Tunnelling and Underground Space Technology 149 (2024) 105813

Contents lists available at ScienceDirect

Tunnelling and Underground Space Technology

incorporating Trenchless Technology Research
journal homepage: www.elsevier.com/locate/tust

Field buried and laboratory investigation of Full-Scale eco-friendly

Spun-Cast concrete pipes under various construction loading regimes
Adeel Faisal a, b, *, Safeer Abbas a, Ammad Hassan Khan a, Imtisal Ahmed a, Sbahat Shaukat c
a
Faculty of Civil Engineering, University of Engineering and Technology, Lahore, Pakistan
b
Civil Engineering Department, University of Engineering and Technology, Lahore (Narowal Campus), Pakistan
c
Punjab Intermediate Cities Improvement Investment Program, Lahore, Pakistan

A R T I C L E I N F O A B S T R A C T

Keywords: Reinforced concrete pipes are widely used under embankments for drainage purposes. Current design and
Buried pipes construction guidelines for the reinforced concrete pipes are known to be overly conservative, putting reinforced
Vehicular load concrete pipes at a disadvantage against flexible pipes. This study was designed to assess the structural per­
Construction equipment
formance of full-scale buried precast concrete pipes with shallow soil cover under the application of vehicular
Three-edge test
Patch load
load representing commonly used construction equipment and daily routine vehicles. The concrete pipes
incorporating fly ash were reinforced with double cage reinforcement with varying amounts of steel. Various
types of vehicles including dump trucks, plain roller, asphalt-filled truck, and transit mixer were passed
perpendicular to the longitudinal axis of the buried concrete pipe and the vertical displacement at the pipe’s
crown was monitored. Maximum deflection of 0.26 mm was observed due to the passage of dump truck over the
buried pipe. No cracking or sign of distress was observed in the buried pipe under the action of tested vehicular
loads. Therefore, pipes were transported to the laboratory and tested under three-edge bearing line load and
patch load to mimic the application of wheel load and to investigate the cracking and ultimate capacities. The
reduction in the steel reinforcement by 75 % and 50 % resulted in the decrease in the ultimate load capacities of
the pipes by 83 % and 57 %, respectively. The 0.3 mm crack load in the patch load test was 0.82 to 0.89 times the
0.3 mm crack load in the three-edge bearing line load test. The 0.3 mm crack load in the laboratory test was
around twice the maximum load applied to the pipe in the field buried condition. This study will help designers
and contractors to better understand the field and laboratory performance of reinforced eco-friendly concrete
pipes buried in shallow soil cover. Moreover, this study acts as a benchmark for evaluating the real field pipe
capacities during the construction phase of various infrastructures development.

1. Introduction improvements such as increased wall stiffness and leak-free joints. In

order to make concrete pipes more competitive, the current design of
Reinforced concrete pipe (RCP) has been in widespread use for reinforced concrete pipes needs to be continuously reassessed.
conveyance of sewage and stormwater for several decades. However, Indirect design method, originally developed in 1930s (Spangler,
RCP has been facing a shrinking market due to the gradual introduction, 1933) is still widely practiced. The indirect design method is based on
over the last fifty years, of new lighter weight pipe materials. This has relating the strength of the pipe in the field condition to the strength in
enabled the manufacturing of longer pipe segments, leading to less the three-edge bearing test in the laboratory through factor known as
costly transportation and easier installation logistics. For instance, the bedding factor. As new knowledge became available regarding soil-
length of one concrete pipe segment is typically 2.44 m, while that of a structure interaction and concrete design procedures, the direct design
plastic pipe can be as long as 13 m. Polyvinyl chloride (PVC) pipe was method was developed (ASCE 15, 2017). The direct design method is
the first flexible pipe developed and later introduced in the market. based on ensuring the pipe capacity that would satisfy all the limit states
Several innovative forms of flexible pipes such as high-density poly­ in the buried condition. The internal forces in the wall of the pipe under
ethylene (HDPE) pipe have been introduced with continuous field conditions are calculated and the pipe wall is designed to resist the

* Corresponding author.
E-mail address: [email protected] (A. Faisal).

https://doi.org/10.1016/j.tust.2024.105813
Received 26 December 2023; Received in revised form 9 April 2024; Accepted 6 May 2024
0886-7798/© 2024 Elsevier Ltd. All rights reserved.
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

action of all the applied loads, similar to the design of other reinforced installed concrete pipe due to the passage of construction equipment. No
concrete structures. Standard installations methods from type I to type guidelines are available in the published literature on the minimum soil
IV were defined. The type I installation requires fill material consisting cover required to protect the concrete pipes from failure under the loads
of well-graded sands and gravels and high level of compaction, while from construction equipment. Installation of concrete pipes in
type IV installation can be provided using the native soil with little or no embankment condition for cross drainage work is followed by
compaction requirements. Direct design method has been shown to be a compaction of the backfill soil for preparation of subgrade for the
superior method of concrete pipe design, compared with the indirect pavement. Compaction is often achieved using plain rollers and backfill
design method (Erdogmus et al., 2010). soil is transported in dump trucks. Previous studies subjected concrete
Several studies in the past have been conducted to evaluate the two pipes to the code-specified traffic load on buried concrete pipes after
design methods. Measurement of soil stresses around concrete pipes completing the installation and compaction of the backfill soil. Scant
buried in a type IV installation over a period of 20 months showed that literature is available on the effect of passing construction equipment
the supporting lateral soil stresses at the springline level were several over the freshly buried concrete pipes with little or no backfill. Concrete
times higher than the estimated values based on the direct design pipes may be cracked while the construction equipment is used to
method (ASCE 15, 2017), which highlighted the level of conservatism in compact the backfill over the freshly installed concrete pipes. The
the design procedure (Wong et al., 2006). A series of full-scale tests on response of buried concrete pipes with shallow soil cover under the
ASTM C76 (ASTM C76, 2022) class V pipes having 600 mm inside action of load from construction equipment has been very limited re­
diameter and buried in a type I installation showed that the application ported. The main objective and the rationale of this study are to phys­
of axle load representing CL-625 design truck loads (CSA S6:19, 2019) ically investigate the buried concrete pipes having very small earth
resulted in no cracking in the pipe (Lay and Brachman, 2014). The cover subjected to the vehicular load due to various construction
maximum tensile strain in the pipe was about half the tensile cracking equipment used during the laying of pipeline infrastructure. This study
strain of the concrete, which showed that the pipe had a factor of safety adds new contribution to the existing knowledge by examining the
of 2 against cracking. This is overly conservative especially considering response of shallow-buried concrete pipes fabricated with varying
that the factor of safety is against the onset of cracking and not on the amount of steel reinforcement, under the action of vehicular loads due to
ultimate failure. MacDougall et al. (2016) conducted physical tests on common construction equipment. The pipes were first tested in the field
full-scale concrete pipes having inside diameters of 600 mm and 1200 and then transported to the laboratory for testing under conventional
mm. The pipes were tested in the buried condition and in the laboratory three-edge bearing line load and patch load to mimic the passage of
under three-edge bearing condition. The pipes were buried in embank­ wheel loads. The comparison of the performance of tested concrete pipes
ment condition in type II installation at various depths of 0.3 m, 0.6 m, under the field-buried conditions and laboratory tests (three-edge
and 0.9 m. Results showed that the pipe exhibited no cracking when bearing line and patch loads) is presented in this study. The main
subjected to the design service load as per AASHTO LRFD Bridge Design advantage of this study is that the pipes are tested under the loading
Specifications (AASHTO, 2020). Moreover, the bedding factors used for from common construction equipment that is used in the pipeline’s
the indirect design of concrete pipes were found to be overly conser­ infrastructure projects. Moreover, it should be noted that the tested
vative. Nehdi et al. (Nehdi et al., 2016) performed full-scale tests on pipes were incorporating fly ash leading towards sustainable and eco-
buried concrete pipes having inside diameter of 600 mm and fabricated friendly construction.
without steel cage. The pipes were buried at depth of 0.6 m in the type IV
installation, which is the poorest of all standard installation types. The 2. Manufacturing and methodology
pipes showed no cracking when Ontario design truck CL-625 (CSA
S6:19, 2019) was passed over the buried pipes. The fact that the pipes 2.1. Manufacturing of full-scale pipes
were unreinforced and buried in type IV installation hints at the overly
conservative nature of the current design methods. Peter et al. (2018) Full-scale concrete pipes were manufactured by spinning method in a
explored the effect of soil erosion on the bending moment in the wall of local industrial-scale facility. The concrete pipes with 750 mm inside
the buried concrete pipe. The effective bedding factors were suggested diameter and 88 mm wall (B) thickness were reinforced with various
for the concrete pipe supported by soil subjected to erosion. percentages of ASTM C76 (ASTM C76, 2022) class IV steel cages (i.e.,
Recent study (Alzabeebee et al., 2018a) conducted using numerical 100 %, 75 %, and 50 % reinforcement). The reinforcement was provided
modelling showed that the earth load bedding factors employed in the in two layers. The area of the reinforcement in the outer cage was 75 %
indirect design method by British Standard BS 9295 (BSI, 2020) were of that in the inner cage. It is worth mentioning that the ASTM C76
overly conservative up to three times the actual bedding factors. Further suggests providing the reinforcement in one layer for pipes of 750 mm
research (Li et al., 2022; Peter et al., 2018; Scheperboer et al., 2021) inside diameter with wall-B thickness to attain the strength re­
showed the vulnerability of the concrete pipes to the underground quirements of class IV. However, for this study, the reinforcement was
cavities resulting from the erosion of soil. Furthermore, it was revealed provided in two layers so that at least one reinforcement cage would be
by previous studies (Alzabeebee et al., 2019, 2017) that the design of positioned close to the tension face at all critical sections of the pipe i.e.,
concrete pipes can be economized by the use of finite element analysis invert, crown and springlines. Both the inner and outer steel cages had
for calculating the loads on the pipes instead of the traditionally prac­ the design concrete cover of 20 mm from the inner and outer surfaces of
ticed indirect design approach. The impact of dynamic loads, such as the pipe, respectively. The details of the pipes reinforcement are
those from traffic and earthquake, can cause distress in concrete pipes, if
such effects are not accounted for at the design stage (Alzabeebee, 2022,
2019; Alzabeebee et al., 2018b). Table 1
Previous studies have consistently reported that the current design Reinforcement details of full-scale precast concrete pipes.
methods of concrete pipe are overly conservative. It leads to the research Pipe Inside diameter Wall thickness As,i As,o As,total
question of whether it is possible to reduce the steel reinforcement used (mm) (mm) (mm2/ (mm2/ (mm2/
in the concrete pipes as per current practice with the objective to m) m) m)
economize the concrete pipes. Moreover, ASCE Standard Installation RC1 750 88 423 317 740
Direct Design (SIDD) (ASCE 15, 2017) requires that a minimum soil RC2 750 88 317 238 555
cover of 0.9 m be provided over the top of the installed concrete pipe if RC3 750 88 212 159 371
construction equipment has to pass over the concrete pipe. Alterna­ As,i = Area of steel in the inner cage, As,o = Area of steel in the outer cage, As,total
tively, the soil cover should be large enough to prevent damage to the = Total area of steel.

