vuong, B 1989, ‘A new linear elastic back-calculation model for back-calculating

Page 8
 layer moduli at fixed Poisson’s ratio’, Australian Road Research, vol. 19, no. 1, pp.
17-28.
White, G 2005, ‘A sensitivity analysis of APSDS, an Australian mechanistic design
tool for flexible aircraft pavement thickness determination’, First European
Aircraft Pavement Workshop, Amsterdam, Netherlands, 11-12 May.
White, G 2017, ‘Expedient runway upgrade technologies’, 10th International
Conference on the Bearing Capacity of Roads, Railways and Airfields, Athens,
Greece, 28-30 June.
Zaniewski, J 1991, Unified Methodology for Airport Pavement Analysis and Design,
Technical Report DOT/FAA/RD-91/15,I, Federal Aviation Administration,
Washington, District of Columbia, USA, June.

TABLES

Table 1. Summary of estimated modulus values

Statistic 40 mm asphalt 160 mm base 200 mm sub-base

Homogenous Section CH 0-370
Minimum 3,971 62 48
25 percentile 13,501 118 186
Average 16,769 158 517
Median 16,709 144 298
75 percentile 20,925 232 441
Maximum 30,916 478 6,217
Standard Deviation 5,050 67 679
CV 30.1% 42.5% 131.5%
Number of Results 185
Homogenous Section CH 380-1370
Minimum 72 53 27
25 percentile 9,925 137 79
Average 13,290 180 160
Median 12,551 168 112
75 percentile 15,752 209 165
Maximum 34,319 764 3,943
Standard Deviation 5,123 71 257
CV 38.5% 39.1% 160.4%
Number of Results 500
Homogenous Section CH 1380-1820
Minimum 2,329 49 47
25 percentile 11,153 124 159
Average 16,293 184 351
Median 15,900 173 257
75 percentile 20,488 231 441
Maximum 48,931 428 2,931
Standard Deviation 6,936 82 312
CV 42.6% 44.6% 88.8%
Number of Results 225

Page 9
 Table 2. Summary of calculated PCN values

Statistic CH 0-370 CH 380-1370 CH 1380-1820

Minimum 4 2 3
25 percentile 8 4 6
Average 11 6 8
Median 10 5 8
75 percentile 10 7 10
Maximum 27 16 20
Standard Deviation 4 2 3
CV 38.3% 44.3% 42.0%
Number of Results 185 500 225

Table 3. Summary of joint load transfer efficiency

Statistic Centre of Slab Longitudinal Transverse Corner of

Joints Joints Slab
Minimum 83% 90% 3% 3%
25 percentile 98% 97% 21% 30%
Average 98% 99% 45% 55%
Median 99% 99% 39% 50%
75 percentile 100% 102% 58% 81%
Maximum 107% 109% 113% 136%
Standard Deviation 3.1% 3.3% 28.4% 33.1%
Number of Results 146 74 74 74

Table 4. Summary of measured deflection values

Statistic D0 CV D900
Minimum 894 170 122
25 percentile 1464 226 344
Average 1690 255 431
Median 1691 249 411
75 percentile 1896 284 498
Maximum 2607 425 1038
Standard Deviation 500 500 500
CV 317 42 132

Page 10
 Table 5. Layer thickness impact on estimated modulus and PCN

Statistic Base Modulus Subgrade Modulus PCN

(MPa) (MPa)
Minimum 0.55 0.36 1.38
25 percentile 0.68 0.66 1.66
Average 0.76 0.71 1.77
Median 0.69 0.73 1.74
75 percentile 0.73 0.76 1.83
Maximum 1.25 1.35 2.45
Standard Deviation 60 60 60
CV 0.16 0.19 0.23

Page 11
FIGURES

3000

2500

2000
Deflection (microns)

1500

1000

500
Centreline 3m left of centreline 3m right of centreline
6m left of centreline 6m right of centreline
0
0.000 0.200 0.400 0.600 0.800 1.000 1.200 1.400 1.600 1.800
Chainage in km from 05 end (km)

FIG. 1(a). Example pavement response – D0

1400
Centreline 3m left of centreline 3m right of centreline
6m left of centreline 6m right of centreline
1200

1000
Curvature (microns)

800

600

400

200

0
0.000 0.200 0.400 0.600 0.800 1.000 1.200 1.400 1.600 1.800
Chainage in km from 05 end (km)

FIG. 1(b). Example pavement response – CV

500

400

300
D-900 (microns)

200

100

Centreline 3m left of centreline 3m right of centreline

6m left of centreline 6m right of centreline
0
0.000 0.200 0.400 0.600 0.800 1.000 1.200 1.400 1.600 1.800
Chainage in km from 05 end (km)

FIG. 1(c). Example pavement response – D900

Page 12
 1000

-1000

-2000

-3000

Cumulative Differences -4000

-5000

-6000

-7000

-8000
cl cl+3 cl-3 cl+6 cl-6 cl+9 cl-9
-9000
0 500 1000 1500 2000 2500 3000
Chainage (m)

FIG. 2. Example cumulative difference for D0

50

45
y = 0.0113x1.5625
40

35
Subgrade Converted CBR (%)

30

25

20

y = 0.1x
15

10

0
0 20 40 60 80 100 120 140 160 180 200
Subgrade Estimated Modulus (MPa)

FIG. 3. Inferred ELMOD 6 subgrade CBR from Modulus

35

30

25
10 perecentile CBR (%)

20

15

10

0
0-740 m 740-1800 m 1800-3000 m
centerline +9 m centerline +6 m centerline +3 m centerline
centerline -3 m centerline -6 m centerline -9 m

FIG. 4. Example estimated subgrade CBR by FWD offset

Page 13
 160

Equivalent Coverages of similar aircraft at ACN-PCN equal to 1.0

140

120

100

80

60

40

20

0
0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5
ACN-PCN Ratio

1 wheel 2 wheels 4 wheels 6 wheels

FIG. 5. Typical aircraft overload damage as a function of ACN ratio

30

25
Calculated PCN

20

15

10

0
0 5 10 15 20 25 30 35 40 45 50
Estimated Subgrade CBR (%)

FIG. 6. ELMOD 6 recommended PCN as a function of calculated subgrade CBR

FIG. 7. Typical joint transfer efficient assessment configurations

Page 14
 160%

140%

120%

Calculated Joint Transfer Efficicny

100%

80%

60%

40%

20%

0%
Centre Longituindal Transverse Corner

FIG. 8. Example calculated joint transfer efficiency summary

1400

1200
Repeated Drop Response (microns)

1000

800

600

400

200

0
0 200 400 600 800 1000 1200 1400
First Drop Response (microns)

FIG. 9. Example comparison of repeated FWD response

800

700

600
FWD Estimated Modulus (MPa)

500

400

300

200

100

0
0 200 400 600 800 1000 1200 1400 1600 1800
Runway Chainage (m)

150 mm - base 150 mm - subgrade 250 mm - base 250 mm - subgrade

FIG. 10. Example effect of base thickness on base and subgrade modulus

Page 15
View publication stats