BY

Ms. SHRADDHA SUNIL KADAM

(REG NO. C1652005)

UNDER THE GUIDANCE OF

DR A.A.BAGE
CONSTRUCTION STAGE ANALYSIS OF SEGMENTAL BRIDGE

MASTERS OF TECHNOLOGY (CIVIL)

WITH

STRUCTURAL ENGINEERING (FULL TIME)

SUBMITTED TO SARDAR PATEL COLLEGE OF ENGINEERING, AN AUTONOMOUS INSTITUTE AFFILIATED TO
UNIVERSITY OF MUMBAI IN ACADEMIC YEAR 2016-17

DEPARTMENT OF CIVIL ENGINEERING

SARDAR PATEL COLLEGE OF ENGINEERING
MUNSHINAGER, ANDHERI (W) MUMBAI 400 058
2016-17

1
 Bharatiya Vidya Bhavans
SARDAR PATEL COLLEGE OF ENGINEERING

Mumbai

CERTIFICATE
This is to certify that Miss. Shraddha Sunil Kadam Seat No. MCSI005
had successfully completed the seminar work entitled
CONSTRUCTION STAGE ANALYSIS OF SEGMENTAL BRIDGE
in the partial fulfillment of M.Tech. in Structural Engineering.

Date:
Place: Mumbai

2
 TABLE CONTENT

CHAPTER 1 INTRODUCTION

CHAPTER 2 CONCRETE BRIDGE ERECTION TECHNIQUES

2.2 Cantilevering Method

2.3 Precast Construction

2.4 Cast-In-Place Construction

2.5 Progressive Placement Method

2.6 The Linn Cove Viaduct

2.7 Concluding the Cantilevering Process

2.8 Cantilever Erection Equipment

2.8.1 Form Travelers

2.8.2 Launching Girders

2.8.2.1 Launching Girder Slightly Longer Than One Span

2.8.2.2 Launching Girder Slightly Longer Than Two Spans

CHAPTER 3 INCREMENTAL LAUNCHING

CHAPTER 4 FALSEWORK

4.1 Stationary Falsework

4.2 Traveling Falsework

4.2.1 Temporary Towers

4.2.2 Span-By-Span Erection

4.3 Consideration of Construction loads and stresses

4.4 Types of Construction Loads and Influences

4.5 Segment Casting cycle

CHAPTER 5 CONCLUSION

3
 FIGURES

Figure 1 Balanced Cantilever Construction

Figure 2 Progressive Placement Method

Figure 3 The Linn Cove Viaduct, North Carolina, U.S. (taken from Rives
1997, p41)
Figure 4 Form Traveler Front View
Figure 5 Working Scheme of Short Launching Girder
Figure 6 Working Scheme of Long Launching Girder
Figure 7 Incremental Launching
Figure 8 Span-By-Span Erection

4
 TABLES

Table 1 Typical Duration of Casting Steps

Table 2 Sequence of Casting Steps for Aichtal Bridge

5
Abstract
Constructability considerations are of importance in segmental bridge construction as the

occurrence of failures of bridge superstructures under construction have highlighted. For a safe

and economical construction process the interactions between construction loads and the

permanent structural system, depending on the chosen erection method, need to be evaluated.

Segmental bridges can be constructed with methods like Balanced Cantilever Construction, where

individual spans are counterweighted about their substructure support. Time-dependent material

properties like strength of the newly cast concrete, as well as shrinkage, creep, and relaxation

influence the structural system resistance. Resulting stresses in the unfinished bridge structure

during construction can even exceed the final stresses during service.

This paper makes an educational contribution by illustrating these concepts with the case

study of the Wilson Creek Bridge in Virginia. This five-span cast-in-place bridge was constructed

using Balanced Cantilever Construction. Two form travelers were used to construct cantilever

arms about the pier tables until the full span was finally connected at midspan; casting cycle

duration for a single segment was 7 days. The contractor implemented major constructability

changes in both the design and the construction of the bridge to facilitate a more economical

construction process.

6
 CHAPTER 1
INTRODUCTION

Construction of segmental bridge superstructures is a complex process due to the multiple

effects that arise from the erection method that is employed. Depending on the erection method,
which relates to the physical means of constructing the foundations, bridge substructure, and
especially its superstructure, the structural resistance of the bridge changes with each construction
stage, with the load conditions, the time-dependent materials behavior, and with environmental
influences. These complex interactions need to be considered in the construction engineering
structural analysis and for developing the flow of work tasks in the construction schedule.
Furthermore, each erection method requires its specific equipment and site installations to carry
out the necessary work tasks, which impacts the project budget.

Part I. Construction of Segmental Bridges

Concrete segmental bridges were built as early as 1925 (Plougastel Bridge) by French
engineer Freyssinet (1879-1962), who was among the first to implement prestressed concrete in
bridge construction, as Menn (1990) reports. Since then, prestressed concrete bridges, both as pre-
tensioned and post-tensioned have seen a rapid development. In most cases post-tensioning is
employed, i.e. the prestressing tendons are stressed with hydraulic jacks after the concrete has been
placed and gained a minimum strength. Usually the post-tensioning tendons are located in steel
ducts within the concrete segment and are terminated in special anchorages.
Concrete segmental bridges utilize box girder superstructures, which according to Troitsky (1994)
have been used in the U.S. since 1973. These superstructures consist of bottom slabs, webs (that
can be inclined), and a cantilevering top slab to provide maximum deck width. Concrete box
girders have multiple advantages, e.g. their versatility in alignment, width, and depth, high
torsional and bending stiffness of the closed cross-section, and an aesthetically pleasing geometric
appearance.

