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

net/publication/355486387

CONSTRUCTION-CENTRIC SUSTAINABILITY OF PRECAST SEGMENTAL BRIDGES

CONSTRUCTED WITH SPAN-BY SPAN METHOD

Article · October 2021

CITATIONS READS

0 171

1 author:

V N Heggade
Indian National Academy of Engineering
72 PUBLICATIONS   28 CITATIONS   

SEE PROFILE

All content following this page was uploaded by V N Heggade on 24 October 2021.

The user has requested enhancement of the downloaded file.

 CONSTRUCTION-CENTRIC SUSTAINABILITY OF PRECAST
SEGMENTAL BRIDGES CONSTRUCTED WITH SPAN-BY SPAN METHOD

VN HEGGADE

V.N. Heggade presently, the Chief Executive Officer (CEO), STUP Consultants & former Executive Director of Gammon is a
senior professional with a rich experience of over three and half decades in Construction sector in the areas of Design Management,
Technical Management, Site Management, Project Management & Contract Management of Highways, Bridges, Energy structures
like Chimneys & Cooling towers, Environmental, Marine and Hydraulic structures.
He is a recipient of around 10 National recognitions in addition to an International prize from various institutions, like IRC, ICI,
NDRF, IBC and IABSE. He has more than 150 publications to his credit and is a member of various IRC (Indian Roads Congress)
and BIS (Bureau Of Indian Standards) committees. He is also a member of TG 10.1 of Federation Internationale Du Beton, which
is a special Task Group working on FIB Model code 2020. Apart from being on Technical Board of ICI (Indian Concrete
Institute) and Academic Board of SPCE (Sardar Patel College Of engineering) is also a Fellow of Indian National Academy of
Engineering (FNAE).

Safety during fast-track construction involving 1. Introduction

manufacture, transportation and erection of heavy
precast/pre-stressed concrete elements is of utmost
importance. Any mishap or severe damage to
concrete elements during these activities may not
only lead to loss of lives but also result in delays
and cost escalation, thus adversely affecting the
basic pillars of sustainability – social, economic
and ecological. The author cites certain cases of
mishaps during the construction of bridges using
precast/pre-stressed segmental method and
analyses the possible root causes of failures. He
also provides useful recommendations. The author
would like to emphasize that the objective of the
paper is to basically create awareness amongst the
engineering fraternity about the dangers of the
‘scale effects’ of the increase in the deck widths
of the spine-and-wing construction and the lessons
to be learnt from the failures.
From the structural engineering point of view, the life
of the structure incepts from initial construction stages
to the point till the structure is destroyed and planned
again by the authorities. During the intervening period
structure experiences various stages such as
construction state, service state, rehabilitation state,
etc. and also exposed to different hazards that can
either be ‘Human-made hazards’ or ‘Natural hazards’
Human-made hazards consists of mistakes like
improper estimation of loading, improper analysis and
design, improper geotechnical assessment, improper
planning, accidental fire, collision and fatigues, choice
of the contractor, extended duration of construction,
new methods of construction/erection, removal of
support and collapse, defective workmanship and
material, mechanical and electrical breakdown during
launching, inadequate site management and defective
temporary works and their design.

