Full-Scale Field Tests on Flexible Pipes

under Live Load Application

Madasamy Arockiasamy1; Omar Chaallal2; and Terdkiat Limpeteeprakarn3

Abstract: This paper describes the procedure and results of the field tests on high-density polyethylene 共HDPE兲, PVC, and metal large
diameter pipes subjected to a highway design truck loading. Numerical simulations using finite element method are performed to
determine pipe-soil system response under live load application. Comparisons of field test data with the predicted responses are made for
soil pressures around and above the pipes, deformed cross-sectional pipe profiles, and pipe deflections. The field test results indicated that
the buried flexible pipes, embedded with highly compacted graded sand with silt, demonstrated good performance without exhibiting any
visible joint opening or structural distress. Under shallow burial conditions, the AASHTO specified deflection limit of 5% is found to be
adequate for installation of the flexible pipes during the construction phase, and a vertical deflection limit of 2% is suggested for HDPE
pipes based on the truck load response and repeated loading effect.
DOI: 10.1061/共ASCE兲0887-3828共2006兲20:1共21兲
CE Database subject headings: Soil-structure interaction; Soil pressure; Flexible pipes; Culverts; Storm sewers; Live loads; Load
distribution; Highways; Trucks.

Introduction other important factors that govern the soil pressure distribution
over the buried flexible pipes.
Flexible pipes, corrugated high-density polyethylene 共HDPE兲, Literature review, conducted on buried flexible pipes subjected
polyvinyl chloride 共PVC兲, and metal pipes are gaining popularity to live load application, indicates that a minimum soil cover over
for use as buried underground conduits for road way and highway the pipe crown appears to be the most important parameter on
gravity-flow applications. In these applications, the pipes must pipe-soil system responses. Watkins and Reeve 共1982兲 performed
support soil overburden, groundwater, and loads applied at the field tests on corrugated polyethylene pipes, with pipe diameters
ground surface due to vehicular traffic. The flexible pipe design ranging from 381 to 610 mm 共15 to 24 in.兲. The test pipes were
and installation are currently based on American Association of subjected to the concentrated wheel load, and it was found that a
State Highway and Transportation Officials 共AASHTO兲 require- soil cover of 305 mm 共1 ft兲 over the pipe crown appeared to
ments and Department of Transportation 共DOT兲 design practice. provide adequate protection against an excessive pipe deflection.
The AASHTO designs of buried thermoplastic and metal pipes Lohnes et al. 共1997兲 conducted the field test on the HDPE pipes
are generally based on the factored thrust and pipe wall resis- with a soil cover of 610 mm 共2 ft兲. It was observed that the load-
tance. The pipe wall resistance must be, indeed, greater than the ing plate with a contact area of 305 by 305 mm 共12 by 12 in.兲
factored thrust to ensure safety against structural failure. Vertical punched into the soil as the bearing capacity of the soil backfill
soil pressure at the pipe crown level and pipe outside diameter are exceeded. This led to a stress increase at the pipe crown and
the two important factors for determination of the factored thrust. eventual failure of the pipe with a reversal of curvature 共local wall
When the flexible pipes are buried under shallow depths, the ver- bending兲. Most pipe failures occurred at vertical pipe deflections
tical crown pressure is mainly influenced by the live load effect. between 1.9 and 2.9% 共Conard et al. 1998兲.
Soil backfill quality, pipe geometrical and material properties, Published literature indicates that there is very limited infor-
pipe installation condition, and loading configuration are also mation on the pipe-soil system behavior with a soil cover of less
than 610 mm 共2 ft兲 and without a pavement structure, especially
1
Professor and Director, Center for Infrastructure and Constructed for pipes with a nominal diameter equal to or greater than
Facilities, Dept. of Civil Engineering, Florida Atlantic Univ., Boca Raton, 900 mm 共36 in.兲. Therefore, full-scale field tests were carried out
FL 33431. at the Florida DOT 共FDOT兲 maintenance yard on Springhill Road
2
Professor and Director, Dev. & Research for Str. and Rehab. in Tallahassee during December 2001 to May 2002 共Arockiasamy
共DRSR兲, Dept. of Construction Engineering, Univ. of Quebec/Ecole de et al. 2004兲. The field test was conducted to investigate the
Technologie Superieure, Montreal PQ, Canada H3C 1K3. behavior of buried thermoplastic and metal flexible pipes under
3
Graduate Student, Dept. of Mechanical Engineering, Florida Atlantic the application of static concentrated wheel loads including a
Univ., Boca Raton, FL 33431. dynamic load allowance factor. The main objective was to evalu-
Note. Discussion open until July 1, 2006. Separate discussions must ate the short-term responses of the pipe-soil system and the pipe
be submitted for individual papers. To extend the closing date by one
joint performance under different soil cover depths without the
month, a written request must be filed with the ASCE Managing Editor.
The manuscript for this paper was submitted for review and possible pavement, which is considered as an extreme case during the
publication on October 6, 2004; approved on March 1, 2005. This paper construction phase.
is part of the Journal of Performance of Constructed Facilities, Vol. 20, In this paper, the pipe installation in the field and the test
No. 1, February 1, 2006. ©ASCE, ISSN 0887-3828/2006/1-21–27/ procedure are first described. Then, the pipe-soil system responses
$25.00. are presented and discussed with emphasis on pipe deflections,

JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES © ASCE / FEBRUARY 2006 / 21

Table 1. Pipe Geometric Properties
Designation PE36a PE36b PE48 PVC36 Al36 St36
Nominal pipe diameter 共mm兲 900 900 1,200 900 900 900
Average inside pipe diameter 共mm兲 914 911 1,208 902 907 910
Average outside pipe diameter 共mm兲 1,059 1,054 1,339 985 947 922
Minimum pipe wall thicknessa 共mm兲 3.18 3.07 3.43 4.91 1.91 2.01
Cross-sectional area per unit length 共mm2兲 10.19 9.53 14.76 10.44 1.19 1.50
Moment of inertia per unit length 共mm4兲 6,555 5,834 10,094 2,393 55 61
a
For the thermoplastic pipes, the minimum thickness denotes the inner liner thickness at the interior pipe wall.

deformation of cross-sectional pipe profile, and soil pressures Burial Depths, Trench Width, Bedding, and Backfill
around and above the pipes. Recommendations are made for the Materials
vertical pipe deflection limit and the minimum soil cover depth
A total of thirty-six 共36兲 pipes were buried and tested under three
based on the field test results and published literature.
different burial depths: 0.5D, 1D, and 2D. These three depths
correspond to 457, 914, and 1,829 mm 共1.5, 3, and 6 ft兲, and 610,
1,219, and 2,438 mm 共2, 4, and 8 ft兲, for the 900- and
Field Installation of Flexible Pipes 1,200-mm-diameter pipes, respectively. A minimum trench width
was maintained so as not to be less than the greater of either 1.5
Description of Flexible Pipes and Instrumentation times the pipe outside diameter plus 305 mm 共12 in.兲 or the pipe
outside diameter plus 406 mm 共16 in.兲 共AASHTO 1998兲. The
Six flexible pipe types, investigated in this field-test program, are
bottom of the trench had a minimum of a 152-mm 共6-in.兲 thick
6.1 m 共20 ft兲 long, and have two different nominal pipe diam-
bedding layer consisting of the 19 mm 共 4 in.兲 crushed limestone
3
eters: 900 and 1,200 mm 共36 and 48 in.兲. The test pipes included
overlaying the undisturbed natural soil. The soil backfill was clas-
high-density polyethylene pipes 共PE36a, PE36b, and PE48兲, poly-
sified as poorly graded sand with silt 共“SP-SM”兲 in accordance
vinyl chloride pipe 共PVC36兲, aluminum pipe 共Al36兲, and steel
with ASTM Standard D2487 共ASTM 1995兲, and the native soil
pipe 共St36兲. The metal pipe profile consists of the spiral-rib type
has a similar soil property. The soil backfill was placed surround-
with 19⫻ 19⫻ 191 mm 共3 / 4 ⫻ 3 / 4 ⫻ 7 − 1 / 2 in.兲 corrugation. The
ing or above the buried pipes in lifts of 152 or 305 mm 共6 or
thermoplastic pipe wall profiles have been presented elsewhere
12 in.兲, and compacted before adding the subsequent lifts. A tight
共Chaallal et al. 2004兲. The pipe geometries and material proper-
control of the backfill compaction was also made to ensure that a
ties are shown in Tables 1 and 2, respectively.
minimum dry density equivalent to 95% of the Standard Proctor
During the pipe installation, linear variable differential trans-
maximum dry density was achieved.
ducers 共LVDTs兲 were used to measure vertical pipe deflections. A
special device, designed and built at the FDOT Structural
Research Center, Tallahassee, was used for measuring cross-
sectional pipe profiles during live load application at different Field Tests of Flexible Pipes
locations along the pipe length. The earth pressure cells were used
to monitor the soil pressure during live load application. These Applied Axle Loads with Impact Factors
pressure cells were located approximately 3,048 mm 共10 ft兲 along
the pipe length from one end. For the 900-mm diameter pipes, the A 142-kN 共32-kip兲 axle load was used to simulate the standard
pressure cells were positioned at locations approximately 25 mm design highway loading in this study. Impact load factors were
共1 in.兲 away from the pipe–soil interface, and above the pipes at included in the simulation using the dynamic load allowance 共IM兲
varying soil depths, as shown in Fig. 1. The positions of the given below 共LRFD AASHTO specifications 1998兲:
pressure cells around the 1,200-mm pipes were very similar to
those of the 900-mm diameter pipes with the exception that the IM = 33 · 共1.0 − 0.125DE兲 共1兲
pressure cells placed above the pipe crown were at different
depths 共Arockiasamy et al. 2004兲. To ensure an accurate reading, where DE = depth of soil cover above the pipe crown 共ft兲. Based
the pressure cells were calibrated in the FDOT laboratory before on computed impact factors, a maximum axle load of 154 kN
installation in the field. A data acquisition system was used to 共34.6 kips兲 was required for 2D burial depth, and 72 and 181 kN
record the measured soil pressures. Strain gauges were also in- 共38.6 and 40.6 kips兲 for 1D and 0.5D burial depths, respectively,
strumented in the interior surface of the buried pipes in the pipe for the 900-mm diameter pipes, For the 1,200-mm diameter
crown, shoulder, springline, haunch, and invert regions. pipes, axle loads of 142, 166, and 177 kN 共32, 37.3, and