Segmental Bridge Construction

Segmental construction is a method of construction in which primary load-carrying
members are composed of individual members called segments post-tensioned together (Podolny

7
and Muller 1982, p10). Segmental construction limitations logically follow from the technical
limitations of erection methods and the construction equipment. Cranes, concrete pumps, form
travelers, and other pieces of equipment have certain physical limitations as to the volume and
weight of material that can be erected at one time.
Subdivision of the superstructure into segments can be made both in the transverse and the
longitudinal direction. Separation of segments in the vertical axis is found less frequently. Vertical
segmentation is used e.g. in composite bridge superstructures that consist of steel beams, trusses,
or steel box girders with a concrete deck slab. Longitudinally divided segments are load-carrying
members that span the complete length of one bridge span, e.g. the use of multiple prestressed
concrete AASHTO (American Association of State Highway and Transportation Officials).
Segments in such a bridge usually are 3 to 5 m long (Fletcher 1984). They can weigh up to 250
metric tons in highway and Transportation Officials) girders, which are laid parallel and then
covered with a deckmetric tons in precast cantilevering construction (Podolny and Muller 1982)
or up to 300 metric tons for cast-in-place (CIP) cantilevering construction (Fletcher 1984). Podolny
and Muller (1982) caution to keep these segments as regular in their geometry and as straight in
alignment as possible and to keep obstructions through e.g. diaphragms in the box girder at a
minimum.
The erection of segments divides the overall construction process into repetitive steps that
facilitate a learning process (Fletcher 1984) and improving productivity in the subsequently
erected spans. Segmental construction thus leads to economic and rapid erection of the bridge
superstructure. A major advantage of segmental construction also is the ease with which it can be
adapted to the specific requirements of the project (e.g. geometry, span lengths, etc.) and to the
capacity of the equipment available to the contractor.

8
 CHAPTER 2
CONCRETE BRIDGE ERECTION TECHNIQUES

Concrete segmental bridges have already been introduced in Section 3.6.1. The following
sections will present the important methods that are used in erecting concrete segmental bridges
nowadays and the equipment employed. Special focus is put on constructability issues, pertaining to
characteristics and requirements, advantages, and disadvantages of each method to prepare for the
case study that is presented in Chapter

2.1. Cantilevering Method

Before used in construction of concrete bridges, the cantilevering method had already been
used in Asia for wooden structures of earliest times, as Podolny and Muller (1982) report. Amongst
the major steel structures that were erected with the cantilevering method are the Firth Rail Bridge
and the Quebec Bridge that are presented in Section 2.1.5. Erection of concrete bridges with the
cantilevering principle led to development of specialized sequences that are discussed further
below.
As already introduced in Section 3.6.2, cantilevering for concrete segmental bridges is a
construction method where segments, either precast or cast-in-place, are assembled and stressed
together subsequently like a chain to form the self-supporting superstructure. Prestressing cables
located in the upper part of the segment cross-section support the cantilever. In the variant of the
progressive placement method stay cables are often used to support the cantilever prior to closure
of the span.
Time-dependent material behavior of the segments under successive load steps requires
comprehensive calculations for all construction stages. Every segment will develop strength with
increasing age of the concrete. Governing for the structural behavior of the cantilever is that every
segment carries and transfers loads from all following segments and construction loads until
closure of the span. From these very basic facts in conjunction with geometry and expected loads
on the structure the calculation of moments and local stresses, as well as calculation of the
deflections that they cause is possible. Optimization of geometry, prestressing, and camber are
then performed.
Depending on the specific segment configuration and erection sequence chosen for the
cantilevering method the cantilever may never be exactly balanced so that the superstructure needs

9
to be balanced to ensure stability. It is possible to fix the supports at the piers of cantilevering
superstructures and install vertical prestressing tendons. Furthermore it is very common to make
use of an additional temporary pier with vertical prestressing that is located close to the permanent
one (Casas 1997). This pier helps withstanding overturning moments from unbalanced load cases
on the bridge superstructure.
Several advantages have contributed to the success of the cantilevering method. Certainly
the most important one is that no falsework or centering is required, leaving traffic under the spans
widely unobstructed during construction. Access from the ground is only necessary for
construction of the piers and abutments and in preparation for the start of cantilevering, which
starts from these locations.
Only relatively little formwork is required due to the segmental nature of the
superstructure. Cantilevering is a very feasible method if the bridge spans are too high above
ground for e.g. economical use of falsework, and if the terrain under the spans is otherwise
inaccessible or unfeasible, being e.g. a deep gorge with danger of flood events. Especially in these
cases rapid construction can be achieved with cantilevering.
Fletcher (1984, p13) notes that especially in cantilevering complete calculations are
required for the construction stage[s] and these are complicated as many stressing effects are time-
dependent. In addition to this, the influence of stepwise construction needs to be considered.
However, the statical system that needs to be analyzed is rather simple and in case of the cantilever
prior to closure at midspan even statically determinate.

2.2. Precast Construction

Precast construction means that bridge members or segments are prefabricated at a location
different that the site, transported to the site, and installed there. Mathivat (1983) gives the
maximum economical span of bridges built in precast segment as about 150 m, since cost for the
placement equipment increase considerably the longer the spans are. Construction with precast
segments has several advantages in comparison with cast-in-place segmental bridges. Casting of
the segments can be performed under controlled, plant-like conditions at the precasting yard
This industrialized process allows easy quality control of segments prior to placement in the
superstructure and saves money through reuse of the precasting formwork. Surface finishing
works, such as texturing, sandblasting, painting, and coating can be performed on the ground level