82 Vol. 50 | Number 4 | September, 2020 The Bridge and Structural Engineer

The human-made hazards can be classified into two
main categories, namely, those during the
implementation of known tasks such as design and
construction and those that are not experienced
previously. Statstically, the former occurs more
frequently than the latter. The latter may not be totally
unforeseeable and could be attributed to carelessness
or deliberate oversight.
In Europe, many long span bridges like Cantilever
bridges, Cable supported bridges etc. were built by
adopting segmental construction technology, utilizing
the fast track construction concept available in the
1950s to 1990s. During such construction, many
mishaps took place like failure of prestressing, toppling
of cranes during construction, failure of form travelers
leading to accidents causing fatalities, huge financial
losses and delays. The creep and shrinkage models
adopted in prestressed concrete bridges have led to
excessive deflections at mid point of the long spans
of hundreds of bridges. In fact, the KB bridge at Palau
having 241.0m span, constructed in 1977 collapsed in
1996 due to excessive deflection of 1.61m and
excessive prestressing loss of 49%. It is worthwhile
to note that during this period in Europe, the measures
taken in design and construction for cost-optimization
and also adopting fast track construction had given
rise to many mishaps during construction and are
presently affecting serviceability of bridges.
It is distressing to see some cases in India following
the the old European practice in the recent past and
witnessing few mishaps during the construction of
some bridges and flyovers.In some bridges built in
Fig.1 : Precast segmental superstructure failures during construction [reformatted from ref.1]
The Bridge and Structural Engineer
the past, excessive deflections and prestressing losses
are observed similar to that of earlier European
experience.
Construction is full of high-risk activities such as
prestressing, erection of heavy components, and
floating of equipment that warrants meticulous
planning and system design to avoid accidents. The
incidents like sinking of barge mounted batching plant,
concrete pump and placer beam, the collapse of PSC
beams at casting yards, accident during launching,
tower crane collapse during erection, damage due to
staging failure, etc. are responsible for accentuation
of unsustainabilty by pushing the project back by years
and causing additional expenditure.
For the accelerated construction of long elevated
corridors and river bridges, the precast segmental
construction is a norm in India. However, there are
some alarming failures of precast segmental
construction in India recently.. Some photographs
published in CE&CR Journal [1] recently are provided
in Fig 1. These photographs show this trend.
The author suspects following common factors in these
failures:
They belong to the category of pre-cast segmental
construction using span-by-span method of
erection.
There was possibly a compression failure due to
crushing of concrete
They possibly adopted only one stage of
prestressing, inducing complete prestressing force
Vol. 50 | Number 4 | December, 2020 83
 into the section when only around 35 to 40 % of
the load is mobilized on the structure, particularly
for spine & wing construction.
Keeping in view the sustainability during construction
of precast segmental span-by-span method of erection,
the paper tries to assess the probable causes of failure
before arriving at some recommendations.
2. Probable causes of failures during design &
construction
One of the evergreen authorities on prestressed concrete
technologies, Dr Ing Fritz Leonhardt, way back in 1958
in his classical book ‘Prestressed Concrete Design and
Construction’ enumerates ten commandments for the
prestressed concrete engineer principally aimed at
mishap-free design and construction of sustainable
presstressed concrete structures [2]. These
commandments have never been as germane as today;
hence worth ruminating and recapitulating in verbatim
as below :
In the Design Office
1. Pre-stressing means compressing the concrete.
Compression can take place only where
shortening is possible. Make sure that your
structure can shorten in the direction of pre-
stressing.
2. Any change in tendon direction produces
“radial” forces when the tendon is tensioned.
Changes in the direction of the centroidal axis
of the concrete member are associated with
“unbalanced forces”, likewise acting
transversely to the general direction of the
member. Remember to take these forces into
account in the calculations and structural
design.
3. The high permissible compressive stresses must
not be fully utilised regardless of circumstances.
Choose the cross-sectional dimensions of the
concrete, especially at the tendons, in such a
way that the member can be properly concreted
– otherwise the men on the job will not be able
to place and vibrate the stiff concrete correctly
that is so essential to pre-stressed concrete
construction.
4. Avoid tensile stresses under dead load and do
not trust the tensile strength of concrete.
84 Vol. 50 | Number 4 | September, 2020
5. Provide non-tensioned reinforcement
preferably in a direction transverse to the pre-
stressing direction and, more particularly, in
those regions of the member where the pre-
stressing forces are transmitted to the concrete.
On the Construction Site:
6. Pre-stressing steel is a superior material to
ordinary reinforcing steel and is sensitive to
rusting, notches, kinks and heat. Treat it with
proper care. Position the tendons very
accurately, securely and immovably held in the
lateral direction, otherwise friction will take its
toll.
7. Plan your concreting programme in such a way
that the concrete can everywhere be properly
vibrated and deflections of the scaffolding will
not cause cracking of the young concrete.
Carry out the concreting with the greatest
possible care, as defects in concreting are liable
to cause trouble during the tensioning of the
tendons.
8. Before tensioning, check that the structure can
move so as to shorten freely in the direction of
tensioning. For distributing the pressure, insert
timber or rubber packings between tensioning
devices and the hardened concrete against
which they may be thrusting. Make it a rule
always to cover up high pressure pipelines.
9. Tension the tendons in long members at an early
stage, but at first only apply part of the pre-
stress, so as to produce moderate compressive
stresses which prevent cracking of the concrete
due to shrinkage and temperature. Do not apply
the full pre-stressing force until the concrete has
developed sufficient strength. The highest
stresses in the concrete usually occur during
the tensioning of the tendons. When tensioning,
always check the tendon extension and jacking
force. Keep careful records of the tensioning
operations.
10. Do not start grouting the tendons until you have
checked that the ducts are free from
obstructions. Perform the grouting strictly in