2
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

presented in Table 1. A relatively larger diameter pipe was selected for Table 2
this study because of its widespread application in drainage works Concrete mixture used for casting of the precast pipes.
associated with roadways. Ingredient Mass/cement mass
The manufacturing process consisted of fabricating the steel cage,
Cement 1.00
preparation of mold, casting, demolding, and curing of pipe. The steel Fly ash 0.11
cage was fabricated manually. The steel cage was made up of longitu­ Fine aggregate 1.46
dinal and circumferential bars (Fig. 1(a)). The circumferential bars were Coarse aggregate 3.02
prepared by bending straight bars in the form of circular rings using bar Water 0.52

bending assembly. The ends of the rings were welded together with an
overlap of 50 mm. The circular rings were then welded to the longitu­ wall thickness over the length of the pipe. The water extracted from the
dinal rings at desired spacing.The inner surface of the pipe mold was concrete as a result of spinning was drained out with the help of a brush
lubricated for ease in the demolding of the pipe (Fig. 1(b)). The steel (Fig. 1(d)). Dry cement, followed by flushing of water, was used to make
cage was inserted into the pipe mold and the mold was fitted with end the final layer of the concrete pipe so as to improve the surface finishing
rings on both sides. The end rings were connected by four tie rods to and the hydraulic performance of the pipe. When the casting was
keep the integrity of the assembly. The pipe mold with end rings was completed, the rotation of the pipe mold was stopped, and the mold
placed on two runners which were powered by a variable speed motor. assembly was removed from the runners. The end rings were removed
The rotational motion of the runners induced rotation to the pipe mold from the ends of the mold immediately after casting and the pipe was
in the direction opposite to that of the runners (Fig. 1(c)). The concrete demolded after 24 h. The curing of the pipe was done in water ponding
mixture for the casting was prepared in two drum-type mixers to feed for 28 days as per local practice.
the concrete into the pipe mold from both ends. The concrete mixture
was prepared using locally available fine and coarse aggregates. Ordi­
nary portland cement and fly ash were used as binders. The concrete 2.2. Test methodology
mixture used for the casting of the pipes is given in Table 2. The mold
was rotated slowly to begin with the casting of the pipe. The concrete The test methodologies consisted of two phases. In the first phase, the
was manually fed into the pipe mold with shovels. The concrete was concrete pipes were buried in the field and subjected to vehicular
poured until the desired wall thickness was achieved. The wall thickness loading. In the second phase, the concrete pipes were transported to the
was controlled by the end rings. laboratory and investigated. Laboratory investigation include the three-
Single-stage centrifugal method (Wen et al., 2017) was employed to edge bearing load and patch load tests.
manufacture the concrete pipe. Accordingly, the concrete feeding was
done in a single layer. The speed of rotation was increased when the 2.2.1. Field buried testing under vehicular load
concrete feeding was completed. A steel rod was used to scrub off the The first phase consisted of excavating a pit 6.4 m × 1.3 m × 1.32 m
concrete in excess of the required wall thickness and to ensure uniform (Fig. 2(a)), preparing the bedding (Fig. 2(b) to (d)), laying of the pipes
(Fig. 2(e)), backfilling, compaction of the backfill soil (Fig. 2(f)), and

Fig. 1. Manufacturing of reinforced concrete pipes (a) steel cage (b) lubricated pipe mold (c) pipe mold fitted with ends (d) brush and scrubber.

3
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

Fig. 2. Installation of concrete pipes (a) excavation (b) leveling of bed (c) compaction of bed (d) laying of crushed stone gravels (e) laying of pipes (f) installed pipes.

passing of various types of vehicles. The pipe had an outside diameter of to the longitudinal axis of the pipe. The vertical displacement of the
930 mm. The minimum horizontal clearance between the springline and crown was recorded through an analogue dial gauge (Fig. 3).
the wall of the pit should be one-sixth of the outside diameter of the pipe Each vehicle was passed three times over the pipe. The first two
(155 mm), according to (ASCE 15, 2017). Accordingly, the minimum passes were done to allow the soil to settle so that the permanent
width of the pit should be 1240 mm. In this study, the excavated pit had deformation in the soil could be eliminated from the recorded
a width of 1300 mm. Crushed stone gravels, classified as GP soil as per displacement. The vertical displacement was recorded for the third pass
Unified Soil Classification System (ASTM D2487, 2017), were used in of each vehicle. The vehicles passed over the buried pipes included the
the bedding and haunch areas of the pipe (Fig. 2(d)). The outer bedding daily routine vehicle and commonly used construction equipment/ve­
and the haunch areas were provided without compaction. Since type IV hicles (i.e., bike, car, single cabin, tractor, transit mixer, dump truck,
standard installation does not require any compaction of the bedding plain baby roller and plain roller). Fig. 4 shows various types of vehicles
and haunch areas, the installation type was classified as Type IV (ASCE passed over the buried pipe. It is important to mention that the recorded
15, 2017), which is the poorest type of installation types. Concrete pipes displacement was due solely to the passage of the vehicular live load
have a major advantage over flexible pipes in terms of installation re­ over the pipe as the dial gauge was zeroed before the passage of each
quirements. Flexible pipes, such as HDPE pipes, heavily rely on the vehicle and the earth load and the pipe self-weight was already acting on
support from the surrounding soil to resist the applied loading and the pipe before the application of the vehicular load. The authors
therefore, stringent installation requirements need to be satisfied for acknowledge the fact that moving loads would have better represented
flexible pipes. On the other hand, concrete pipes resist the applied the actual loading scenario. However, the pipe was buried under a small
loading through internal forces in the walls of the pipe and therefore, earth cover of 190 mm and the top surface of the earth fill was unpaved,
can be installed in any type of installation. This also aligned with the limiting the potential for replicating the actual speed of the vehicle. The
authors’ aim to test the concrete pipes under the most severe con­ pipe was thoroughly observed for any cracks or distress due to the
struction loadings with the least amount of support from the soil. vehicular loading. Each of the two replicate pipes was loaded indepen­
The two replicate pipes were lowered into the trench. In order to dently, one at a time. Later, the pipes were removed from the pit and
allow space for monitoring of potential cracks in the pipe, the length of transported to the laboratory for further testing.
the test pit was kept larger than the total length of the two pipes
together. The same material as used for the bedding was then placed up 2.2.2. Laboratory testing
to the springlines of the pipes. The excavated soil was then backfilled up One of the two replicate pipes in each category of pipes was tested
to ground level. The compaction of the backfill soil was achieved using under three-edge bearing condition according to ASTM C497 (ASTM
manual compactor in two layers. Each layer of the backfill soil was C497, 2020). Accordingly, the concrete pipe was placed on the two
approximately 175 mm thick. The workers sprinkled water over the lower wooden bearings measuring 50 mm × 25 mm in cross-section. The
backfill soil for ease in compaction based on their experience with the lower bearings were attached to a rigid steel section resting on the test
same type of soil. The ambient temperature at the time of compaction floor. The distance between the inner edges of the lower bearings was
varied between 19 ◦ C to 21 ◦ C. The top surface was compacted and 65 mm. A line loading uniformly distributed over the length of the pipe
leveled to allow for the passage of various types of vehicles. The top was applied to the top of the pipe through an upper bearing as shown in
surface was not paved. The top of the pipe was 190 mm below the top Fig. 5 and illustrated in Fig. 6.The upper bearing consisted of a rigid steel
finished surface. This was done to maximize the effect of vehicular live section, a wooden section and a 25 mm thick rubber pad bearing against
loading on the pipe. The buried concrete pipes were subjected to various the top of the pipe. The load was recorded by a load cell placed between
types of vehicular loads. Each type of vehicle was passed perpendicular the hydraulic jack and the rigid steel section. Two linear variable

4
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

Fig. 3. Set up for testing of buried concrete pipe (a) experimental representation (b) schematic representation.