10
without scaffolding when the segments are still accessible from all sides prior to installation in the
superstructure.
Another major advantage mentioned by Mathivat (1983, p212) is that the complete casting
of the superstructure can be removed from the critical path of the overall construction schedule,
since superstructure segments can be precast during construction of the substructure. Assembly
of the bridge superstructure takes much less time than cast-in-place construction, as precast
segments do not need to cure on site before being prestressed together. Through the early casting
of segments material properties are also influenced positively. As segments are usually stored at
the precasting yard or on site for a while the concrete will have gained more strength until
installation than cast-in-place elements have when being loaded. The time-dependent effects of
concrete shrinkage and creep will occur with reduced extent because of the increase age of the
concrete segments (Mathivat 1983) and will cause smaller deflections of the superstructure than
with cast-in-place construction.
This industrialized process allows easy quality control of segments prior to placement in
the superstructure and saves money through reuse of the precasting formwork. Surface finishing
works, such as texturing, sandblasting, painting, and coating can be performed on the ground level
without scaffolding when the segments are still accessible from all sides prior to installation in the
superstructure.
Another major advantage mentioned by Mathivat (1983, p212) is that the complete casting
of the superstructure can be removed from the critical path of the overall construction schedule,
since superstructure segments can be precast during construction of the substructure. Assembly
of the bridge superstructure takes much less time than cast-in-place construction, as precast
segments do not need to cure on site before being prestressed together. Through the early casting
of segments material properties are also influenced positively. As segments are usually stored at
the precasting yard or on site for a while the concrete will have gained more strength until
installation than cast-in-place elements have when being loaded. The time-dependent effects of
concrete shrinkage and creep will occur with reduced extent because of the increase age of the
concrete segments (Mathivat 1983) and will cause smaller deflections of the superstructure than
with cast-in-place construction.
However, cost for the precasting yard, storage, transportation, and installation of precast
segments needs to be evaluated in comparison with cost for the form travelers for cast-in-place
construction to achieve an economical solution.The precasting yard requires investment in
equipment. Adjustable formwork to form the bridge geometry and alignment needs to be installed.

11
Lifting equipment is also required to put the segments into the storage area and later load them on
truck to be hauled to the construction site.
It is common practice to use the match-cast method to achieve high accuracy in segment
prefabrication. Match-casting means that the segments are cast in the formwork between a
bulkhead at one end and a previously cast segment at the other (Levintov 1995, p46). Segment
joint faces need to be clean of any dirt for match-casting.
Levintov (1995) distinguishes concrete segment prefabrication into short-line casting and long-
line casting. Short-line casting would comprise formwork of the length of only one segment; with the
previously cast segment being moved into position for match-casting on a mobile carriage. Short-
line casting can be carried out in the horizontal position or with the segments tilted facing upward
(Podolny and Muller 1982), however, the normal horizontal position facilitates match-casting. The
overall bridge alignment requires careful adjustment of the formwork prior to each concrete
placement. Short-line casting does not take much workspace.
Long-line casting on the other hand means erection of formwork for about a complete
bridge span. According to Levintov (1995) the formwork can be erected stationary for the
superstructure soffit only, with smaller movable forms for web sides and interior formwork. This
formwork will be cheaper than the flexibly adjustable formwork for short-line casting, but will
require much more workspace. Levintov cautions that the long-line casting is feasible for straight
superstructures or superstructures with constant curvature. Segments are match-cast progressively
on the long-line formwork by step-by-step advancement of the movable formwork units and a
movable bulkhead.
Phipps and Spruill (1990) describe the precasting cycle that was used in construction of the
Biloxi Interstate I-110 viaduct. According to them, the freshly cast segments were steam cured in
a movable shed covering the casting bed of the short-line formwork. The pretensioning strands
were released by cutting them, quality control and testing of concrete samples was performed, and
internal formwork units were removed from the new segment. After lifting the previously cast
segment from its position for match-casting into the storage area, the new segment was rolled out
of the formwork. It was positioned for match-casting according to the required overall alignment.
Cleaning of the joint face and the bulkhead was done prior to casting the next segment.
Reinforcement bars were preassembled in reinforcement cages to speed up placement. Pre-
tensioning strands were used in the box girder segment, being stressed prior to concrete placement.
After concrete placement and consolidation with vibrators the segment was screeded and given a
surface finish before the curing shed was set up over the casting bed. With the sequence described

12
a casting cycle of one superstructure segment per day could be achieved. In the final superstructure
post-tensioning cables were installed to stress the precast segments together
Precast segments have joints that require special attention. An epoxy agent is usually
applied to the joint faces shortly before putting a segment into its location in the superstructure.
Joints are usually only a few millimeters thin. Podolny and Muller (1982) explain the functions of
the epoxy agent that is applied to the joint faces when placing precast segments. During segment
placement the epoxy serves to lubricate the joint faces, which are cleaned by sandblasting and
compensate for minor imperfections in the match-cast surfaces (Podolny and Muller 1982,
p485). In the finished structure the hardened epoxy seals the joints against moisture and thus
additionally protects the tendons in their ducts. Furthermore, the epoxy is able to transmit
compressive forces and shear forces. Information on mixing, handling, and properties of the two
main ingredients, the epoxy resin and the hardener, is provided by Podolny and Muller (1982).
Interestingly, the epoxy agent can reach a higher final strength than the concrete itself.
In addition to the epoxy transmitting shear forces between segments the joint faces are
given a special shaping to transmit shear. So-called shear keys are cast into the joint faces to lock
the segments together. They transmit shear forces and also help in exact alignment of the segments
during assembly. Segments of the so-called second generation facilitate many smaller shear keys
that are located not only in the box girder webs, but also in top and bottom flanges (Podolny and
Muller 1982).