accordance with the relevant directives or
specifications.
The Bridge and Structural Engineer
Deck cross section evolution and scale effects
Precast segmental construction grew out of mass
production concept to prefabricate off site as much of
the bridge deck as possible which has many advantages
that include :
Enhanced quality control of casting in factory-like
conditions.
Ability to adjust the rate of casting to suit the rate
of erection by increasing the number of moulds
thereby flexibility to control the construction
programme.
Reduction of disruption to the existing users of the
site.
The flexibility to run the casting of the deck in
parallel with the construction of foundations and
sub-structures.
Overall reduction in construction programme due
to possible parallel activities and hence cost
savings.
The first application of span-by-span precast segmental
construction was realized in India for Krishna bridge at
Deodurg in 1971(Fig.2). Infact, this should be termed as
the turning point in the Indian history of precast
segmental construction. This 540 m long submersible
bridge across River Krishna at Deodurg in Karnataka,
Fig.2: Turning point in the history of precast segmental bridges in India
(Source: Gammon India Ltd)
The Bridge and Structural Engineer
having 18 spans of 30m each, was considered to be the
most ideal bridge for adoption of this technique. The
bridge was also the first submersible bridge in pre-
stressed concrete. The cross section of the decking was
a three-celled box catering for 5.5m wide deck, in the
shape of an aerofoil required from design considerations
of 5.5 m submergence and a maximum mean velocity
of 5.5 m/s under peak flood condition. The decking is
semi-continuous, with six span continuity for
superimposed and live loads. To facilitate proper access
for shuttering, concreting, etc. the 1.2 m long segments
were precast one above the other and the weight of the
segment was restricted to 10t. The segments were
supported on false work and the erection of segments
were done using specially designed travelling gantry
moving on top of already placed segments.
Since the decking was designed as six span semi
continuous structure, it was necessary to erect each
span on temporary staging. The precast segments
were assembled over pipe staging erected in the span.
A specially designed gantry moving over already
placed segments was used for erection of segments.
The segments were checked for proper alignment
prior to gluing together. Then the cables were stressed.
After establishing semi-continuity through gap
concreting, the decking was lowered over permanent
bearings. The gluing material was selected by
conducting field tests on various materials like epoxy,
Vol. 50 | Number 4 | December, 2020 85
mortar etc. Ordinary Max Pulls were used to impart
temporary pre-stress while gluing the segments
together.
In late nineties, India embarked upon the mammoth
infrastructure development especially in urban areas
where elevated corridors for both roads and metro
became popular. Especially where ground accessibility
is available, the span by span method of construction
became very popular due to its simplicity, speed of
construction and also as it could accommodate sharp
curvatures. Later, this method of construction was
extended to river bridges too.
The method which was started for 1.5 lanes of
vehicular traffic was progressively scaled up to 4 lanes
of traffic increasing the deck widths from 5.5m to
22.0m (Fig.3).
In the formative stages, up to 2 lanes to 3 lanes of
vehicular traffic which requires the deck width of
around 15m, the single cell box girder cross sections
could be effectively exploited. Wherever, more than
3 lanes of traffic was envisaged, 2 independent
carriageways were provided keeping the efficiency
of box girder cross sections in view. As the contract
periods started shrinking, to reduce the number of
operations by decreasing the number of foundations
to nearly half, the two independent carriageways
began to be clubbed together to accommodate in one
box girder cross section instead of two independent
Fig.3 : Increase in deck widths (Bot left:Godavari bridge Rajmundry,Bot right:Kosi bridge) (Source :
Gammon India Ltd)
86 Vol. 50 | Number 4 | September, 2020
cross sections. The avoidance of twin piers at a cross
sectional elevation with larger deck widths could also
provide space under the deck for traffic movement
which is vital in urban areas. This warranted twin
celled box girder; in certain cases deck was
transversely prestressed or long cantilevering wings
were supported by struts. This phenomena started
having what the present author calls as ‘scale effects’
during design as well as at construction stage.
The span-by-span method of precast segmental
construction by very nature of its fast track oriented
construction has certain peculiarities like, only one
stage of prestressing. Hence the entire prestressing
force or pre compression is induced in to the girder
during construction at one go. Thus the stresses are
stretched to the limits during construction stage .As
the deck widths started increasing, the weight of the
segments too started posing problems in handling
capacities of Bridge Deck Erection Equipment
(BDEs). Sometimes, the transportation of very wide
decks created issues while negotiating the curves in
congested urban areas. This gave rise to the latest
version of precast segmental box girder cross section
viz., spine and wing construction where the transverse
segment is broken in to three units, a spine and two
wings before finally made monolithic at the erection
location by transverse prestressing. The flyover from
JJ hospital to JJ School in Mumbai was first such
construction to be adopted in the country.
The Bridge and Structural Engineer
The main viaduct was 1932 m long between abutments
with overall width of 16.2m ( Fig 4) and 56 spans of
34.5 m each. Eight span continuity was provided
resulting in a distance of 276 m between expansion
joints. Minimum clearance of 7.5 m is provided from
road level to soffit of superstructure. Sharp curvature
had to be provided in the alignment at some locations,
minimum radius of curvature being 69 m.The decking
consists of three precast segments in the cross-
section, one central spine segment and two side
frames. This facilitated transportation and erection
of segments within the narrow access roads and
working space available. Maximum weight of segment
was restricted to about 45t. The three segments were
integrated by transverse prestressing.The deck is
supported on twin pot PTFE bearings on a central
pier located on 4 m wide central verge of the existing
road below. Bored cast in-situ piles of 1200 mm
diameter are socketed into rock. M75 grade high
performance concrete with micro-silica was used first
time in the country for a bridge structure to ensure
long term durability.
The problems associated with one stage prestressing
and stretching the permissible stresses to the limits
during construction stage immediately after
prestressing is much more accentuated in the spine-