Fig. 4. Various types of vehicles passed over the buried pipes (a) car (b) single cabin (c) tractor (d) transit mixer (e) dump truck (f) roller.

displacement transducers (LVDTs) were used to measure the change in through the LVDTs positioned against the pipe at the springline level.
vertical diameter. One LVDT was positioned against the invert and the The data from the load cell and the LVDTs were logged to a digital data
other against the crown of the pipe. The LVDTs were attached to a steel acquisition system. The crack width was monitored using standard leaf
box section that was passing through the pipe and supported indepen­ gages.
dently at the ends. Similarly, horizontal deformation was recorded The second replicate of pipe in each category was also tested under

5
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

Fig. 5. Experimental laboratory test set up for three-edge bearing line load test on the concrete pipes (a) test set up (b) close-up of LVDTs.

Fig. 6. Schematic diagram of three-edge bearing test on the concrete pipes in the laboratory.

patch loading (Faisal et al., 2023) in the laboratory. The pipe was sup­ AASHTO truck load consideration for a dual-tire wheel (AASHTO,
ported on the same lower bearings as used in case of the three-edge 2020). It is worth mentioning that the dimensions of the loading patch
bearing test per ASTM C497 (ASTM C497, 2020). However, instead of used in this study are not to be taken as the general guidelines for
the line loading distributed over the entire length of the pipe, the pipe replicating the vehicular load on any buried pipe. Rather, the di­
was loaded at mid-length through a patch consisting of a pair of two mensions of the loading patch used in this study are to be considered
rubber bearings placed side by side, a wooden block on the rubber suitable for replicating the passing of vehicular load over a pipe having
bearings, a steel plate, and a rigid steel section on the steel plate (Fig. 7). negligible earth cover (such as 190 mm in this study). The vertical
The load from the hydraulic jack is applied to the rigid steel section, the displacement of the top of the pipe was recorded by an LVDT at 75 mm
steel plate, the wooden block, and the pair of rubber bearings, which distance from the edge of the loading patch.
transfer it to the top of the pipe. Each rubber bearing measuring 250 mm
× 500 mm in plan and 25 mm thick was placed on the pipe such that its 3. Results and discussion
500 mm side was in the circumferential direction of the pipe and 250
mm side was in the longitudinal direction of the pipe. Thus, the two 3.1. Behavior of field buried pipes under vehicular loading
rubber pads together measure 500 mm along the longitudinal direction
of the pipe and also 500 mm in the circumferential direction. The Table 3 presents the vertical displacement of the crown of the pipe
dimension of the loading patch along the longitudinal direction of the recorded under the action of various vehicular live loads. As expected,
pipe is more significant than the dimension of the loading patch in the the needle of the dial gage did not move when very small load (i.e., bike
circumferential direction of the pipe. This is because of the curved load) was passed over the pipe. Negligible deflection was observed due
surface of the pipe due to which only a portion of transverse dimension to the passage of car, single cabin, and tractor. This was because of the
of the loading patch is effective in transferring the load to the pipe and relatively small axle loads of these vehicles and the relatively large
the rest of the loading patch is “overhanging”. The dimensions of the bending stiffness of the concrete pipe. The maximum displacement of
rubber bearings were selected to mimic the loading of a truck passing 0.26 mm was observed due to the passage of the dump truck over the
over a shallow buried pipe. Furthermore, the selected patch dimensions RC3 buried pipe. No cracks or signs of distress were observed in the
match with the footprint area of 250 mm × 500 mm adopted in the buried pipes due to the passage of the various vehicles employed in this

6
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

Fig. 7. Laboratory test set up for patch loading on the concrete pipes (a) experimental (b) schematic longitudinal view (c) schematic cross-sectional view.

material.
Table 3
In order to corroborate the experimentally observed response of the
Vertical displacement recorded for the buried pipes due to various vehicles.
pipe, the loads on the pipe due to earth fill, the self-weight of the pipe
Type of vehicle Vertical displacement (mm) and the load transmitted to the pipe due to the passage of various ve­
RC1 RC2 RC3 hicles were calculated using the guidelines in ASCE 15 Standard Practice
Bike 0 0 0 for Direct Design of Buried Precast Concrete Pipe Using Standard In­
Car 0.01 0.01 0.01 stallations (SIDD) (ASCE 15, 2017). The bending moments and thrusts at
Single cabin 0.03 0.04 0.03 the critical locations i.e., invert, crown, and the springlines were
Tractor 0.02 0.02 0.03 calculated. The strains at the inner and outer sides of the wall of the pipe
Transit mixer 0.15 0.16 0.18
Dump truck 0.24 0.23 0.26
were computed at each critical location. Various parameters used for the
Plain baby roller 0.13 0.15 0.14 computation of earth load (kN/m) on the pipe and self-weight (kN/m) of
Plain roller 0.16 0.14 0.15 the pipe are presented in Table 4.
One of the important considerations in calculation of the earth load
on the pipe is whether the pipe is buried in trench condition or
study. This is in agreement with previous studies on full-scale buried
embankment condition. In an embankment condition, the soil directly
concrete pipes (Lay and Brachman, 2014; MacDougall et al., 2016;
above the concrete pipe settles less as compared to the neighboring soil
Nehdi et al., 2016). Lay and Brachman (2014) reported maximum
on the sides. The relative downward movement of the neighboring soil
displacement of 0.42 mm at the crown of their 600 mm diameter buried
results in the additional downward load on the pipe beyond the original
pipes at the load of 186.7 kN. The relatively higher displacement
prism load of the soil directly above the pipe by a factor known as the
recorded in their study was likely due to the heavy load corresponding to
vertical arching factor (VAR). In a trench condition, the friction between
fully factored AASHTO load (AASHTO, 2020) applied through a hy­
the walls of the trench and the neighboring soil helps to reduce the
draulic jack to simulate the traffic load. Nehdi et al. (2016) reported
relative downward movement between the soil directly above the pipe
maximum vertical displacement of 0.22 mm at the crown of the pipe
and the soil along the sides, thus effectively reducing the earth load on
under the load of precast concrete blocks placed over the pipe to
the pipe (ACPA Concrete Pipe Design Manual, 2011). However, as the
simulate the loading on the buried pipe in field condition. Moreover, the
width of the trench increases, the restraining effect of the wall of the
observed deflections in this study were nearly the same for the three
trench diminishes and the earth load on the pipe becomes equal to the
types of pipes reinforced with varying amount of steel. This was to be
earth load in case of embankment condition. The transition width of the
expected because the concrete pipes remained uncracked, and therefore,
trench is the width at which the trench starts behaving like an
the behavior of the pipes was governed primarily by the concrete

7
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

Table 4
Parameters for calculation of earth load and self-weight of the pipe.
Inside Wall Outside Height of Installation Unit weight of Vertical Arching Unit weight of Earth load, Self-weight of
diameter thickness diameter fill (mm) type fill (kN/m3) Factor (VAF) concrete (kN/m3) We (kN/m) pipe, Wp (kN/
(mm) (mm) (mm) m)