2.3. Cast-In-Place Construction

Podolny and Muller (1982) provide an example for a typical casting cycle. As outlined in
Section 3.6.3.1, any previously cast segment needs to have developed at least the specified strength
to be prestressed to previous elements and support the subsequent one. After finishing all work on
a segment the form traveler is detached from the previous position and moved forwards on rails
that are mounted on the bridge superstructure. In order to remain balanced during advancement
the form traveler may be equipped with a counterweight. Upon arrival at the new position it is
adjusted and anchored to the existing superstructure at its rear to be able to withstand overturning
moments that will occur from the weight of new concrete. The external formwork is cleaned and
aligned to the required geometry of the next segment, also incorporating the desired camber
When the form traveler has thus been prepared the reinforcement and tendon ducts for bottom
slab and webs will be installed and connected with the previous ones. In cast-in-place

13
 construction it is possible to have continuous mild reinforcement in the superstructure, whereas
in precast segmental construction only the longitudinal prestressing tendons will cross the
segment joints.
Reinforcement can be pre-assembled into cages that are lifted into place by crane.
Prestressing tendons are already inserted into their ducts prior to placement of concrete because
of better accessibility. After these preparations, concrete is placed. Accessibility of the bottom
part of the box girder may require that the bottom slab is cast before internal formwork for webs
and top slab is advanced and aligned. After curing sufficiently for strength and durability, the
tendons in the newly cast concrete segment can be prestressed. Finally, the casting cycle starts
all over again to cast the next segment.
Concrete placement can be carried out by various means, e.g. with buckets that are hoisted
by crane, or by pumping. While placing the concrete in lifts care needs to be taken that no
segregation of the concrete mixture occurs, and that proper consolidation will be achieved. The
most common method is to vibrate the concrete in the formwork by means of internal or external
vibrating devices. Most important for the quality of the concrete is curing to achieve strength
and durability. Upon gaining enough strength, the tendons in the newly cast segment will be
stressed to some degree and the cycle starts all over again. Based on Mathivat (1983, p201) an
overview of casting steps and typical values for their duration is be given in Table 4-2 It should
be noted that the sequence of steps given in this table is only a generic example and would be
broken down into more steps for planning an actual construction project:

Table 1: Typical Duration of Casting Steps

Duration Activities
Post-tensioning tendons in previous segment
Stripping formwork
1 day Advancing form traveler
2 days Placing reinforcement, ducts, and tendons
1 day Placing concrete for bottom slab, webs, and top slab
3 days Curing concrete (including Sunday)

Mathivat (1983) also gives information on means of accelerating this process. Use of
special form travelers with lateral main beams or self-supporting carriages (Mathivat 1983,
pp201f) will leave the bottom slab widely unobstructed and make construction easier. Secondly,
increasing the length of segments can be considered during the design phase while keeping in
mind the increasing weight and cost of bigger form travelers. Stepwise construction of the box

14
girder is also possible, with relatively simple formwork for the top slab following a few segments
behind the main form traveler. With this method less concrete has to be placed in one single step
of concrete placement. Yet it requires careful structural design of the vertically segmented
superstructure. Finally, an example of a segmental bridge is given for combination of cast-in-place
and precast segment sections. Webs of the superstructure box girder of the Brotonne Bridge in
France were precast and placed into the form travelers, which were used to fabricate the remaining
cast-in-place parts of the cross-section (Mathivat 1983).

2.4. Balanced Cantilever Construction

Balanced cantilever construction denotes building a bridge superstructure from both sides
of the pier table in a scales-like fashion. This erection method is also known under the name free
cantilever construction (Podolny and Muller 1982). Fletcher (1984) gives information that the pier
table element, serving as a base from which cantilevering is begun, is usually between 6 and 12 m
long. In order to balance the weight of both arms of the cantilever superstructure the segments will
be about equally placed at both ends. Actual placement of new segments will hardly proceed
exactly at the same times as Mathivat (1983) expresses. Therefore the pier can undergo overturning
bending moments and needs to be designed accordingly. Temporary towers with vertical
prestressing or counterweights can provide additional support. Figure 4-1 schematically shows a
typical construction stage in Balanced Cantilever Construction.

Figure 1: Balanced Cantilever Construction

15
 Balanced cantilevering can be carried out with cast-in-place or precast segments. For cast-
in-place balanced cantilevering a set of two form travelers is required, one for each arm of the
cantilever. For multi-span bridges the form travelers can be dismantled after finishing cantilevering
from one pier and can be set up for new use on the next cantilever.

In case of a bridge with variable box girder depth the pier table segment will be the most
massive segment of the superstructure. This segment needs to be constructed prior to cantilevering
to provide a working platform from which the two form travelers can start. It also includes
diaphragms that facilitate the flow of forces from the cantilever arms into the piers. Because of
size, geometry, and construction separate from the rest of the superstructure the pier table segment
will take a considerable amount of time to construct. It can be put into place either with large
precast segments or as cast-in-place with formwork mounted on the pier shaft.

An interesting pier design specifically feasible for cantilevering is mentioned by Fletcher

(1984), who points out that a pier consisting of transverse twin walls is advantageous as it provides
stability for cantilevering but allows horizontal movement of the superstructure from thermal
elongation through flexing of the wall panels.

2.5. Progressive Placement Method

The progressive placement method, in comparison with the balanced cantilevering method,
is a one-directional process as shown in Figure 4-2. All cantilever segments are subsequently
placed at the tip of a cantilever that is built across all spans. Both cast-in-place and precast
segmental construction can be used. Often stay cables from the tip of a temporary tower on the
superstructure support the cantilever. With growing cantilever superstructure this support
mechanism has to be advanced. Another method of support is use of temporary towers, which are
mentioned in Section 4.2.3.3. According to Mathivat (1983) this method is competitive for spans
between 30 to 50 m in length, whereas incremental launching and balanced cantilevering are also
used for much longer spans.

16
 Figure 2: Progressive Placement Method
Progressive placement has several advantages, as Mathivat (1983) points out. First of all,
the placement process does not have to switch sides as it occurs in the balanced cantilevering
method. Thus process control is simplified. In addition to this, good access to the placement
location is given on the already completed part of the bridge superstructure. With the progressive
placement method horizontal curves can easily be accommodated.