Fig.4 : Dotted arrows indicating growth trend in deck widths of spine and wing construction
The Bridge and Structural Engineer
and-wing construction as is deliberated in succeeding
paragraphs. To compound the matter worse, the
widths of such construction is steadily growing (Fig.4)
to cater even up to 8 lanes (34m wide deck) which
started with 16.2m wide deck.
Joachim Scheer [3] in his book titled ‘Failed Bridges’
avers while talking about scale effects and
extrapolation that “enlarging or slenderizing tried and
tested designs when the ‘protective cover ’ of
experience is overstepped, hitherto insignificant
influences become predominant.” In load-bearing
structures, the geometric parameters, the dimensions
and their relationships are decisive for extrapolation,
because one feature of the development of structural
engineering was and is the tendency towards ever
larger and more slender structures. Innovative
structures are more difficult to assign to a
“continuation of relationships beyond the area in which
they are defined”
In each case of extrapolation one can identify a
situation where, in early examples of the structural
form, a certain factor was of secondary importance
with regard to stability or strength. With increasing
scale, however, this factor assumed primary
importance and led to failure. The accidents happened
Vol. 50 | Number 4 | December, 2020 87
not because the engineer neglected to provide
sufficient strength as prescribed by the accepted
design approach and code requirement, but because
of the unwitting introduction of a new type of
behaviour. Up to certain magnitude of the structure,
the secondary effects may not be so important and
don’t govern the design and as such can be neglected.
However such practice has limits of validity up to
certain scale of structure. Following a period of
successful construction the designer has become
complecent and simply extended the design method
that could have been the possible reason for failure.
2.2 Scale or Extrapolation effect during design
for construction stage
The cracking normally triggers the mishap in the post
tensioned segmental construction. Cracking in
segmental box girders may occur due to the
inadequacies arising out of the design for construction
stage, construction techniques adopted and quality
and workmanship of construction.The following
inadequacies during the design stage may initiate
the cracking, consequently resulting in mishaps.
Inappropriate consideration of stresses due
to moment redistribution,
Not considering thermal stresses during
construction stage (expansion or contraction),
Not considering thermal gradient effect during
construction stage.
Over or under estimation of initial prestress
resulting from optimistic or conservative
coefficients of friction and wobble.
Under or over estimating relaxation loss in
prestressing.
Tendon profiling where a large variation of
prestress eccentricity occurs and not checking
the stresses at critical locations.
Not considering the forces of Bridge Deck
Equipment (BDE) due to the coupling of BDE
with the permanent structure.
Not considering wind action on BDE and its
effect on permanent structure during
construction stage.
As has been demonstrated above, the width of the
cross sections for spine and wing construction has
88 Vol. 50 | Number 4 | September, 2020
been indiscriminately increased over the period to
enable accelerated construction as well as for
providing space under the deck in urban environment.
In certain spine and wing construction where the deck
width is increased indiscreminately the mishaps
occurred during construction and the recent mishaps
possibly has a pattern. The author is of the opinion
that there is a scale or extrapolational influence where
the secondary effect which could have been neglected
hitherto for small deck widths have started playing
role in these failures. The design and construction rules
may be in accordance with code requirements which
were extrapolated unwittingly and applied by
designers and constructors. These may not be valid
to cater for pronounced secondary effects for larger
widths of spine and wing precast segmental
construction.
Unlike in single unit tranverse segment where
cantilevering wing also contributes to the properties
of the section, in 3 units spine and wing cross sections,
the wings don’t contribute to the properties of the
section as such during construction. 100%
prestressing is applied on spine cross section when
merely about 50% load has arrived. Thus in spine
and wing construction, the stresses are stretched to
limits much more than in single transverse segment.
The above said scale effect is well demonstrated by
a live case study presented by Mr Hapreet Singh [4]
in the recently concluded refresher course on
Prestressed Concrete Technology where the author
was the moderator. The contract was a joint venture
between two contractors where nearly 6.1 km
elevated road was to be constructed catering for 6
lane traffic with the deck width of 25m. Out of the
total lengh, 3.3 km was passing from a congested area
requiring negotiation of sharp curves and as such it
warranted spine and wing construction to be cast
separately. The rest 2.8km could be single 25 m wide
transverse segment. However, from aesthetic
considerations and for uniformity, the same cross
sections were maintained for both the stretches (Fig.5),
though in the first case spine and wings were cast
seperately and in second case spine and wings were
cast monolithically. During the construction stage, at
the time of full transfer of prestress, for the first case
the load mobilised is 34 % with maximum compressive
stress of 16 MPa while for the second case the load
mobilised is 60% with the maximum cppressive stress
of 11.5 MPa.
The Bridge and Structural Engineer
 Fig.5 : Comparison of single transverse segment with spine & wings portions cast
separately [4]
Let’s synthesise the aspect of what author calls as
‘stresses stretching to limits’ in case of spine and wing
construction. Under the construction stage while
carrying out the SLS varification for ‘rare combination’
where thermal gradient also needs to be taken in to
account, the compressive stress is to be limited to
0.48fck. Though the IRC:112 code specifies that
thermal and wind conditions need not be considered
together for this varification, the author recommends
that the same should be considered during the
construction stage. During the construction stage in
span by span spine and wing construction, 100 %
prestressing is induced in one go when nearly 50%
Fig.6: Buckling of presrtessed concrete members (Source: concept drawn from ref [2])
The Bridge and Structural Engineer
load is on the structure. This makes the member
reaching its compressive stress almost it’s permissible
liimit of 0.48fck. For any reasons if there are any
unintended flexural stresses or axial compression
causing the buckling, there will not be any margin to
absorb this additional stresses in the spine beam. There
are sufficient reasons for unintended forces on the