Table 2. Pipe Properties and Idealized Pipe Wall Thickness for Three-Dimensional Finite Element Modeling
Designation PE36a PE36b PE48 PVC36 Al36 St36
Modulus of elasticity 共MPa兲 760 760 760 2,760 68,950 199,960
Poisson’s ratio 0.40 0.40 0.40 0.37 0.33 0.30
Idealized pipe wall thickness 共mm兲 48.26 36.83 53.34 26.67 5.08 7.11

22 / JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES © ASCE / FEBRUARY 2006

Fig. 1. Location of pressure cells for 36-in. diameter pipes with 共a兲
0.5D; 共b兲 1D; and 共c兲 2D burial depths

39.9 kips兲 were required for the pipes with 0.5D, 1D, and 0.5D
burial depths, respectively.
Fig. 3. Two FDOT trucks in position over the test pipe with 0.5
burial depth
Critical Load Case Determination and Test Vehicles
The positions of test vehicles along the pipe length were deter- Test Vehicle Position and Data Recording
mined to be one of the critical parameters for evaluation of the Fig. 2共b兲 shows the tire positions of the two rear axles of the
performance of the pipes and the joints. Three different load cases FDOT truck for pipes with 0.5D burial depth. The rear axle of
共I, II, and III兲 were initially proposed and evaluated in the study each truck was exactly over the center of the buried pipe. Fig. 3
关Fig. 2共a兲兴. In the load case II, a second vehicle was added to shows a typical test setup for the pipes with 0.5D and 1D burial
simulate the two-lane traffic. Boussinesq 共1883兲 introduced the depths. The tire positions for 1D burial depth are also similar to
solution for the stresses produced at any point in a homogeneous, that of the 0.5D burial depth, except that the location of the first
elastic, and isotropic medium as the result of a point load applied FDOT truck from the pipe end was 762 mm instead of 556 mm.
on the surface of an infinitely large half space. Using the Bouss- In the case of 2D burial depth, the two rear axles of each tandem
inesq theory, the critical load case was determined based on the dump truck were positioned symmetrically with respect to the
computed stresses at the pipe crown level. It was found that the buried pipes.
load case II was the critical one 共the measured soil pressures also The pipe deflections, soil pressures, and pipe wall strains at the
confirmed this finding兲, and used for all test pipes. instrumented pipe sections were monitored and recorded 共1兲 prior
Two different trucks, the tandem dump truck and the FDOT to, 共2兲 immediately upon load application, and 共3兲 immediately
truck, were used for the required load simulation. These trucks after load removal. Additional measurements of pipe wall strains
were loaded with concrete blocks based on the required axle loads were also recorded after 15– 30 min of load application.
for different test pipes and burial depths. The tandem dump truck
was chosen for the pipes with 2D burial depth due to the ease of
maneuverability over the narrow and limited space in the field. Analytical Investigations
Using the Boussinesq theory, it was determined that the tandem
dump trucks only produced 70–80% of the required pressures at
Finite Element Analysis for Pipe-Soil System
the pipe crown for the test pipes with 1D and 0.5D burial depths
共Arockiasamy et al. 2004兲. Therefore the FDOT trucks were Numerical simulations based on finite element 共FE兲 method were
selected to simulate the required truck loads. carried out to study the pipe-soil system behavior under the