From a structural point of view the progressive placement method is advantageous in

substructure design. Only vertical forces from the dead load of the superstructure under
construction are experienced. In comparison with incremental launching and balanced
cantilevering, a simpler flow of forces takes place between superstructure and the piers. No
horizontal forces are introduced in the piers and no unbalanced bending moments have to be
withstood by the piers. It is therefore possible to immediately install the permanent bearings
(Mathivat 1983).

Some disadvantages of the progressive placement method need to be dealt with during
design and construction. As construction only progresses at the tip of one cantilever, progress is
slower than in balanced cantilevering. Progressive placement resembles incremental launching in
that the superstructure undergoes stresses very different from the permanent service conditions,
including even stress reversals. In both cases the structure needs to incorporate temporary
prestressing tendons to account for these stresses. Mathivat (1983) also points at the difficulty in

17
erecting the first span with progressive placement. Other construction methods may have to be
employed for this stage.

2.6. The Linn Cove Viaduct

An interesting example for use of the progressive placement method is the Linn Cove
Viaduct (Anon. 1984), shown in Figure 4-3. Its location in an environmentally sensitive area in
North Carolina, the inaccessibility of the sloping site at the mountain face, and the highly curved
alignment of the viaduct provided a set of difficult conditions for this project. Match-cast
prefabricated segments for the highly curved viaduct were delivered by truck directly to the end
of the already completed part of the structure, where they were placed and attached with temporary
thread bars and tendons. Deviating from the method outlined above, no cables were used above
the spans, but steel bents were installed under the spans during construction to provide support.
Only drilling so-called microshaft piles for the pier foundations needed to be done directly on
ground, as the piers themselves also consisted of precast segments that were craned into position
from above and post-tensioned vertically. Implementation of the progressive placement method in
connection with precast pier segments that were lowered into place from above, called top down
construction, helped protecting the natural environment at the site as far as technically possible.

Figure 3: The Linn Cove Viaduct, North Carolina, U.S. (taken from Rives 1997, p41)

18
2.7. Concluding the Cantilevering Process

The cantilevering process will finally have reached its end when both girders meet at
midspan and need to be connected. Three different ways exist to achieve this connection in the
structural system (Mathivat 1983). A hinged connection can be installed that allows horizontal
movements in the superstructure. As Mathivat (1983) writes, this system is structurally relatively
simple, yet the hinges are complicated details and the overall structural redundancy of the system
is reduced. Podolny and Muller (1982) also mention the lower ultimate load-carrying capacity of
the hinged system and the higher susceptibility to creep and relaxation phenomena. Furthermore,
the two superstructure halves can have a slight angle between them as deflections occur, which is
detrimental to the appearance of the bridge and the users comfort (Podolny and Muller 1982,
p36). Secondly, part of the midspan superstructure can be designed as a suspended span sitting on
bearings between the cantilevers. In this configuration the deflection angle between the shorter
cantilevers and the suspended span will be much smaller, and differential settling of the supports
can better be accounted for (Podolny and Muller 1982, p38). Still, the connections require special
details in the structural system.

Finally, the whole superstructure can be made continuous at midspan. Achieving this
statically indeterminate system is the most common way in building cantilever bridges for several
reasons. Mathivat (1983, p41) names specifically that the deflections in the stiffer continuous
superstructures are indeed far smaller than those met in hinged structures and both visual
appearance and drivers comfort are better than in hinged superstructures. He also notes the
necessity for expansion joints in very long continuous multi-span superstructures and advises to
provide expansion joints about 300 to 600 m apart (Mathivat 1983, p45) in points of small
moments in the superstructure. Horizontal movements of the bridge superstructure can be
accommodated by the flexibility of the piers themselves, or by using elastomeric bearings or
sliding supports (Mathivat 1983, p45).

Continuity is generated by casting a closure segment into the gap at midspan, through
which continuity tendons, the so-called integration cables as mentioned in Section 4.1 run in the
bottom part of the box girder (Mathivat 1983). Prior to casting this segment, misalignments of the
two superstructure halves are corrected with hydraulic jacks. It should, however, be tried to keep

19
these additionally imposed stresses small by paying close attention to the correct alignment
including camber when casting the superstructure halves. Additionally, as mentioned in Section
3.6.2.7, the two girders are often jacked apart to compensate for future effects of long-term creep
and shrinkage of the superstructure on the substructure as Matt et al. (1988, p37) report. They
further mention casting the closure segment of their bridge project at night to avoid problems from
temperature gradients in the superstructure. For casting and curing of the midspan closure segment
the girders need to be fixed in their position. Finally, continuity tendons can be inserted into the
newly cast segment and post-tensioned. Upon closure of midspan internal stress redistribution
takes place, shifting the moments from the supports more towards midspan. The formerly free
cantilevers are now restrained in deflection and rotation. Podolny and Muller (1982) provide a
sample calculation for the effect of stress redistribution with consideration of time-dependent
effects.

2.8. Cantilever Erection Equipment

Different erection equipment is used in bridge construction. With the specific focus on
cast-in-place and precast cantilever segmental bridges, form travelers and launching girders will
be introduced in the following sections. Other equipment that can be used for placement of precast
segments is e.g. cranes.

2.8.1. Form Travelers

Cast-in-place cantilever construction requires formwork that is attached to the tip of the
growing cantilever for casting. The following paragraphs will deal with girder cross-sections only.
As the cantilever grows the forms travel are set forth in steps. These form travelers give shape to
the segment, support the weight of the newly cast concrete until it has gained enough strength to
be post-tensioned to the previous cantilever segments, and transfer the segment weight to the
already existing superstructure. Determining for the capacity of the form travelers is the maximum
size and initial weight of the biggest segment in the bridge superstructure, including other
construction loads.