prestressed spine beam which will be deliberated later
in this paper.
The buckling phenomenon of presressed concrete
member (Fig.6) is quite complex.The critical buckling
load for axially compressed member hinged at the ends
Vol. 50 | Number 4 | December, 2020 89
is given by well known Euler’s formula Pcr = 2EI/L2.
If the compressive force in the prestressed member
is produced by presressing tendons without touching
the member in anyway, the critical buckling load will
be similar to that of ordinary compressive member.
On the other hand, if the prestressing tendons are
continuously touching the member like in bonded/
grouted tendons, the buckling due to presressing is
completely ruled out. Gustav Magnel proved by
experiments that if the prestressing tendon is
connected at center of the member, the critical
buckling load for the member will increase by 4 times.
If the member is connected at ‘n’ equi-spaced points,
then the critiacal buckling load will be ‘n’2 times the
normal buckling load, provided that the concrete
compressive strength of the concrete is not exceeded
before the critical buckling load.
During the construction stage when full transfer of
prestress takes place in span by span construction
though the cables are not grouted, the tendons would
have been in contact with the prestressed member at
several points increasing the critical buckling load
manifold. However, to a large extent it alters the
compressive stress-strain property of the concrete
with regards to the buckling load. For the single
transverse segment, the stress strain curve above 01
(see Fig 6) would be available while for spine and
wing construction the curve above 02. At that point 02
of the curve, the modulus of elasticity of concrete is
much smaller than s = 0 and decreases drastically
reaching the danger zone much earlier than the
member without precompression. Hence spine beam
streched to the compressive stress limits are
susceptible to buckling phenomenon arising out of even
marginal unintended axial force (C).
The permissible compressive stress is checked under
SLS rare combination, hence in case of the uncertainty
of unintended buckling load on fully stressed spine
beam by one stage prestressing, one of the ways to
avoid the mishap due to buckling is to make sure that
the prestressed member is not a slender compression
member. This can be achieved by making sure
L/D 12 where L = spine beam length and D = depth
of the spine beam.
Further, as the deck width increases the width of the
spine beam also increases. If the design of the spine
beam is done by beam analogy, it is imperative to
ensure that the ratio W/D 4 where W = Width of
90 Vol. 50 | Number 4 | September, 2020
the spine beam at top, so that the beam does not start
behaving like a wall.
2.3 Potential unintended forces on spine beam
It has been demonstrated in the previous section that
the extrapolation of the design rules to a very wide
spine and wing construction may result in secondary
effects, which were insignificant for smaller decks,
but the same may start governing the design. Some
of the unintended forces during the construction stage
which may cause these secondary effects are
deliberated below.
It has been widely reported that there is a
conservatism in assuming the wobble and friction
coefficients leading to more prestressing in reality than
actually assumed in design calculations. Most of the
design codes specifies prestressing actions to be
considered as sup= Yinf= 1.0 during the construction.
This needs to be changed to Ysup=1.1 & Yinf = 0.9.
Further, it is imperative to establish Wobble and Friction
coefficients on the basis of actual tests at site to
ensure that very high pre-compression is not already
induced in to the section.
Also, during the launching of the segments, as can be
seen from the Fig.7, the coupled actions between
launching girder can induce unscheduled forces in the
girder (which is under erection) during the load
transfer and also on already installed back span.
While the segments are being erected in the span Pn+1
to Pn+2, the launching girder may induce restraints on
already installed deck by virtue of its fixity at ‘middle
support’ location and friction by ‘counter weighted
trolley’ at the rear end. This may induce additional
compression in the back span.The lauching girder is
supported on middle support and rear trolley by portal
frame action in transverse direction. For any reasons,
if the middle support and rear trolley are not exactly
on top of diaphragm segments, due to the wind action
on 2 to 3 m deep launching girder, there is a possibility
of torsion being induced.
In addition to thermal gradient across the depth of
the girder, uniform effective temperature variation may
also induce forces into the section, especially for the
backspan which is clamped between bearings and the
launching girder reactions.
As the widths of the box girders increase, the higher
widths warrant twin celled box - whether it is spine
The Bridge and Structural Engineer
 Fig.7 : A stage of Kinemetics of launching
(Source: Gammon India Ltd)
and wing construction or single transverse box
segment. It introduces new transverse deformability
issues when there is a vertical component of
prestressing especially when cross section
transversely becomes slender.
At around 1/3rd or 1/4th the span, the prestressing
cables are deviated vertically (which is depicted from
the two photos in the Fig.8 by the positions of
anchorages in the diaphragm segment and sheathing
ducts at midspan segments), inducing vertical
component of prestressing. The distribution of shear
stresses for such vertical component is made
conventionally with the assumption that the rigid cross
section is transversely not deformable (Fig.8 left
bottom). However, depending upon the stiffness of
component parts of the cross section this assumption
may not be correct (Fig. 8 right bottom). The rigid
un-deformable cross section in transverse direction
assumes that a centered vertical component of the
prestressing is evenly distributed among the webs.
In fact, a center pre vertical component of the
prestressing may induce a transverse deformation in
the top and bottom flanges with a displacement of the
The Bridge and Structural Engineer
center web larger than that of the external webs,
leading to cracks in the flanges.
Thus there is a very likely possibility of additional
unintended compression being induced in to cross
section for variety of reasons explained above.
2.4 ‘Scale effects’ during Construction
The following inadequacies during the construction
stage also may result in poor workmanship, ramifying
into mishaps:
Fabrication defects in precasting yard.
Inappropriate and not following the timely
(with in pot life of epoxy polymerisation)
epoxy jointing of segments.
Breaking of shear keys during handling of
segments.
Inadequate Geometric control during
matchcasting.
Non uniform temporary prestressing between
the segments.
Vol. 50 | Number 4 | December, 2020 91
 Fig. 8 : Transverse stiffness of wide two celled box
(Source : Concept is taken from the ref [5])