Fig. 2. 共a兲 Three different load cases. 共b兲 Tire dimension and position of the first FDOT truck for pipes with 0.5D burial depth

JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES © ASCE / FEBRUARY 2006 / 23

 Fig. 5. Vertical pipe deflection during the pipe installation

methods: the modified Iowa formula 关Eq. 共2兲兴 by Watkins and

Spangler 共1958兲 and the modified Meyerhof’s formula 关Eq. 共3兲兴
by Sargand et al. 共1998兲.
DLKWcr3
⌬x = 共2兲
Fig. 4. Three-dimensional FE modeling of the pipe-soil system: EI + 0.061E⬘r3
solid modeling for 2D burial depth 共a兲 and for 1D and 0.5D 共b兲, where ⌬x = horizontal deflection; DL = deflection lag factor;
boundary condition for 2D burial depth 共c兲, and for 1D and 0.5D K = bedding constant; E⬘ = modulus of soil reaction 共Hartley and
burial depth 共d兲 Duncan 1987兲; Wc = vertical soil pressure⫻ o.d.; o.d.= pipe out-
side diameter; r = pipe mean radius; EI = stiffness factor
共PS · 0.1493· r3兲; and PS = pipe stiffness.
application of truck loading. Both two- and three-dimensional FE
analyses were performed based on modeling the pipe–soil inter- ⌬x 33.75␴␯ 关1 + 2共H/D兲兴2
共%兲 = 共3兲
face conditions as fully bonded. The pipes with 2D burial depth D E⬘ 关1 + H/D兴共H/D兲
were modeled using a two-dimensional FE code, CANDE-89 pro-
where H = average soil cover over the pipe= H0 + 0.215⫻ r;
gram 共Musser 1989兲. The CANDE-89 modeling is based on a
H0 = height of soil cover over the pipe crown; ␴v = vertical stress
two-dimensional plane strain formulation, in which the loaded
at depth H; and D = nominal pipe diameter.
footprint length is assumed to be infinite. The finite footprints,
caused by the required wheel loads, were idealized as a single
strip load with an infinite length using the approach suggested by
Katona 共1990兲. Gravel and sandy silt with 95% compaction were Field Test Results and Comparisons
selected for the bedding and in situ soil materials, and sandy silt
was used for both the trench fill and the backfill with the same Vertical Pipe Deflections during Installation
compaction level.
The vertical pipe deflections during installation were measured
A general-purpose three-dimensional FE program, ANSYS
and recorded only for the metal pipes and the HDPE
共Version 5.7, 2000兲, was used to model the pipes with all three
1,200-mm-diameter pipe with 2D burial depth. Fig. 5 shows the
共2D, 1D, and 0.5D兲 burial depths. Three assumptions made in the
measured pipe deflection versus depth of soil backfill. It can be
three-dimensional modeling are small displacement theory, time-
seen that the values undergo an increase during the placement and
independent response, and linearly isotropic elastic material. For
compaction of the soil backfill 共from the pipe invert level to the
2D burial depth, the pipe-soil system was modeled taking advan-
pipe crown兲, then exhibit a slight decrease after the soil backfill
tage of the symmetry of axle loads 关Fig. 4共a兲兴. In the case of 1D
was placed above the pipe crown, and have no significant change
and 0.5D burial depths, the buried pipe was subjected to unsym-
until the end of the installation. These observations were similar
metrical axle loads. Therefore the system was idealized consider-
to those reported by Fleming et al. 共1997兲 where an increase
ing the full pipe together with the soil 关Fig. 4共b兲兴. All three trans-
in the pipe diameter along the vertical axis was observed in the
lational degrees of freedom were fixed at the bottom boundary,
heavily compacted sand installation. The maximum pipe
and only the horizontal translational degrees of freedom were
restrained along the two lateral boundaries 关Figs. 4共c and d兲兴. The
pipe material properties and idealized pipe wall thicknesses used
for the models are shown in Table 2. For the soil modeling, the
Poisson’s ratio was assumed to be 0.38, and the soil moduli for
pipes with 0.5D, 1D, and 2D burial depths were taken as 19.6,
31.4, and 43.2 MPa 共2,845, 4,552, and 6,259 psi兲, respectively.
The detailed procedures for obtaining the idealized thickness and
soil moduli were discussed elsewhere 共Arockiasamy et al. 2004兲.