750 88 926 190 IV 18.8 1.45 23.5 7.32 5.56

embankment condition. The transition width of a trench depends on the weight of the pipe was computed as the product of unit weight of the
installation type, the fill height above the top of the pipe and the pipe, taken equal to 23.5 kN/m3, and the cross-sectional area of the pipe.
geotechnical properties of the soil. The pipes in this study were buried in Since the pipe was tested with no fluid inside, the fluid weight was not
a trench having width of 1.3 m. The effective friction angle of the soil considered. For the purpose of design as per AASHTO (ASCE 15, 2017),
was taken to be 13◦ based on a geotechnical study (Malik, 2015) of the HL-93 vehicular live loads have to be considered. The loads applied by a
area. The ratio of active lateral earth pressure to the vertical pressure wheel, the longitudinal and transverse spacings between the wheels and
was calculated to be 0.633 using Eq. (1) (Table 5). The product K.tanϕʹ the footprint area of the wheels are specified in the design procedure.
was 0.146. For the current study, the transition width was calculated to The footprint area of the wheel is assumed to be a rectangle a × b, where
be 1.50 m, which is nearly same as the actual width of the trench (1.30 a is the dimension of the footprint area transverse to the direction of
m). Therefore, the earth load was calculated based on the embankment travel and b is the dimension along the direction of travel of the wheel.
condition (positive projecting). This is also in agreement with the design The values of a and b are taken to be 0.50 m and 0.25 m, respectively, in
guidelines of AASHTO LRFD Bridge Design Specifications (AASHTO, the design procedure. The footprint area was assumed considering a
2020), and ASCE 15 (ASCE 15, 2017) which suggest that the earth load single dual wheel of a standard truck.
should be calculated based on embankment condition, irrespective of The wheel load applied to the surface was then distributed over the
the actual burial condition of the pipe. subsoil plane at the level of the top of the buried concrete pipe in the
The earth load, We, presented in Table 4, was calculated using Eqs. form of average pressure intensity (kN/m2). The average pressure in­
(2) and (3) (Table 5). The vertical arching factor for a type IV standard tensity was calculated by dividing the wheel load (kN) by the assumed
installation is 1.45 (ACPA Concrete Pipe Design Manual, 2011). The self- spread load area (Aspread). Depending upon the fill height over the top of
the pipe, the design may be governed by a single dual wheel, two single
dual wheels with transverse spacing of 6 ft or, two single dual wheels of
Table 5 LRFD alternate loads with transverse spacing of 6 ft. For pipes with a fill
List of equations and description of various parameters used in the equations. height less than 0.76 m, the case of single dual wheel governs, and the
Eq. Equations Description of parameters Reference spread load area is calculated using Eqs. (4) and (5) (Table 5). The
Nos. pressure intensity was converted into a line load (kN/m) at the top of the
(1) 1 − sinϕʹ K = ratio of active lateral (Sivakugan, 2021) pipe by multiplying the average pressure intensity (kN/m2) by the
K = horizontal out-to-out dimension (m) of the pipe (which is equal to the
1 + sinϕʹ earth pressure to the vertical
pressure.φ’ = effective outside diameter (m), in case of a circular pipe). Due to the longitudinal
friction angle of the soil
stiffness of the pipe and its ability to distribute load over its length, the
(2) We = VAF × PL We = earth load (kN/m) (AASHTO, 2020)
(3) PL = VAF = vertical arching line load calculated at the top of the pipe was reduced by the factor L/Le,
(
4− π
)
factor where L is the dimension of the spread load area parallel to the longi­
γs H + Do Do
8 PL = prism load (kN/m) tudinal axis of the pipe and Le is the effective length of the pipe
γs = unit weight of the soil considered to be resisting the load. The effective length Le is calculated
(kN/m3)
Do = outside diameter of the
using Eq. (6) (Table 5).
pipe In this study, various types of vehicles were passed over the buried
(4) Aspread = aspread × a = width of footprint area (ASCE 15, 2017) pipes. The vehicles represent the typical construction equipment on a
bspread of the tire (m) construction site. Thus, the vehicles used in this study are different from
(5a) aspread = a + 1.75H b = length of footprint area
the assumed vehicular live loading used for design purpose. Therefore,
(5b) bspread = b + 1.75H of the tire (m)
aspread = width of spread various parameters required for the computation of live load on the
load area (m) buried pipe in this study had to be determined for each vehicle passed
bspread = length of spread over the pipe. An important consideration regarding the computation of
load area (m) the effect of live load on buried pipes is the impact factor or dynamic
Aspread = area of the spread
load area (m2)
load allowance. AASHTO LRFD Bridge Design Specifications (AASHTO,
H = fill height over the top 2020) requires that the impact factor due to the live load should be
of the pipe (m) considered for all pipes with fill heights less than 2.44 m. However,
(6) 3 L = dimension of the spread (ACPA Concrete ASCE 15 (ASCE 15, 2017) states that impact factor is not required due to
Le = L + 1.75 Do
4 load area parallel to the Pipe Design
the “cushioning” effect of the fill over the pipe, if the fill height is equal
longitudinal axis of the pipe Manual, 2011)
Do = outside diameter of a to or greater than 300 mm. In this study, the fill height was 190 mm,
circular pipe which is less than the minimum fill height required for ignoring the
(7)
Mi =
∑
Cm,i Wi
Dm Wi = load type (Wp, We, or (ASCE 15, 2017) impact factor. However, in this study, the vehicles were passed over the
∑ 2 WL) in kN/m pipe at a speed much lower than the usual speed of passing. In fact, the
(8) Ni = Cn,i Wi Dm = mean diameter of the
vehicles were slowly progressed towards the pipe and stopped when the
pipe (m)
Mi = bending moment due wheels were directly over the pipe. This was done to allow for the
to the load type (kN-m/m) observation of vertical displacement at the crown of the pipe through an
Ni = thrust due to the load analogue dial gauge and observe the pipe for any cracking/distress.
type (kN/m)
Therefore, the impact factor was not considered in the calculation of
Cm,i and Cn,i are the moment
and thrust coefficients, loads on the pipe due to the passage of the vehicles.
respectively, for the load Various parameters required for computation of live load effect on
type. the buried pipe due to the passage of various vehicles over the pipe are

8
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

listed in Table 6. The models of the vehicles used in the study are listed Table 7
in Table 6. The characteristics of the vehicles such as weight, axle load, Coefficients for bending moments and thrusts due to various types of loads.
and width of the tires were obtained from the manufacturers’ manuals Load type Invert Crown Springlines
for the vehicles. The width of the footprint area was measured for the
Cm Cn Cm Cn Cm Cn
vehicles onsite and the length of the footprint was taken as the fraction
of the circumference of the tire that was in firm contact with the ground. Pipe weight, Wp (kN/ 0.235 0.077 0.079 − 0.077 − 0.101 0.287
m)
The bending moment and thrust in the wall of the pipe were calculated Earth load, We (kN/ 0.191 0.128 0.118 0.079 − 0.127 0.504
using the ASCE 15 (ASCE 15, 2017) procedure. Accordingly, coefficients m)
for bending moment and thrust due to various types of loads were used Live load, WL (kN/m) 0.241 − 0.017 0.267 − 0.005 − 0.172 0.492
for calculating bending moment and thrust at the critical locations in the Cm = coefficient for bending moment, Cn = coefficient for thrust.
wall of the pipe i.e., invert, crown, and springlines. The coefficients are
listed in Table 7. The bending moment at a location in the pipe due to a
given type load is calculated as the product of the bending moment Table 8
coefficient (Cm), the load value, and the mean radius of the pipe as Bending moments and thrusts at the critical locations in the buried pipe.
shown in Eqn. (7). The net bending moment at the location of the pipe is Type of Invert Crown Springlines
determined by summing the bending moments due to all load types. The vehicle
Bending Thrust Bending Thrust Bending Thrust
thrust at a particular location due to a given load is determined as a moment (kN/m) moment (kN/m) moment (kN/m)
product of the thrust coefficient (Cn) and the load value. The net thrust at (kN-m/ (kN-m/ (kN-m/
the location of the pipe is determined by summing up the thrusts due to m) m) m)
all load types as represented by Eq. (8). Bike 1.19 1.36 0.60 0.15 − 0.67 5.54
ASCE 15 (ASCE 15, 2017) provides two coefficients for the load ef­ Car 1.82 1.25 1.30 0.12 − 1.12 8.61
fects (bending moment, thrust) due to live loads, i.e., when the fill height Single 2.09 1.21 1.59 0.1 − 1.31 9.9
exceeds 1.75 Do (1.621 m) with uniformly spread live load and when the cabin
Tractor 1.63 1.28 1.08 0.13 − 0.98 7.66
fill height is 300 mm with narrowly distributed load over the top of the Transit 4.26 0.84 4.00 0.00 − 2.86 20.49
pipe. For the intermediate fill heights, the coefficient may be interpo­ mixer
lated between the two values. In the present study, the fill height was Dump 4.33 0.83 4.08 − 0.01 − 2.91 20.8
190 mm. Therefore, the loads were expected to have narrower distri­ truck
Plain 3.01 1.05 2.61 0.06 − 1.96 14.38
bution. Although no guidelines are available in the published literature
baby
for live load coefficient for the case when the fill height is less than 300 roller
mm, the coefficients for live loads in this study, have been obtained by Plain 3.56 0.96 3.22 0.03 − 2.36 17.1
extrapolating the values for the case when the fill height is 300 mm and roller
when the fill height is 1.75 Do (1.621 m). Moreover, in ASCE 15 (ASCE
15, 2017), the load on the top of the pipe (kN/m) was obtained by
buried pipe were calculated. The calculations were based on the
multiplying the average pressure intensity (kN/m2) by the outside
assumption that the pipe behaved in a linear elastic manner. The cur­
diameter of the pipe. However, in this study, the length of the spread
vature φ (1/m) of the wall of the pipe due to bending moment was
load area (bspread) or the outside diameter of the pipe, whichever is
calculated using φ = M/EI, where M is the bending moment (kN-m/m),
lesser, was used in the calculation, instead of the fixed value of outside
E is modulus of elasticity of concrete (kN/m2), and I is the moment of
diameter of the pipe. This was because the fill height in this study was
inertia of the unit length of the wall of the pipe (m4/m). The modulus of
much lesser than the normally encountered fill heights and the length of
elasticity of concrete was estimated based on the compressive strength
the spread load area at the level of the top of the pipe was lesser than the
of the companion concrete cylinders sampled from the concrete mixture
outside diameter of the pipe (0.926 m). The bending moment and thrust
used for casting of the concrete pipes The average compressive strength
due to each type of load were calculated using Eqs. (7) and (8) (Table 5),
of the concrete was 30.5 MPa. The modulus of elasticity of the concrete
respectively. √̅̅̅̅
All the calculations were done for the unit length (m) of the pipe. The was computed using E = 4700 fcʹ, where fc’ is the compressive strength
bending moments and thrusts at the critical locations in the pipe are of concrete (MPa) and E is the modulus of elasticity of concrete (MPa)
given in Table 8. The bending moment is considered positive if produces (ACI Committee 318, 2022). The EI value was calculated to be 1474 kN-
compression on the outer side of the wall of the pipe. The negative sign m2/m for the unit length (m) of the wall of the pipe.
of the bending moment at the springline represents compression on the The strain due to bending moment was calculated based on the
inner side of the wall. The bending moment at the invert was larger than assumption that the inner and outer sides of the wall of the pipe would
at the crown for all types of vehicles. The bending moment at the invert experience normal strains of same magnitude but opposite sense. The
was largest for the loading of the tested dump truck vehicle. Based on the magnitude of the strain was calculated using εm = φ.h/2, where φ (1/m)
bending moments and thrusts computed at each critical location of the is the curvature due to bending moment, h is the thickness of the wall
buried pipe, the strains at the inner and outer sides of the wall of the (m), εm is the normal strain due to bending moment. The sense of the