Lower strength of materials.
Over or Under prestressing
Unscheduled forces on permanent structure
due to faulty operation of Bridge Deck Erection
Equipment (BDE).
Most important distinction between full span precast
construction to that of segmental construction is the
joints between the segments that has to be perfectly
fitted with each other.These joints have to transfer
the compression through 100% contact area which is
ensured by match casting, either by long line method
or by shortline method.
Guidelines for design and construction of segmental
bridges, IRC:SP:65 [6]covers in detail the design and
construction requirement for precast segmental
bridges. Despite these elaborate guidelines, there
seems to be recurrent issues regarding the quality of
match casting of the segments leading to mishaps
during the construction due to following reasons.
Improper placement of concrete, resulting in
voids/empty pockets in bottom slab and also
honeycombing at the edges of the segments
near the end stopper.
The epoxy formulation for wet joints not
properly applied and temporary prestressing
not done within the potlife leading to less than
100% contact area between the match cast
surfaces.
The shearkeys being broken while
deshuttering or handling the segments leading
to imperfect fitting of the segments.
Less than 100% contact area between the
compression joints leading to longitudinal
cracks after final prestressing.
There are some issues which make the match casting
of precast segments sensitive to certain types of
defects. Due to the short pouring cycle between
segments (usually a day) the whole process of
segment construction can lead to different type of
construction defects if specific inspection and
measures are not taken adequately. Some of the most
common defects during the fabrication and erection
of segments are:
The continuous reuse of movable parts of the
formwork might affect its water tightness
after several reuse cycles. This can lead to a
leakage of the grout component of the
concrete mix.
The segment pouring sequence, particularly
if a self-compacting concrete mix is not used,