Pipe Deflection
In addition to the results from FE analyses, the maximum pipe
deflection was also computed using two different semi-empirical Fig. 6. Comparison of pipe deflections for the PE36a pipe

24 / JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES © ASCE / FEBRUARY 2006

 the data obtained prior to load application, and during load appli-
cation with and without 5-time magnification. A pipe profile com-
puted from the three-dimensional FE analysis also indicates a
similar pattern. It can be seen that the pipes generally tend to
deform to the “heart shape” defined by Rogers 共1988兲, in which
the pipe crown flattens and the shoulders become relatively more
curved. This behavior confirmed that the buried pipes were em-
bedded in a well-compacted quality soil.

Soil Pressures
Table 3 shows the measured and predicted soil pressures around
and above the pipes with 0.5D burial depth. It is readily seen that
Fig. 7. Measured cross-sectional profiles with and without the predicted soil pressures from the three-dimensional FE analy-
magnification for the PE36a pipe sis at the pipe crown and springline are generally in the same
range with the measured values. The magnitudes of soil pressures
for pipes with 1D and 2D burial depths are small when compared
deflections, recorded when the backfill was filled up to the pipe to those with 0.5D burial depth, and the data can be found else-
crown level, were approximately 0.2, 1.1, and 0.5% for the steel, where 共Arockiasamy et al. 2004兲.
aluminum, and HDPE 48-in. diameter pipes, respectively. Thus
the deflection design limit of 5% specified in AASHTO 共1998兲 is
Visual Observations of Pipe Joint Opening and Pipe
found to be adequate for flexible pipes during the installation
Distress
under shallow burial applications.
Observations of the performance of pipe joints were made only
for the pipes with 0.5D burial depth via visual examination. Due
Pipe Deflections during Live Load Application
to the safety concern, the pipe joints were not inspected during
The vertical pipe deflection due to the effect of live load applica- live load application; therefore the joints were photographed at
tion was found to have a maximum value of 0.6% 共5 mm兲 in the one of the pipe ends using a zoom camera to observe joint dislo-
PE36a pipe with 0.5D burial depth. This value compared reason- cation, if any, and inspected for joint opening immediately after
ably with the finding in the field test on HDPE pipes reported by removal of the truck loading. Neither visible pipe joint openings
McGrath et al. 共2002兲. Fig. 6 shows comparisons of the measured were observed in this study nor did the test pipes exhibit any
horizontal deflection 共⌬x兲 and vertical pipe deflection 共⌬y兲 with visible localized bulging, wall buckling, wall crushing, cracking,
the predicted values based on the modified Iowa formula, the or tearing of the pipe.
modified Meyerhof formula, as well as CANDE and ANSYS
analyses for the PE36a pipe. It can be seen that the analytical
value based on the three-dimensional FEM prediction 共ANSYS Discussions
analysis兲 compares reasonably with the measured value. The
other methods tend to overestimate to a considerable extent. In
Pipe Deflection Limit for HDPE Pipes
the case of pipes with 1D and 2D burial depths, the measured and
predicted values were small 共less than 3 mm兲. Under shallow burial conditions and static loading configuration,
pipe deflection limits have been reported by Iowa State University
共Klaiber et al. 1996; Lohnes et al. 1997; Conard et al. 1998;
Cross-Sectional Pipe Profile
Phares et al. 1998兲, Texas Tech University 共Jayawickrama et al.
No significant deformation was observed for all the test pipes 2002兲, Ohio University 共Sargand et al. 1998兲, and Florida Atlantic
except that the two HDPE 900-mm-diameter pipes with 0.5D University 共Reddy 1999兲. The deflections at pipe failure were
burial depth experienced a small deformation at the pipe crown found to be varying from 2 to 8%. These published data indicate
region. Fig. 7 shows different measured cross-sectional pipe pro- that vertical pipe deflection limits are dependent to a large extent
files for the PE36a pipe. These pipe profiles are plotted based on on the pipe diameter and the position of the loaded contact area.