Form travelers available in todays construction industry are made by specialized

manufacturers and are reusable and very flexible (Levintov 1995) with respect to changing geometry
of the bridge superstructure and its alignment, including camber. They can be enclosed in a heated tent

20
to enable concrete placement and curing to proceed during adverse weather conditions, especially low
temperatures. In comparison with a precasting yard, form travelers often offer the less costly solution,
since transportation and storage of prefabricated segments is avoided, and they integrate all the
functions of the precasting plant into a relatively small device. Fletcher (1984) notes that by use of
form travelers the formwork is reused several times, while adjustments to variable segment geometry,
especially depth, remain possible with relatively little effort. Disadvantages of cast-in-place
cantilevering with form travelers are discussed in Section 3.6.3.1. Figure 4-4 shows a typical view
of a form traveler.

Figure 4: Form Traveler Front View

Form travelers consist of a sufficiently stiff steel frame to which form panels for the box
girder segments are attached at the front. According to Levintov (1995, p43), the steel frame is

21
mostly composed of two parallel diamond- or triangular-shaped frames that are connected and
stiffened by diagonal bracing and transverse trusses at the upper front and rear. Wheels allow
longitudinal movement of the form traveler on rails on top of the superstructure. For stability of
the traveler it can be held down by a counterweight at its rear end (Mathivat 1983). After
advancing, the traveler is anchored down to the cast parts of the superstructure with tendons at its
back, as Levintov (1995) writes. Thus, the overturning moment from the load at the traveler tip
can be resisted. It is mentioned that for structural reasons the form traveler main beams are located
above the webs of the concrete box girder, so that construction loads can be transferred into the
main load-carrying system of the bridge directly. The longitudinal main beams of the form
travelers need not necessarily be located above the webs. Form travelers are also used in
configurations with the main beams in a lateral position, leaving the bridge deck free (Podolny and
Muller 1982). Mathivat (1983) further distinguishes so-called self-supporting assemblies, where
the stiffening effect of the form panels contributes to the stiffness of the whole form traveler.

Suspended from the traveler are not only the adaptable forms for exterior and interior of
the concrete segment, but also working platforms on different levels that can be accessed from
above.

2.8.2. Launching Girders

Apart from various types of cranes that can be used to place precast segments, launching
girders are widely used for this purpose. Levintov (1995) mentions limited access under the
cantilever and great height of bridge superstructures above ground as reasons why launching
girders would be used. They are very feasible for bridges with several spans, as due to their length
they can be advanced over gaps that are still to be bridged. During construction they are moved
forward on rails whenever a major part of the bridge superstructure has been completed.

Launching girders, also called launching gantries, are large trusses that are placed
longitudinally on the bridge superstructure. One or more movable crane devices for transportation
of the precast segments can be attached to them, running along the chords of the girders. Precast
segments are delivered to the girder by special heavy-duty vehicles.

22
 If launching girders are built of high strength steel their weight can be reduced
considerably. At the same time, however, larger deflections occur that are limited by additional
support of the girders with a king post system with stay cables (Mathivat 1983).

Launching girder trusses can have triangular or rectangular cross-sections and can be
constant in depth or higher towards the middle. They can be disassembled into parts that are
connected with high-strength friction bolts (Mathivat 1983) for transportation, very similar to
tower crane booms.

Most launching girders are overhead trusses that have three leg supports. The three legs are
called rear, central, and guide leg. Some of these legs, often the guide leg, are not permanently
fixed to the girder to allow the advancing movement as will be described below. Very often these
legs form a bent above the superstructure, leaving space for the precast segments that are turned
90 sideward to be moved through the gap. Pivoting the whole launching girder around the rear
support leg (Mathivat 1983) accommodates bridge superstructure curves in the horizontal plane.

A major feature of launching girders is their length in comparison with the span length of
the bridge superstructure. Levintov (1995, p45) writes that launching girders composed of single
or double trusses may range from slightly longer than a span length to slightly longer than twice a
span length. Erection sequences for these two extremes shall be briefly described in the following
paragraphs.

2.8.2.1. Launching Girder Slightly Longer Than One Span

Construction of the bridge superstructure with a launching girder about a span long is
performed as follows. After advancement the girder rests with its rear leg on the cantilever tip at
midspan and with its center leg on the next pier. Around this pier new segments will be placed
with balanced cantilevering, filling the remaining half-span behind the pier and advancing the
cantilever to the next midspan. Afterwards, the next pier table segment is placed and the cantilever
advances half a span so that it comes to rest on guide leg and rear leg. The guide leg will remain
on the pier as the girder advances further. When the central leg arrives at this pier and the rear leg
is at midspan the next placement position has been reached. Figure 4-5 shows the construction
sequence with a launching girder that is slightly longer than one span

23
 Figure 5: Working Scheme of Short Launching Girder

A major drawback of this method is found in the previous description. The bridge superstructure
will take considerable loads during construction, since the heavy launching girder rests with one
leg at the midspan cantilever tip.

24
2.8.2.2. Launching Girder Slightly Longer Than Two Spans

Construction of the bridge superstructure with a launching girder about two spans long
does not incur the aforementioned detrimental load condition. At all stages the load-carrying girder
legs will be located above piers. This is shown in Figure 4-6. In the normal placement position the
launching girder rests with its rear leg above a previous pier and with its central leg above a free
pier. It is easily possible to also support it at the guide leg once the third pier table has been placed.
Placement of the segments will then proceed on both sides of the pier table in the middle of the
girder. To speed up construction, the girder can be equipped with two crane devices to place
segments on both sides simultaneously. After the remaining gap in the superstructure has been
closed and the cantilever has grown into the next span, the girder is advanced one complete span.
During advancement the girder rests on the guide and central leg that remain on the newly finished
pier and the one that lies ahead. This second way of employing launching girders is the more recent
technique (Mathivat 1983).