92 Vol. 50 | Number 4 | September, 2020 The Bridge and Structural Engineer

 is very sensitive to workmanship skills. The
compaction and pouring sequence within each
segment, honeycombs, voids or cold joints due
to improper spacer fixing are common
problems that might arise.
In very wide segments, there can be ‘bow
effect’ during match casting. The previously
cast segment which acts as a shutter face
may undergo curved face due to thermal
gradient and induce that curvature
permanently in newly cast segment. The old
segment will come back to its original shape
after cooling to ambient temperature. This
effect can be particularly severe when the
ratio of width to length of the segments is more
than 6.
The placing of the pre-assembled segment
reinforcement within the formwork and the
difficult access for inspection and location of
spacers can lead to lack of cover in some
locations.
During the segment movement to storage or
erection, damages can arise from improper
handling of the segment, such as chipping or
breaking of shear keys and local cracking if
temporary lifting points and support points are
not adequately designed or detailed.
The ‘glue joints’ with epoxy resin is provided
from durability considerations to ensure water
tightness in the joints to avoid any ingress of
water in to internally post tensioned tendons.
Normally, the preparation, application and
temporary prestressing of epoxy glue to
induce uniform pressure across the jointed
surface should be completed with in an hour
before the hardening of epoxy glue. The delay
in the process or the non-uniform pressure
across the depth of the segment may result
in undulated surface.
The segments should be repaired prior to their erection
as long as the defects do not affect the contact faces
between segments that are match cast. These faces
consequently should not be repaired prior to erection
and prestressing of the segment. Conventional
concrete repairing techniques such as the use of micro
concrete or grout mixes with low shrinkage, after
cleaning the affected area, are common repairing
strategies.
The Bridge and Structural Engineer
Among the other factors deliberated above, epoxy
gluing, temporary prestressing and bow effect can
be attributed to extrapolating the cross sections to
wider deckwidths and applying standard construction
methods similar to time tested ones for smaller
deckwidths.
In a tropical country like India where rain fall is quite
considerable, the epoxy glued wet joints are preferred.
The objective of the epoxy glue is to make sure that
there is no water ingress in to tendon ducts leading to
corrosion of tendons in the long run. The epoxy glue
does not increase or contribute to the structural
efficiency in any way. In fact, this is an additional
time consuming activity. However, during the
investigation of recent failures, it has been observed
by investigating committees that the epoxy application
in the joints are not uniform but acts as shim reducing
the contact area between the segments thereby
increasing the compressive stresses. As per the
requirement IRC : SP:65 , 1mm thick epoxy shall be
applied on both the joint surfaces, 0.3 MPa temporary
prestressing uniformly across the surface to be
maintained for 24 hours. The mixing resin and hardner,
application and temporary prestressing shall be
completed within an hour. These specifications are
same irrespective of the area of the epoxy to be
applied. With the ever-increasing deck width ( In the
initial days one side of the segment area of 6 m2 has
now gone beyond 14 m2) and corresponding joint
area, the practicality of application of epoxy with in
potlife has become cumbersome. Assuming, that there
are 13 joints in a span of 40m, including temporary
prestressing the total time required for epoxy
application is 13 hours. After the last joint is
prestressed, the pressure needs to be maintained for
24 hours. This means the total activity of the epoxy
application requires 37 hours for a 40-m span. Based
on this, the additional time required for 10 km elevated
corridor, the time required for epoxy-related
applications will work out 385 days. Thus epoxy glued
joints not only increase the risk of additional
compressive stress due to poor workmanship but also
time and cost overrun. The more sustainable precast
segmental construction is dry joints with external
presressing. By avoiding the internal prestressing, the
apprehension of water ingress leading to possible
corrosion of the tendons is obviated apart from
external tendons being accessible for inspection and
maintenance.
Vol. 50 | Number 4 | December, 2020 93
 Fig.9: Bow effect [7] as the segments become slender transversly
Another scale effect during match-casting due to the
ever-increasing deckwidth is ‘bow effect’ (Fig. 9).
In precast segmental construction, adjacent segments
must match perfectly to ensure the load transfer (shear
and normal forces) through the un-reinforced joint
between two precast elements, hence the match
casting is used for segments production. However,
the heat of hydration and the ambient temperatures
may result in curved segments.
The curvature of both segments is formed by the
thermal gradients in the match cast segments. The
face of the already cast segment is used as a
shuttering to cast the new segment. Because of the
heat of hydration in the new segment a thermal
gradient is formed in the match cast segment leading
to a curved side between the two adjacent segments.
After the seperation when it comes back to original
ambient temperature, the match cast segment returns
to its original shape whereas the new segment has a
permanent curvature. Factors influencing the blow
effect are, width to length ratio of the segment, heat
of hydration and heat diffusion at the segment
surfaces.
3. Quality control during design and
construction
In comparison to other conventional prestressed
concrete bridges, for precast segmental bridges, the
94 Vol. 50 | Number 4 | September, 2020
influence of construction method, construction
sequence and construction accuracy are significantally
important. Therefore, it is imperative that qualified
engineer(s) with the sufficient knowledge and
experience are constantly involved in the design
scrutiny and Bridge engineers are stationed at the
construction site to provide appropriate construction
management.
The design and construction of precast segmental
bridges require a tight integration between several