Table 3. Comparisons of Measured Soil Pressures with Predicted Values 共FE Analysis兲 for Pipes with 0.5D Burial Depth
Soil pressure 共kPa兲
Pressure PE36a PE36b PE48 PVC36 Al36 St36
cell
positions FEM Field FEM Field FEM Field FEM Field FEM Field FEM Field
P1 8 21 8 14 6 27 9 14 11 29 11 59
P2 26 38 25 17 22 N/A 27 30 27 25 26 26
P3 52 51 50 58 39 21 53 73 50 76 48 70
P4 40 39 41 66 37 N/A 37 50 37 44 36 45
P5 104 125 83 106 66 46 94 90 86 113 113 72
P6 509 101 508 438 211 46 502 492 503 166 505 222
Note: N/A: data not available; FEM: finite element method.

JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES © ASCE / FEBRUARY 2006 / 25

Table 4. Changes in Vertical Pipe Diameter Due to Repeated Loading
Effect 共adapted from Faragher et al. 2000兲
Vertical pipe diameter change
共%兲
Soil
type 50 load passes 1,000 load passes Ratioa
Gravel 0.4 1.1 2.8
Sand 0.75 2.4 3.2
a
Ratio of the pipe diameter changes at the 50th load pass to those at
1,000th load pass.