Even longer launching girders have been used in construction, as reported by Mathivat
(1983). Due to the long span required for launching girders and the mechanical parts, such as crane
and advancement devices, launching girders can be quite costly. If possible, launching girders
should be adaptable for reuse or should be rented. Launching girders can reach lengths of more
than 150 m, depending on the requirements of the bridge spans, and weights of up to about 400 t
(Mathivat 1983). In addition to these specialized, expensive and heavy pieces of construction
equipment it is also possible to use simple lifting devices that are located at the cantilever tip. In
the case of the Linn Cove Viaduct, which has been presented in Section 4.2.1.4, a derrick was
mounted to the bridge superstructure that placed segments as they were delivered by truck (Anon.
1984). Other types of deck-mounted equipment are imaginable and mentioned by Levintov (1995,
p44), e.g. a longitudinal beam fitted with lifting tackle and winches.

25
Figure 6: Working Scheme of Long Launching Girder

26
 CHAPTER 3

INCREMENTAL LAUNCHING

Incremental launching was developed by the German engineers Fritz Leonhardt and Willi
Baur for the Rio Caron Bridge in Venezuela (Podolny and Muller 1982), which was built from
1962 to 1964. The incremental launching technique, as opposed to other methods presented in this
chapter, consists of casting a continuous chain of segments at one particular location on site and
then pushing the growing superstructure out over site to be bridged. A casting bed with adjustable
formwork for the superstructure segments is set up. This casting bed can also be enclosed in a
heated tent so that controlled casting and curing conditions are achieved. The normal cycle time,
regardless of segment length is one week. Segment lengths according to Liebenberg (1992)
typically range between 15 and 30 m.
Two different techniques for launching the bridge superstructure from the casting bed exist.
Hydraulic jacks can pull the superstructure with steel rods, as it was done for the Rio Caron Bridge
(Podolny and Muller 1982). The second, more common method is to employ a pair of hydraulic
jacks acting vertically and horizontally. Continuous repetition of lifting the superstructure off the
abutment and then pushing it forward as far as the jack allows will achieve the launching in
incremental steps. Figure 4-7 shows the launching process schematically. Podolny and Muller
(1982) caution to design the jack capacity for more than the usual friction coefficient of 2 %
because of imperfections that can occur during construction.

27
Figure 7: Incremental Launching

28
 In front of the cantilevering superstructure a lightweight steel launching nose is attached
with tendons that reaches the next support before the bridge superstructure itself arrives. Its
purpose is to keep the bending moments in the superstructure smaller. Mostly the launching nose
has a length of about 60 % of the bridge spans (Podolny and Muller 1982). Another way of
reducing the bending moments is to implement temporary towers between the bridge piers. These
towers need to be able to take the horizontal forces that arise from launching.

On top of all supports, including abutments, piers, and temporary towers temporary sliding
bearings are installed during construction that will later be replaced with the permanent ones.
Stainless steel plates are installed on the bearings. While the superstructure is advanced, Neoprene
pads coated with Teflon and reinforced with steel plates are inserted between concrete and steel to
reduce friction (Liebenberg 1992). Very low friction coefficients of 2% or less can be achieved
with this method.

Several advantages make incremental launching a very competitive erection method. As with
any cantilevering method it leaves the site below completely unobstructed during construction. Only
for very long spans temporary towers or cable stays from above as supports are needed. Except for
these the equipment necessary is reduced to the jacking mechanism, the adjustable stationary casting
bed, and temporary sliding bearings, all of which may possibly be reused, which reduces the capital
investment considerably. Podolny and Muller (1982) furthermore mention the cost savings due to
avoidance of segment transportation and heavy construction equipment. They also point at less
maintenance cost due to the higher prestressing of the superstructure. The controlled casting and curing
conditions allow steady and quick construction progress.

Bridges that are erected with the incremental launching method should, according to
Podolny and Muller (1982), have a constant cross-section, especially in depth, and have a straight
superstructure. It is possible to accommodate small variations in alignment and horizontal and
vertical curvatures provided that they have a constant radius. Close control of the bridge geometry
during casting and launching is very important. Sloping grades at the bridge site are also
accommodated, in this case the launch is usually in the downward direction, more than 2 %
slope would require a retarding mechanism to stop the movement of the superstructure (Liebenberg
1992, p165)

29
 Liebenberg (1992, p164) also gives a very clear statement of the main difficulty of the
incremental launching method: During launching, the section undergoes complete stress reversals
as it progresses from a cantilever to the first support and thereafter over the following spans to its
final position. Clearly, this erection sequence generates a bending moment envelope in the
structure depending on the span lengths that needs to be accounted for in designing the cross-
section properties and the amount of reinforcement and prestressing tendons. The stresses due to
the aforementioned high bending moments require much longitudinal prestressing both at top and
bottom of the cross-section. Another disadvantage is the large workspace that is needed for the
casting bed at the abutment and the adjacent storage areas (Podolny and Muller 1982).

The Aichtal Bridge in Germany that was built mainly between 1981 and 1983 serves as a
good example of the incremental launching method. According to Basse et al. (1985) this bridge
with its total length of 1,161 m is the longest one ever built with incremental launching. It crosses
two valleys at a maximum of 48 m and 50 m above ground, respectively. A fixed bearing is located
at the pier between the two valleys. At the same location a second jacking system was installed for
use in later construction stages.

The normal pier spacing for the 21 spans of the Aichtal Bridge is 51 m, reaching a maximum of
80 m and 84 m respectively at the deepest parts of the valleys. These wide spans required use of
temporary towers that were braced with stay cables from the ground to resist the horizontal forces
from launching. The whole bridge superstructure consists of two parallel single cell box girders
that are 3.50 m deep, 5 m wide at the soffit and carry 13.50-m wide decks. After completion of
one girder all construction equipment was relocated for the second box girder.