interfaces as part of an industrialised process, and
requires specific measures to be taken into account
in the quality control of the geometry, materials and
erection and also specialised workmanship in different
areas.
Detailed documents including method statements,
fabrication and erection procedures and testing
requirements need to be specified in the contract
documents to assist the relevant teams involved in
the project to reduce the risks during construction.
The current organisational structure for managing the
EPC contract consist of a design consultant to the
contractor, proof-checking consultant to approve the
design of the contractor’s designer, safety consultant
to approve the project highway safety and finally
Authority Engineer (AE) representing the Authority
acting as a supervision consultant including final design
The Bridge and Structural Engineer
approval and quality control at the site. The contractor
engages the prestressing system supplier and Bridge
Deck equipment supplier. Some times contractor has
his own BDEs.
On many occasions, the proof consultant and safety
engineers are engaged, without appropriate
consideration to their qualification and experience in
precast segmental construction. The expectation
from proof-checking and safety consultants is that
the design has to be approved immediately after
being submitted by the contractor’s designer. In the
author’s opinion, there is a strong need to look in to
the pre-qualification of proof checking and safety
consultants.
Some EPC contract specifies that the entire design
with approvals and the 10 % of the work has to be
completed within a very short period of the award of
the work including geotechnical investigations,
surveying, etc. In many cases, this is impractical and
the infeasible haste leads to many erroneous designs,
approvals and bad quality of construction resulting in
mishaps. This criteria has to be reviewed going
forward for safe construction.
Bridge Deck equipment design including launching
system is very important part of precast segmental
construction. There are instances where design
consultant and the supplier of BDE are the same.
There is a strong conflict of interest here and this
practice of design consultant and BDE supplier to be
a same agency should be discontinued from the safe
construction point of view. Presently, there is no
system of proof-checking the launching system or
BDE which needs to be introduced with immediate
effect.
In the present Authority Engineer (AE) contract, there
is no provision of structural engineer with necessary
qualification and experience as a key personnel. This
needs to be introduced with immediate effect to add
another layer of checking the designs.
4 Conclusions
The bridge failures using precast segmental-type
construction should be a matter of great concern to
the engineering fraternity. Ensuring sustainablity
during the construction is a very important component
of the overall sustainable construction practices. The
accidents and collapses during construction have
The Bridge and Structural Engineer
enormous impact on all the three pillars of sustainability
- viz. Social, Ecological & Economical.
After due synthesis of the probable causes of failures,
the article provides some recommendations for
ensuring sustainability during the precast segmental
span-by-span method of construction as below.
For construction stage, SLS check for rare
combination involving wind and thermal
loads should be carried out as per IRC:112.
For span-by-span precast segmental
construction with one stage prestressing,
prestressing action should be ãsup=1.1 &
ãinf = 0.9. For SLS verification under rare
combination, wind load should be
considered in combination with thermal
loads.
The spine cross section for spine-and-wing
construction should be proportioned such
that LD 12 where L = spine beam length
and D = depth of the spine beam, and
WD 4 where W = width of the spine beam
at top, so that the beam does not start
behaving like a wall.
There are some issues which make the
match casting of precast segments
sensitive to certain defects. The segments
should be repaired prior to their erection
as long as the defects do not affect the
contact faces between the segments that
are match cast. These faces consequently
should not be repaired prior to erection and
prestressing of the segment without risking
further damage.
In case of segments with width-to-length
ratio higher than 4, the mock-ups for match
casting of segments should be done to
arrive at optimum mix design of concrete.
This should be made mandatory.
The wet joints should be prohibited and dry
joints with external prestressing should be
encouraged for span-by-span method of
precast segmental construction.
There is a strong need to look in to the
technical competence of pre-qualification
of proof checking and safety consultants.
Vol. 50 | Number 4 | December, 2020 95
View publication stats
The criteria in EPC contract to complete
the design with approvals and the 10 % of
the work within a very short period of the
award of the contract including
geotechnical investigations, surveying,etc.
need to be reviewed.
The design of launching system or BDE
should be proof-checked and load-testing
should also be conducted.
An experienced and qualified structural
engineer should be introduced as a key
personnel in Authority Engineer (AE)
contract.
5 References
1. Bhowmick, Alok. Learning from Bridge
Failures-Challenges and Opportunities, CE&CR
Journal, February 2021,Vol.34, No.02. pp 22-
33.
2. Leonhardt, Fritz, Prestressed Concrete Design
& Construction, Wilhelm Ernst & Sons, 1964.
Berlin, Munich.
96 Vol. 50 | Number 4 | September, 2020
3. Scheer, Joachim, Failed Bridges Case Studies,
Causes and Consequences, 2010 Wilhelm Ernst
& Sohn, Verlag fu¨r Architektur und technische
Wissenschaften GmbH & Co. KG, Rotherstraße
21, 10245 Berlin, Germany.
4. Singh, Harpreet, Design & Construction of Spine
& Wing Segmental, and Strutted Box Segmental
Construction, Online Refresher Course on
“Pre-Stressed Concrete for Bridges &
Buildings” from 20thMarch - 1stMay 2021.
5. Podolny, WalterJr, The cause of cracking in Post-
tensioned Concrete box girder bridges and
retrofit procedures, PCI Journal, March-April
1985,pp 82-139.
6. IRC: SP:65, Guidelines for design and
construction of segmental bridges, Indian Roads
Congress, New Delhi.
7. Fib bulletin 82, Precast segmental Bridges,
Guide to good practice.
The Bridge and Structural Engineer