Fig. 8. Normalized measured soil pressures for 共a兲 HDPE pipes and
Faragher et al. 共2000兲 conducted field tests on buried plastic 共b兲 PVC and metal pipes
pipes with diameters of 610– 1,067 mm 共24– 42 in.兲. These pipes
were buried under a soil cover of 914 mm 共3 ft兲 without a pave-
ment structure and tested with 1,000 passes of a heavy test
stiffer—PVC and metal—pipes appeared to distribute concen-
vehicle to study the effect of the repeated loading on vertical pipe
trated wheel loading at the soil surface to pipes’ various regions
deflections. They reported that the pipe deflections increased rap-
better when compared to the HDPE pipes.
idly during the initial loading cycles, and the rate of deformation,
It is also interesting to point out that the pressures at the
thereafter, reduced considerably with increasing number of cycles
haunch region 共at 135°兲 for all of the pipes except the PE48 pipe
of loading. In this study the backfill soil was normally compacted
are only approximately 15–35% of those at the pipe crown. An
共Faragher et al. 2000兲. In order to account for higher soil com-
examination of the measured soil pressures at the pipe invert 共at
paction, the vertical pipe deflection corresponding to the 50th
180°兲 also shows that the pressures are generally in the same
load pass may, then, be assumed as the value of the initial deflec-
range as those at the haunch region 共Fig. 8兲. Therefore one can
tion. The effect of the repeated loads on HDPE pipes may be
infer that the test pipes received good soil support throughout the
estimated using the deflections at the 1,000th load pass. Table 4
lower portion of the pipes.
shows that the vertical deflections at the 1,000th load pass are
approximately three 共3兲 times those at the 50th load pass
共Faragher et al. 2000兲.
The maximum vertical pipe deflection can be estimated to be Summary and Conclusions
1.8% based on a magnification of three 共3兲 over the observed
The present study investigates the pipe-soil system responses
maximum deflection of 0.6% in the present study. Therefore it is
under live load application including dynamic impact load effects.
suggested that the maximum deflection be limited to 2% for
One of the important factors for the current design method of
HDPE pipes with shallow burial conditions. In the present study,
buried thermoplastic and metal pipes is the vertical soil pressure
a soil cover of 457 mm 共1.5 ft兲 was used for the
at the pipe crown. Field test results show that when an additional
900-mm-diameter HDPE pipes; however, further studies may be
soil cover of 152 mm 共6 in.兲 is introduced, the vertical soil pres-
necessary to examine the repeated loading effect.
sure, at the pipe crown, of the larger HDPE pipe reduces 2 to 3
times compared to those of the smaller HDPE pipes. The other
Soil Pressures two important factors are the vertical pipe deflections during in-
The computed soil surface pressures were 648 and 634 kPa stallation and live load application. The AASHTO specified de-
共94 and 92 psi兲 for the 900- and 1,200-mm-diameter pipes, flection limit of 5% is found to be adequate for the flexible pipes
respectively. These values were calculated based on the tire-truck during the installation phase. A vertical deflection limit of 2% is
loads of 44.4 and 45.1 kN 共10,150 and 9,975 lbs兲 and the tire- suggested for the HDPE pipes during the construction phase for
contact area of 229 by 305 mm 共9 by 12 in.兲. The measured soil road way and highway applications. Although a soil cover of
pressures at the pipe crown are 19, 16, 14, 11, and 18% of the 457 mm 共1.5 ft兲 was used for the 900-mm-diameter HDPE pipes
computed soil pressures for the PE36a, PE36b, PVC36, St36, and in the present study, further studies may be necessary to examine
Al36 pipes, respectively. Moreover, in the case of the PE48 pipe, the repeated loading effect.
the measured pipe crown pressure was only 7%. It is readily seen
that the measured soil pressure for the HDPE 1,200-mm-diameter
pipe is approximately 2 to 3 times less than those of the HDPE Acknowledgments
900-mm-diameter pipes. This indicates that an additional soil
cover of 152 mm 共6 in.兲 significantly contributes to lower the soil The writers wish to express sincere thanks to Florida Department
pressure at the pipe crown. of Transportation 共FDOT兲 for the financial support of the study
Fig. 8 shows the measured soil pressures normalized by the presented in this paper 共Research Project: Experimental and Ana-
values at the pipe crown 共at 0°兲 for pipes with 0.5D burial depth. lytical Evaluation of Flexible Pipes for Culverts and Storm
It can be seen that the pressures at pipe springline 共at 90°兲 are Sewers, Contract No. BC-775, Principal investigator: Dr M.
only 40–55% of those at pipe crown for the HDPE pipes, and that Arockiasamy, Project Manager: Marc Ansley兲. The writers wish
the values are about 70, 80, and 98%, for the aluminum, PVC, to express their appreciation to Dr. P. Scarlatos, Professor and
and steel pipes, respectively. This suggests that for pipes with the Interim Chairman, Department of Civil Engineering, and Dr. Karl
same diameter and soil cover depth, the HDPE pipes attracted K. Stevens, Dean, College of Engineering, Florida Atlantic
more load at the pipe crown region than other three pipes; the University for their continued interest and encouragement.

26 / JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES © ASCE / FEBRUARY 2006