Table 6
Parameters for calculation of live load due to the passage of various vehicles.
Type of vehicle Model Weight of vehicle (kN) Wheel load (kN) a (m) b (m) aspread (m) bspread (m) Le (m) Live load, wL (kN/m)

Bike Honda CD-70 1.49 0.8 0.057 0.029 0.39 0.361 1.605 0.5
Car Suzuki WagonR 13.5 4.35 0.145 0.073 0.478 0.405 1.693 3
Single cabin Toyota Hilux 4x2 28.1 8 0.226 0.113 0.559 0.446 1.774 4.5
Tractor Millat 375 23 6 0.429 0.25 0.762 0.583 1.977 3
Transit mixer LPK 2518 RMC 190 40 0.509 0.253 0.843 0.583 2.058 19.4
Dump truck Howo Sinotruck 6x4 422.9 65 0.509 0.253 0.843 0.583 2.058 31.6
Plain baby roller 3411 HAMM 113.1 61 2.14 0.5 2.47 0.83 3.69 18.5
Plain roller Dynapac CA302D 127 79.4 2.13 0.5 2.46 0.83 3.68 21.6

Note: a = width of footprint area, b = length of footprint area, aspread = width of spread load area, bspread = length of spread load area, Le = Effective length of pipe.

9
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

strain was taken tensile (positive) on the inner side of the wall of the pipe during the performance of the three-edge bearing test or determined
at location of positive bending moment and vice versa. The normal later from the load–deflection curve. D-load is defined as the load per
strain due to the thrust in the wall of the pipe was calculated using εN = unit length per unit diameter of the pipe. D-load is reported in N/m/mm
N/EA, where N is the thrust in the wall of the pipe (kN), A is the cross- units. Furthermore, the ratio R0.3 of the actual Dult load to the actual D0.3
sectional area of the unit length of the wall of the pipe (m2/m), EA is the load has been calculated for each tested pipe. The ratio R0.3 depicts the
axial stiffness of the unit length (m) of the wall of the pipe (kN/m), and ability of the pipe to carry additional load after it has reached its 0.3 mm
εN is the normal strain due to the thrust in the wall of the pipe. The net crack load. The higher value of R0.3 reflects higher ductility and higher
normal strain was calculated by superimposing the normal strains due to factor of safety against collapse.
the bending moment and thrust i.e., ε = εm + εN . The values of εN were The D-loads corresponding to various stages of the loading and the
substituted with a negative sign as the thrust in the wall of the pipe R0.3 values for the tested pipes are given in Table 9. The RC1 pipe had
produces compressive (negative) strain in the wall of the pipe. The the highest D0.3 load, followed by RC2 and RC3. The RC2 and RC3 pipe
strains at the inner and outer sides of the wall of the pipe resulting from had 22 % and 37 % lower D0.3 loads, respectively, as compared to the
the passage of various vehicles are presented in Fig. 8. RC1 pipe. This showed that the decrease in the reinforcement area
The strains produced are the net result due to the earth load, the self- resulted in the decrease in the D0.3 load. This can be explained by the
weight of the pipe and the effect of live load transmitted to the pipe due fact that the steel rings play a major role in resisting the load after
to the passage of various vehicles over the pipe. Positive (tensile) strains cracking of the concrete matrix. The reduction in the area of steel
are produced on the inner side of the wall at the invert and crown and on resulted in the decrease in the D0.3 load. Like D0.3 load, the Dult loads also
the outer side of the wall at the springlines. The strain at the invert was trended down with decreasing steel area. However, the effect of the
higher than elsewhere in the pipe for all tested vehicles. The modulus of reduction in the Dult load with decreasing steel area was more pro­
√̅̅̅̅
nounced. The RC2 and RC3 pipes had 20 % and77% lower Dult loads,
rupture of concrete can be estimated using fr = 0.62 fcʹ, where fr is the
respectively, relative to the RC1 pipe. This is likely because the reduc­
modulus of rupture of concrete (MPa) and fc’ is the compressive strength tion in the steel area resulted in the higher stresses in the steel rings,
of concrete (MPa) (ACI Committee 318, 2022). The average compressive which resulted in the change in failure mode from flexure to radial
strength for the pipes in this study was 30.5 MPa. Thus, fr was estimated tensile cracking. Thus, the RC2 and RC3 pipes were unable to reach their
to be 3.42 MPa. The tensile cracking strain of concrete can be estimated full flexural capacities because of the “premature” failure due to radial
from ∊cr = fr /E, where E is the modulus of elasticity of concrete. The tensile cracking. Furthermore, the R0.3 value also decreased with the
value of ∊cr thus, computed was 132 × 10-6. reduction in steel area. This is to be expected because the reduction in
The highest tensile strain occurred at the inner side of the wall of the steel area had a more pronounced effect on the Dult load as compared to
pipe at the invert due to the passage of the dump truck with a value of the D0.3 load.
129 × 10-6 (Fig. 8(a)), which is less than the computed tensile cracking While ASTM C76 (ASTM C76, 2022) designates the pipe strength
strain of the concrete (132 × 10-6). Although this could be an explana­ classes as I to V with minimum requirements of D0.3 and Dult for each
tion for why the buried pipes did not crack under the action of the strength class. The Canadian standard CSA A257.2 (CSA A257.2, 2019)
applied vehicular live load on the surface, both the maximum tensile uses the minimum required D0.3 as the pipe designation. For instance,
strain occurring in the pipe and the tensile cracking strain of concrete ASTM C76 (ASTM C76, 2022) requires a class IV pipe to attain minimum
have been computed based on the available guidelines in the literature D0.3 and Dult of 100 N/m/mm and 150 N/m/mm, respectively. The
and the two values can only be considered as the best guess of their Canadian standard CSA A257.2 (CSA A257.2, 2019) has the same
actual values. Previous studies on buried pipes have also reported that strength requirements for a class of pipe designated as 100D (100 N/m/
no cracks appeared in the pipe under the action of design level vehicular mm is the minimum required D0.3). Thus, ASTM C76 class IV is same as
load. For instance, Nehdi et al. (2016) reported the maximum tensile CSA A257.2 class 100D. Based on the D0.3 and Dult loads recorded for the
strain of 91 × 10-6 recorded by the strain gage at the invert of the buried tested pipes, an attempt has been made to classify the pipes per ASTM
pipe under the wheel load of 68.5 kN from a standard CL 625 Ontario C76 and CSA A257.3 (Table 10). The RC1 pipe meets the Dult load
truck. The maximum tensile strain recorded was less than the cracking requirement of 175 N/m/mm for ASTM C76 class V (and CSA A257.3
strain of concrete and therefore, all the tested concrete pipes including class 140D). However, its D0.3 load falls short of the required D0.3 load
the plain concrete pipe without any reinforcement showed no cracking. (140 N/m/mm) for ASTM C76 class V (and CSA A257.3 class 140D).
It is worth mentioning that the pipes in their study were buried at a Therefore, the RC2 pipe was classified as satisfying the strength re­
depth of 0.60 m as compared to 0.19 m in the current study and had quirements of ASTM C76 class IV (and CSA A257.3 class 100D). The RC2
inside diameter of 600 mm. The presence of more fill height over the top pipe meets the requirements of ASTM C76 class IV (and CSA A257.3
of the buried concrete pipe in their study might be the reason for the class 100D) in terms of Dult load but not in terms of D0.3. Therefore, it
relatively smaller strain at the invert due to the spreading of the load could be said to satisfy the requirement of ASTM C76 class III (and CSA
over a larger area at the level of top of the concrete pipe. Moreover, A257.3 class 65D). The RC3 pipe satisfies both the D0.3 and Dult load
Fig. 8 (a) and (b) show that the strain at the neutral surface was less than requirements of ASTM C76 class III (and CSA A257.3 class 65D).
0.5 × 10-6 at invert and crown. This showed that the strain due to the According to both ASTM C76 and CSA A257.2, the safety factor
thrust in the wall was negligible. This was because of high axial stiffness depends on the class of the pipe. The safety factor is taken 1.5 for ASTM
of the wall of the pipe as compared to its flexural stiffness. Similar results C76 classes I through IV (and CSA A257.3 classes 50D through 100D)
have been obtained in the previous studies (Lay and Brachman, 2014; pipes and taken 1.25 for ASTM C76 class V (and CSA A257.3 class
MacDougall et al., 2016; Nehdi et al., 2016) on buried concrete pipes. 140D). In terms of minimum required Dult, the factor of safety (FS) may
For instance, Macdougall et al. (2016) reported a normal strain of be stated as: FS = 1.5 for Dult less than or equal to 150 N/m/mm, FS =
approximately + 2 × 10-6 (the positive sign indicates tensile strain) at 1.25 for Dult equal to or greater than 175 N/m/mm. FS is interpolated
the neutral surface of both invert and crown of pipe buried at 0.3 m linearly between 1.5 and 1.25 when Dult is between 150 N/m/mm and
depth at a simulated wheel load of 65 kN. The pipe in their studies had 175 N/m/mm. The actual Dult load of a pipe tested in TEBT may be
inside diameter of 600 mm and wall thickness of 94 mm. larger than its actual D0.3 load by a different factor. Therefore, the
equivalent class of a pipe may be taken as its actual D0.3 or Dult/FS,
3.2. Three-edge bearing line load laboratory test whichever is lesser (Ramadan et al., 2020; Younis et al., 2020). For
instance, the D0.3 and Dult of the RC1 pipe tested in this study were 94.4
The loads corresponding to the first visible hairline crack, 0.3 mm N/m/mm and 158.5 N/m/mm, respectively. The equivalent class of the
crack, the first peak load, and the ultimate load were either recorded pipe is lesser of 94.4 and 158.5/1.42. Thus, the equivalent class of the