References pipe under vehicle loading.” Transportation Research Record 1288,
Transportation Research Board, National Research Board, Washing-
American Association of State Highway and Transportation Officials ton, D.C., 127–135.
共AASHTO兲. 共1998兲. AASHTO LRFD bridge design specifications, Klaiber, F. W., Lohnes, R. A., Wipt, T. J., and Phares, B. M. 共1996兲.
Washington, D.C. Investigation of high density polyethylene pipe for highway applica-
tions, Final Rep.: Phase I, Iowa DOT Project HR-373, Engineering
Arockiasamy, M., Chaallal, O., Limpeteeprakarn, T., and Wang, T.
Research Institute, Iowa State University, Ames, Iowa.
共2004兲. Experimental and analytical evaluation of flexible pipes for Lohnes, R. A., Wipt, T. J., Klaiber, F. W., Conard, B. E., and Ng, K. W.
culverts and storm sewers, Vol. III: Field experimental work and nu- 共1997兲. Investigation of high density polyethylene pipe for highway
merical analysis, Contract No. BC-775, Florida Department of applications, Final Rep.: Phase II, Iowa DOT Project HR-373A, Iowa
Transportation. State University, Ames, Iowa.
ASTM. 共1995兲. “Classification of soils for engineering purposes: Unified McGrath, T. J., DelloRusso, S. J., and Boynton, J. 共2002兲. “Performance
soil classification system.” D 2487-93, Annual Book of ASTM of thermoplastic culvert pipe under highway vehicle loading.” 81st
Standards, ASTM, Philadelphia. TRB Annual Meeting, Transportation Research Board, Washington,
Boussinesq, J. 共1883兲. Application des potentials a l’etude de l’equilibre D.C.
et du mouvement des solides elastiques, Gauthier-Villars, Paris. Musser, S. C. 共1989兲. CANDE-89 user manual, Publication No. FHWA-
Chaallal, O., Arockiasamy, M., and Limpeteeprakarn, T. 共2004兲. “Labo- RD-89-169, Federal Highway Administration, U.S. Department of
ratory testing for determination of stiffness characteristics of large Transportation, Washington, D.C.
diameter thermoplastic gravity-flow drainage pipes.” 83rd TRB Phares, B. M., Wipf, T. J., Klaiber, F. W., and Lohnes, R. A. 共1998兲.
Annual Meeting, Transportation Research Board, Washington, D.C. “Behaviour of high-density polyethylene pipe with shallow cover.”
Transportation Research Record 1624, Transportation Research
Conard, B. E., Lohnes, R. A., Klaiber, F. W., and Wipf, T. J. 共1998兲.
Board, National Research Council, Washington, D.C., 214–224.
“Boundary effects on response of polyethylene pipe under simulated
Reddy, D. V. 共1999兲. “Evaluation of plastic piping for culverts and storm
live load.” Transportation Research Record 1624, Transportation Re-
sewers.” Final Rep., FDOT Work Program Number: 0510757, Depart-
search Board, National Research Board, Washington, D.C., 196–205.
ment of Ocean Engineering, Florida Atlantic University.
Faragher, E., Fleming, P. R., and Rogers, C. D. F. 共2000兲. “Analysis of Rogers, C. D. F. 共1988兲. “Some observations on flexible pipe response to
repeated-load field testing of buried plastic pipes.” J. Transp. Eng., load.” Transportation Research Record 1191, Transportation Research
126共3兲, 271–277. Board, National Research Board, Washington, D.C., 1–11.
Fleming, P. R., Faragher, E., and Rogers, C. D. F. 共1997兲. “Laboratory Sargand, S. M., Hazen, G. A., and Masada, T. 共1998兲. “Structural evalu-
and field testing of large-diameter plastic pipe.” Transportation ation and performance of plastic pipe—Volume I.” Rep. No. FHWA/
Research Record 1594, Transportation Research Board, National OH-98/011, Final Report to the Ohio Department of Transportation
Research Board, Washington, D.C., 208–216. and the Federal Highway Administration.
Hartley, J. D., and Duncan, J. M. 共1987兲. “E⬘ and its variation with Watkins, R. K., and Reeve, R. C. 共1982兲. Effect of heavy loads on buried
depth.” J. Transp. Eng., 113共5兲, 538–553. corrugated polyethylene pipe, Advanced Drainage Systems, Inc.,
Jayawickrama, P. W., Amarasiri, A. L., and Regino, P. E. 共2002兲. Mini- Columbus, Ohio.
mum cover requirements for large diameter HDPE pipe installed with Watkins, R. K., and Spangler, M. G. 共1958兲. “Some characteristics of
granular backfill. 81st TRB Annual Meeting, Transportation Research modulus of passive resistance of soil: A study of similitude.” Proc.,
Board, Washington, D.C. 37th Annual Meeting, Vol. 37, Highway Research Board, Washington,
Katona, M. G. 共1990兲. “Minimum cover heights for corrugated plastic D.C., 576–583.

JOURNAL OF PERFORMANCE OF CONSTRUCTED FACILITIES © ASCE / FEBRUARY 2006 / 27