10
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

Fig. 8. Strain at the inner and outer sides of the wall of the buried concrete pipe due to the passage of various vehicles (a) invert (b) crown (c) springlines.

11
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

Table 9 length.
D-loads corresponding to various stages of the loading in TEBT. The next significant point on the D-load versus vertical deflection
Pipe DHL(N/m/ D0.3(N/m/ Dfirst-peak(N/ Dult(N/m/ R0.3 = curve was the point at which the pipe exhibited a drop in the load,
mm) mm) m/mm) mm) Dult/D0.3 followed by regain in the load-carrying capacity. This point, marked as
RC1 78.6 115.1 125.5 193.2 1.68 Dfirst peak, marked the beginning of the extensive cracking and/or the
RC2 76.8 94.2 110.2 160.4 1.70 significant widening of the cracks already formed (Fig. 10(c) and (d)).
RC3 75.9 83.8 94.2 109.4 1.31 The drop in the load was more pronounced for the RC3 pipe as compared
DHL = Hairline crack load, D0.3 = 0.3 mm crack load, Dfirst-peak = load at first- to RC1 and RC2 pipes. All the tested pipes regained their load-carrying
peak in the load–deflection curve, Dspringlines = load at springline cracking, capacities such that the second peak load was higher than the first peak
Dult = ultimate load, R0.3 = ratio of Dult to D0.3. load for all the tested pipes. The second peak load is the ultimate load,
represented as Dult. However, the Dult load of the RC3 pipe was only 16
% higher than its Dfirst peak load. For comparison, the Dult loads of the
Table 10 RC1 and RC2 pipes were 54 % and 45 % higher than the corresponding
Classification of the tested pipes based on their actual D0.3 and Dult loads. Dfirst peak loads, respectively. The lower regain of the RC3 pipe is
Pipe ASTM CSA FS Dult/ Equivalent Governing explained by the early appearance of the radial tensile crack (Fig. 10(c))
C76 class A257.3 FS class factor and the inability to fully utilize the steel strength.
class

RC1 III 65D 1.25 155 115D D0.3 3.3. Patch load laboratory test
RC2 III 65D 1.40 115 94D D0.3
RC3 II 50D 1.50 73 73D Dult
The load corresponding to various cracking events was recorded and
presented in Table 11 for the patch load laboratory testing. P-load is the
pipe is 94D. It is worth mentioning that the equivalent class of RC1 and total load in kN applied to the pipe through the patch. The P0.3 load was
RC2 pipes was governed by D0.3 and that of the RC-3 pipe by Dult. the highest for the RC1 pipe, followed by RC2 and RC3 pipes, which
Fig. 9 shows the D-load versus vertical displacement curves for the showed the P0.3 load reduction with the decrease in the reinforcement
pipes tested in TEBT. The critical points corresponding to the appear­ area. This is to be expected as the steel reinforcement helps to keep the
ance of hairline crack, the crack width of 0.3 mm, the first peak load and crack width in check by contributing the tensile force across the crack.
the ultimate loads have been marked on the D-load versus deflection Similarly, the Pult loads followed the same trend, i.e., the Pult load de­
curves. All the tested pipes showed linear increase in the D-load with the creases with the decrease in the reinforcement area. The Pult / P0.3 ratio
vertical deflection in the initial phase of the loading, as expected. The was also high for the pipes with more reinforcement ratio. This showed
initial stiffness of all the tested pipes was similar as all the pipes were that reducing the reinforcement area significantly decreased the factor
fabricated with same wall thickness and concrete. In the next phase, of safety beyond the P0.3 load.
nonlinear behavior initiated with stiffness gradually decreasing as the The load–deflection curves for the tested pipes under patch loading
loading increased. The beginning of the nonlinear behavior was char­ are shown in Fig. 12. All the tested pipes showed four distinct phases of
acterized by the appearance of the first visible hairline crack at the loading,.i.e., the initial linear phase, the second nonlinear phase with
invert of the pipe (Fig. 10(a)). The decrease in the stiffness of the pipe in decreasing stiffness until the first peak, the third phase characterized by
this phase was the result of the widening of the already formed hairline regain in the load-carrying capacity until the ultimate load, and the final
crack at the invert and the appearance of the hairline crack at the crown post-ultimate phase. No cracks were observed in the initial linear phase
and widening thereafter. The cracks were monitored using crack of the loading, as expected. The second phase i.e., the nonlinear phase
measuring leaf gauge (Fig. 11). The stiffness of the pipe in this phase was was characterized by the gradual propagation of cracks through
highest for the RC1 pipe, followed by RC2 and RC3 pipes. The higher widening of the already formed cracks (Fig. 13(a)) and the formation of
stiffness of the RC1 pipe is possibly due to its higher resistance to new cracks, leading to reduced stiffness. The third phase was marked by
cracking and its ability to develop multiple cracking (Fig. 10(b)) as the appearance of punching shear cracks around the perimeter of the
opposed to the RC-2 and RC-3 pipes, which mostly developed single loaded patch (Fig. 13(b)) and the formation of radial cracks until the
cracks. The D0.3 load was attained in this phase of the loading when the ultimate load. The RC1 pipe was able to significantly regain its load-
crack width in the pipe reached 0.3 mm along a minimum of 300 mm carrying capacities past the first peak. The RC2 pipe also showed
similar behavior but with lower regain. However, the RC3 pipe showed
little regain in its load-carrying capacity past its first peak. This was
attributed to the difference in the failure modes of the pipes. The RC1
pipe exploited its full flexural and shear resistance with no radial cracks.
The RC2 and RC3 pipes were unable to utilize their full flexural strengths
as the highly stressed steel rings led to radial tensile failure of the pipes.

3.4. Comparison of three-edge bearing line load and patch load tests

The three-edge bearing test has been used in the industry for over a
century to assess the structural performance of concrete pipes in the
laboratory (Younis et al., 2020). The three-edge bearing test subjects the
pipe to much more severe loading regime as compared to the actual field
condition (Erdogmus et al., 2010). This is because the loads and soil
reactions on the pipe are applied in distributed manner in buried con­
dition as opposed to line load and reactions in case of three-edge bearing
test. This difference is accounted for by the application of bedding factor
to the results of three-edge bearing test. While this approach may be
reasonable for a pipe with enough burial depth to distribute the wheel
load over the length of the pipe, it may not be reasonable for a pipe with
Fig. 9. D-load vs. vertical deflection of the tested pipes in TEBT. very shallow burial depth and poor-quality installation technique. The

12
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

Fig. 10. Cracking in the tested pipes under TEBT (a) crack at the invert (b) multiple cracking in RC-1 pipe (c) radial crack (d) shear crack.

Fig. 12. Load-deflection curve of the tested pipes under patch loading.

patch load test conducted in the present study could more accurately
Fig. 11. Monitoring of crack width using crack measuring leaf gauge. mimic this loading scenario. Table 12 shows the comparison of results
obtained from the three-edge bearing test and patch load test on the
concrete pipes. All the pipes showed lower load capacity tested under
Table 11 patch load in comparison with the three-edge bearing test at both
P-load results corresponding to various stages of loading. cracking and ultimate stages. This is because a concentrated load is
Pipe PHL (kN) P0.3 (kN) Pult (kN) Pult/P0.3 applied to the pipe over a smaller area in case of patch load test as
RC1 115.2 172.3 305.6 1.77 compared to the three-edge bearing test.
RC2 113.8 153.3 220 1.44 The load (kN) in three-edge bearing test has been divided by the
RC3 112.3 134.2 153.3 1.14 length of the pipe (2.44 m) to obtain the distributed w-load (kN/m). In
PHL = Hairline crack load, P0.3 = 0.3 mm crack load, Pult = ultimate load. case of patch load test, although the load is applied over a small region,

13
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

Fig. 13. Cracking in the tested pipes under patch load test (a) longitudinal cracks along the invert and crown (b) cracks near the loaded patch.

Table 12
Comparison of results from three-edge bearing line load test and patch load test.
Pipe Three-edge bearing test Patch load test Effective length factor (experimental)

P0.3 (kN) Pult (kN) w0.3 (kN/m) wult (kN/m) P0.3 (kN) Pult (kN) w0.3,eq (kN/m) wult, eq (kN/m)

RC-1 210.4 353.2 86.3 144.9 172.3 305.6 100.2 177.7 0.82
RC-2 172.3 293.3 70.7 120.3 153.3 220 89.1 127.9 0.89
RC-3 153.3 200 62.9 82 134.2 153.3 78.0 89.1 0.88

P = load (kN) applied by the hydraulic jack; w = distributed load intensity (kN/m), weq = P-load in patch load test divided by the effective length calculated using Eq.
(9).

the longitudinal cracks due to the flexure action in the wall of the pipe that the patch load test equivalent distributed load results were higher
have been observed over the full length of the pipe, which implies that than the three-edge bearing distributed load results. For instance, at the
the length of the pipe effective in resisting the load is more than the stage of 0.3 mm crack width, the equivalent distributed load in the patch
loaded length of 0.5 m. This is because of the relatively high longitudinal load test for RC1 is 16 % higher than the distributed load in the three-
bending stiffness of the pipe. However, the response of the pipe was edge bearing test. This showed that using Eq. (9), the patch load test
more localized around the centrally loaded patch in terms of increased would result in somewhat underestimation of the effective length of the
crack widths and increased spalling of concrete cover around the cen­ pipe. An experimentally determined effective length factor of the pipes
trally loaded patch as compared to the ends of the pipe. Therefore, the tested in the present study can be determined using Eq. (10).
total length of the pipe cannot be considered effective in resisting the
P0.3,patch
load. This is also obvious from the fact that the pipes could sustain lower FLe = (10)
P0.3,TEBT
loads in the patch load test as compared to the three-edge bearing line
load test. Thus, a portion of the total length of the pipe should be where P0.3, patch is the load at 0.3 mm crack width in the patch load test
considered effective to obtain the equivalent distributed load, weq (kN/ (kN), P0.3,TEBT is the load at 0.3 mm crack width in the three-edge
m) by dividing the P-load by the portion of the total length. This portion bearing test (kN), and FLe is the experimentally determined effective
of the length may be called as the effective length of the pipe for the length factor. The effective length factors calculated using Eq. (10) are
purpose of calculating the equivalent distributed load (kN/m). There are given in Table 12. The effective length factors for patch load test
no published guidelines available to calculate the effective length of a calculated based on the experimental data in the present study ranged
pipe under the patch load test. Both the Concrete Pipe Design Manual from 0.82 to 0.89, which are somewhat higher than the effective length
(ACPA Concrete Pipe Design Manual, 2011) and Standard Installation factor calculated using Eq. (9) from Standard Installation Direct Design
Direct Design method (ASCE 15, 2017) calculate the effective length of a procedure (ASCE 15, 2017) or the Concrete Pipe Design Manual (ACPA
buried concrete pipe by Eq. (9). Concrete Pipe Design Manual, 2011). Although the effective length
3 factor may vary for different sizes of concrete pipes, the results of the
Le = L + 1.75 Do (9)
4 pipes tested in the present study seem to suggest some level of conser­
vatism in the effective length equations used in the present design
Where, L is the length of the spread load area at the level of top of the practice (ACPA Concrete Pipe Design Manual, 2011; ASCE 15, 2017).
buried concrete pipe (m), Do is the outside diameter of the pipe (m), and
Le is the effective length of the pipe in resisting the load. In the patch 3.5. Comparison of field-buried and laboratory tested pipes
load test conducted in the present study, the length of the loading pad
was 0.5 m. The effective length calculated using Eq. (9) was 1.72 m, The indirect design method of concrete pipes is based on calculating
which is 70 % of the total length of the pipe. The equivalent distributed the load (N/m) on the concrete pipe due to earth fill, fluid weight, and
loads (kN/m) obtained by dividing the P-load results in the patch load live load. The required D-load (N/m/mm) under the three-edge bearing
test by the effective length (1.72 m) are presented in Table 12. condition is then calculated using Eq. (11).
Comparing these equivalent distributed load results, it was observed

14
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

( )
We + Wf WL FS Table 13
D0.3 = + (11) Comparison of laboratory and field capacities of the tested pipes.
Bf,e Bf,LL D
Pipes Load capacity in Load capacity in field condition (kN/m)
where We, Wf, WL are the load (N/m) due to earth fill, fluid weight, and laboratory condition
(AASHTO, (BSI, (Alzabeebee
live load, respectively, Bf,e and Bf,LL are the bedding factors for the earth (kN/m)
2020) 2020) et al., 2018a)
and live loads, respectively, D is the inside diameter of the pipe (mm), FS
RC1 86 146 95 129
is the factor of safety and D0.3 (N/m/mm) is the minimum required D- RC2 71 121 78 107
load corresponding to 0.3 mm crack width in a three-edge bearing test. RC3 63 107 69 95
The FS was taken equal to 1. The earth load bedding factor for pipe
having inside diameter of 750 mm and installed in type-IV installation is
1.7 as per AASHTO LRFD bridge design specifications. The bedding the passage of dump truck. The observed deflections were identical
factor for equivalent bedding class D or N in BS 1295–1 (BSI, 2019) is for all types of tested pipes.
1.1. Several previous studies (Becerril García and Moore, 2015; Erdog­ 3) Results of the three-edge bearing test showed that the reduction in
mus et al., 2010; Lay and Brachman, 2014; MacDougall et al., 2016) the amount of flexural reinforcement reduced the D-load strength
have reported that the bedding factors used for the indirect design of corresponding to 0.3 mm crack width, but more prominent reduction
concrete pipes were overly conservative. For instance, Macdougall et al. was observed in the ultimate D-load strength. This also changed the
(MacDougall et al., 2016) reported that the live load bedding factor was failure mode from flexure to radial tensile failure when the steel
1.17 to 1.75 times higher than the AASHTO (AASHTO, 2020) bedding reinforcement was reduced.
factors. The earth load bedding factor was reported to be more conser­ 4) The comparison of the structural performance of the pipe in the
vative i.e., 1.77 times the AASHTO bedding factor. Although the pipes three-edge bearing and the patch load tests showed that the patch
were not subjected to cracking limit state test in the field in the present load test could more accurately mimic the vehicular live load on a
study, the absence of cracking or any distress in the pipe even at the shallow buried pipe. The length of the pipe that can be considered
burial depth of 0.19 m suggests that the existing bedding factors could effective in resisting the load under patch loading seemed to be 80 %
be reduced to economize the pipe design. A comparison of the load to 90 % of the total length of the pipe.
capacities of the tested pipes in the laboratory and field conditions is 5) The absence of cracking in the field tests showed that the pipes
given in Table 13. buried in the shallow soil cover of 0.19 m can withstand the live load
The field capacities of the tested pipes have been calculated using the due to the passage of typical construction equipment/machinery.
corresponding bedding factors as per AASHTO (AASHTO, 2020), BS
1295–1 (BSI, 2020) and Alzabeebee et al. (Alzabeebee et al., 2018a). The This study reinforced the findings from previous studies that the
maximum load applied to the pipes in the field condition in the present current design and construction practices of the precast concrete pipes
study was about 39 kN/m. The field capacities of the pipe based on are overly conservative. This study provides important experimental and
bedding factors by AASHTO (AASHTO, 2020) and Alzabeebee et al. analytical evidence to economize the existing design and construction
(Alzabeebee et al., 2018a) were 2.7 to 3.7 times higher than the practices. This study will act as a benchmark for experimental evalua­
maximum load applied to the pipes tested in the field in the present tion of the behavior of locally produced spun-cast pipes through labo­
study. This explains why the pipes did not show any cracking or sign of ratory and field-buried tests. Further investigations are required by
distress when tested in the field condition. Although the heaviest considering various parameters that influence the pipe behavior under
available construction equipment was passed over the buried pipes, the various loading regimes through a dedicated finite element analysis
pipes were not cracked under the heaviest available loads. The pipes based study in future.
could have cracked in the field if subjected to additional loading through
specially designed vehicular load or through loading blocks on site, CRediT authorship contribution statement
which is the main limitation of this study. This comparison shows that
the current design and construction practices are overly conservative. It Adeel Faisal: Writing – review & editing, Writing – original draft,
also shows that there is a need to adopt more rational guidelines that can Visualization, Methodology, Investigation, Funding acquisition, Formal
economize and simplify the present practice on designing and installa­ analysis, Data curation, Conceptualization. Safeer Abbas: Writing –
tion of the precast concrete pipes. review & editing, Supervision, Project administration, Funding acqui­
sition, Conceptualization. Ammad Hassan Khan: Writing – review &
4. Conclusions editing, Validation, Resources. Imtisal Ahmed: Resources, Methodol­
ogy, Investigation, Data curation, Conceptualization. Sbahat Shaukat:
This study explored the structural performance of reinforced con­ Writing – review & editing, Resources, Methodology, Funding
crete pipes locally manufactured through spun-cast centrifugal method acquisition.
of casting. Full-scale precast concrete pipes, having an inside diameter of
750 mm and wall thickness of 88 mm, were fabricated with various
levels of reinforcement in the form of double steel cage. The concrete Declaration of competing interest
pipes were tested in the field under vehicular load. Typical construction
equipment, such as dump truck and rollers, used at a local construction The authors declare that they have no known competing financial
site, were passed over the buried concrete pipes. The pipes were later interests or personal relationships that could have appeared to influence
tested in the laboratory under widely practiced three-edge bearing line the work reported in this paper.
load test and novel patch load test for comparison purpose. The
following specific conclusions can be drawn from the present study: Data availability

1) The pipes showed no cracking or sign of distress under the action of No data was used for the research described in the article.
various vehicular loads in the field condition. It can be concluded
that the pipes behaved in a linear elastic manner with the load Acknowledgement
resistance offered mostly by the concrete material.
2) The passage of heavy vehicles such as transit mixer resulted in minor The research grant provided by University of Engineering and
vertical deflection with the maximum deflection of 0.26 mm due to Technology Lahore for this study is gratefully acknowledged.

15
A. Faisal et al. Tunnelling and Underground Space Technology incorporating Trenchless Technology Research 149 (2024) 105813

References BSI, 2020. BS 9295. Guide to the structural design of buried pipes. British Standards
Institute, London, UK.
CSA A257.2, 2019. Reinforced circular concrete culvert, storm drain, sewer pipe, and
AASHTO, 2020. AASHTO LRFD Bridge Design Specifications, Ninth edition. ed.
fittings. Canadian Standard Association, Ontario, Canada.
American Association of State Highway and Transportation Officials, Washington,
CSA S6:19, 2019. Canadian Highway Bridge Design Code. Canadian Standards
DC.
Association (CSA), Mississauga, ON.
ACI Committee 318, 2022. Building Code Requirements for Structural Concrete and
Erdogmus, E., Skourup, B.N., Tadros, M., 2010. Recommendations for design of
Commentary (Reapproved 2022). American Concrete Institute, Farmington Hills,
reinforced concrete pipe. J Pipeline Syst Eng Pract 1, 25–32. https://doi.org/
MI, USA.
10.1061/(ASCE)PS.1949-1204.0000039.
ACPA Concrete Pipe Design Manual, 2011. Concrete Pipe Design Manual.
Faisal, A., Abbas, S., Ahmed, A., 2023. Mechanical performance of spun-cast full-scale
Alzabeebee, S., 2019. Seismic response and design of buried concrete pipes subjected to
precast pipes incorporating hybrid conventional rebar cage and steel fibers.
soil loads. Tunn. Undergr. Space Technol. 93, 103084 https://doi.org/10.1016/J.
Structures 52, 104–116. https://doi.org/10.1016/j.istruc.2023.03.176.
TUST.2019.103084.
Lay, G.R., Brachman, R.W.I., 2014. Full-scale physical testing of a buried reinforced
Alzabeebee, S., 2022. A comparative study of the effect of the soil constitutive model on
concrete pipe under axle load. Can. Geotech. J. 51, 394–408. https://doi.org/
the seismic response of buried concrete pipes. Journal of Pipeline Science and
10.1139/cgj-2012-0256.
Engineering 2, 87–96. https://doi.org/10.1016/J.JPSE.2021.07.001.
Li, B., Fang, H., Yang, K., Zhang, X., Du, X., Wang, N., Guo, X., 2022. Impact of erosion
Alzabeebee, S., Chapman, D., Jefferson, I., Faramarzi, A., 2017. The response of buried
voids and internal corrosion on concrete pipes under traffic loads. Tunn. Undergr.
pipes to UK standard traffic loading. Proceedings of the Institution of Civil Engineers
Space Technol. 130, 104761 https://doi.org/10.1016/J.TUST.2022.104761.
- Geotechnical Engineering 170, 38–50. https://doi.org/10.1680/jgeen.15.00190.
MacDougall, K., Hoult, N.A., Moore, I.D., 2016. Measured load capacity of buried
Alzabeebee, S., Chapman, D.N., Faramarzi, A., 2018a. Development of a novel model to
reinforced concrete pipes. ACI Struct J 113. https://doi.org/10.14359/51688059.
estimate bedding factors to ensure the economic and robust design of rigid pipes
Malik, A., 2015. Geotechnical statistical evaluation of lahore site data and deep
under soil loads. Tunn. Undergr. Space Technol. 71, 567–578. https://doi.org/
excavation design. Portland State University. https://doi.org/10.15760/CEEMP.17.
10.1016/J.TUST.2017.11.009.
Nehdi, M.L., Mohamed, N., Soliman, A.M., 2016. Investigation of buried full-scale SFRC
Alzabeebee, S., Chapman, D.N., Faramarzi, A., 2018b. A comparative study of the
pipes under live loads. Constr Build Mater 102, 733–742. https://doi.org/10.1016/J.
response of buried pipes under static and moving loads. Transp. Geotech. 15, 39–46.
CONBUILDMAT.2015.10.203.
https://doi.org/10.1016/J.TRGEO.2018.03.001.
Peter, J.M., Chapman, D., Moore, I.D., Hoult, N., 2018. Impact of soil erosion voids on
Alzabeebee, S., Chapman, D.N., Faramarzi, A., 2019. Economical design of buried
reinforced concrete pipe responses to surface loads. Tunn. Undergr. Space Technol.
concrete pipes subjected to UK standard traffic loading. Proceedings of the
82, 111–124. https://doi.org/10.1016/J.TUST.2018.08.003.
Institution of Civil Engineers - Structures and Buildings 172, 141–156. https://doi.
Ramadan, A., Younis, A.A., Wong, L.S., Nehdi, M.L., 2020. Investigation of structural
org/10.1680/jstbu.17.00035.
behavior of precast concrete pipe with single elliptical steel cage reinforcement. Eng
ASCE 15, 2017. Standard practice for direct design of buried precast concrete pipe using
Struct 219, 110881. https://doi.org/10.1016/J.ENGSTRUCT.2020.110881.
standard installations (SIDD), Standard Practice for Direct Design of Buried Precast
Scheperboer, I.C., Luimes, R.A., Suiker, A.S.J., Bosco, E., Clemens, F.H.L.R., 2021.
Concrete Pipe Using Standard Installations (SIDD). American Society of Civil
Experimental-numerical study on the structural failure of concrete sewer pipes.
Engineers (ASCE), Reston, VA. doi: 10.1061/9780784413074.
Tunn. Undergr. Space Technol. 116, 104075 https://doi.org/10.1016/J.
ASTM C497, 2020. Standard Test Methods for Concrete Pipe, Concrete Box Sections,
TUST.2021.104075.
Manhole Sections, or Tile. American Society for Testing and Materials, West
Sivakugan, N., 2021. Soil mechanics and foundation engineering: fundamentals and
Conshohocken, PA, USA. doi: 10.1520/C0497-20E01.
applications, 1st Edition. ed. McGraw-Hill Education, New York.
ASTM C76, 2022. Standard Specification for Reinforced Concrete Culvert, Storm Drain,
Spangler, M.G., 1933. The Supporting Strength of Rigid Pipe Culverts. Bulletin 112, Iowa
and Sewer Pipe. American Society for Testing and Materials, West Conshohocken,
State College.
PA, USA. doi: 10.1520/C0076-22.
Wen, Q.-J., Jing, H.-W., Sanda, S., Zhuang, S.-S., 2017. Experimental investigation of
ASTM D2487, 2017. Standard Practice for Classification of Soils for Engineering
mechanical properties of centrifugal concrete in circular pipes. J. Mater. Civ. Eng.
Purposes (Unified Soil Classification System). American Society for Testing and
29, 04016251. https://doi.org/10.1061/(ASCE)MT.1943-5533.0001771.
Materials, West Conshohocken, PA, USA. doi: 10.1520/D2487-17E01.
Wong, L.S., Allouche, E.N., Dhar, A.S., Baumert, M., Moore, I.D., 2006. Long-term
Becerril García, D., Moore, I.D., 2015. Behaviour of bell and spigot joints in buried
monitoring of SIDD Type IV installations. Can. Geotech. J. 43, 392–408. https://doi.
reinforced concrete pipelines. Can. Geotech. J. 52, 609–625. https://doi.org/
org/10.1139/t06-012.
10.1139/cgj-2013-0072.
Younis, A.A., Ramadan, A.S., Wong, L.S., Nehdi, M.L., 2020. New rational test for
BSI, 2019. BS 1295-1 Structural design of buried pipelines under various conditions of
reinforced-concrete pipe eliminating subjective crack-width criteria. Structures 28,
loading - General requirements. British Standards Institute, London, UK.
2507–2522. https://doi.org/10.1016/J.ISTRUC.2020.10.076.

16