An enclosed 25.50-m long casting bed with adjacent assembly yard for the reinforcement
cages was erected behind the abutment with the launching jacks. For winter construction work
another 50-m long tent with large heaters was set up for proper curing and the piers were built with
thermally insulated climbing formwork. Casting of segments was done in a weekly cycle. Both
longitudinal and transverse limited prestressing was implemented. Basse et al. (1985, p23) give
information on the sequence of casting steps that is compiled in Table 4-3:

30
 Table 2: Sequence of Casting Steps for Aichtal Bridge

Day Activities
Monday Post-tensioning, stripping of formwork, incremental launching of segment
Tuesday Placement of reinforcement cages for bottom slab and webs,
Installation of ducts and tendons, installation of interior web formwork
Wednesday Completion of formwork installation, concreting of bottom slab and webs
Thursday Removal of web formwork, installation of interior top slab formwork,
Placement of top slab reinforcement, ducts, and tendons
Friday Concreting of top slab
Saturday Curing of concrete
and Sunday

Launching of the cured and post-tensioned segments required many personnel for
supervision of all sliding bearings. An overall longitudinal slope of the bridge reduced the jacking
forces necessary for launching. An overall curvature with radius 1,500 m in the horizontal plane
and a constant 3.5 % cross slope of the bridge superstructure induced stresses due to restrained
deformations during launching. Furthermore, the changing span lengths had to be considered in
coming up with the prestressing program to optimize the tendon profile and prestressing forces.

A surveying program had been prepared to control geometry during all construction steps.
Several fixed surveying points were located along the site. Overall, tolerances for deviations
from the planned bridge geometry during casting and launching were less than 1.5 mm (Basse
et al. 1985).

31
 CHAPTER 4
FALSEWORK

Falsework has been used in construction since ancient times, when Roman bridge builders
erected their semicircular stone arches for bridges, aqueducts, and vaults on wooden centering.
Falsework provides continuous support for the formwork that gives shape to the superstructure.
In most cases falsework is used for cast-in-place concrete structures. It requires firm, relatively
even ground on which it can be erected.

4.1. Stationary Falsework

Apart from custom-built timber structures a wide range of steel elements and modules
for falsework is available in the construction industry. Liebenberg (1992) gives a range of up to
300 m in length and about 10 m in height for bridges to be built with this method. He also
specifically points at the necessity for stable foundations of the falsework, sufficient bracing of
the falsework structure, and consideration of the deflection of the falsework in the overall
superstructure camber.
Falsework, either stationary or traveling, can also be configured as casting girders that
hold the formwork into which the concrete is placed. In that, these girders resemble the erection
girders of the span-by-span method, which are used to assemble precast segments. Liebenberg
(1992) further distinguishes the girders depending on their location to the bridge superstructure
as overhead or supporting it from below, or combination of both. He also reminds that use of
major pieces of equipment, such as casting girders needs to be considered carefully because of
the high capital investment that is necessary.
Stationary falsework is the simplest method of erecting the bridge superstructure.
Advantages mentioned by Liebenberg (1992) are that stationary falsework can be erected by less
specialized workers to any desired shape. Use of modern standard elements of which the falsework
is put together allows uncomplicated erection. There are, however, several disadvantages related
to stationary falsework. A lot of material is required for stationary falsework, which requires much
time and manual labor to be spent on its erection, in addition to the cost of purchasing or renting
the materials themselves. Therefore, Liebenberg (1992) concludes that it needs to be erected some
spans in advance to keep up with rates of placement of concrete that can be achieved.

32
 Today, falsework is competitively used e.g. for the complex alignments of highway
interchanges, as Cassano (1987) reports for the example California. He states that apart from the
feasibility for the highly curved superstructures, use of major falsework systems also allows
placement of very large volumes of concrete at the same time and thus speeds up construction.
Cassano (1987)mentions that all falsework in California has to be designed and built according to
the Falsework Manual of the California Department of Transportation and is reviewed and
inspected by qualified engineering personnel.

4.2. Traveling Falsework

Traveling falsework alleviates some of the problems associated with stationary formwork.
This type of falsework, including formwork, is assembled to larger units that can be moved to the
next span to be cast. As with stationary falsework, this method requires the site to have relatively
level and firm ground to allow movement of the falsework on a wheel assembly. In case the ground
conditions are less favorable, using the span-by-span method might be advisable, where the
superstructure segments are assembled on erection girders. Span-by-span erection is introduced in
the following Section 4.2.4. Traveling falsework is shown in Figure 4-8

4.2.1. Temporary Towers

A kind of falsework used frequently in construction is the use of temporary towers to

support the superstructure that is under construction at intermediate positions. These towers can
e.g. be used in the balanced cantilevering method to stabilize the structure against tipping over
from construction loads. They can be additionally strengthened with prestressing rods to withstand
the forces that are imposed on them during construction. Other applications are found during
incremental launching, where together with the launching nose they serve to keep the range of
bending moments small. Instead of using temporary towers as supports Liebenberg (1992, p156)
also mentions stiffening girders by means of prestraining with a king post and adjustable inclined
ties.

4.2.2. Span-By-Span Erection

Levintov (1995, p47) lays out the characteristics of the span-by-span erection method as
assembling all segments for a span in a set, which is then aligned, jointed, and longitudinally

33
post-tensioned together to make a complete span. The principle of span-by-span erection is shown
in Figure 4-9. Span-by-span erection is typically limited to bridges that consist of box girders with
constant depth. The actual construction can have several variants, the segments can be assembled
on the ground and lifted in place as a group by a heavy-duty crane or they can all be put into their
final position on erection girders along the spans to be completed. The second method was e.g.
used for constructing the Biloxi Interstate I-110 Viaduct. In this project different types of erection
girders were used. The authors report that on some spans triangular trusses were implemented, and
on the other hand steel box girders came into use, which left more clearance for traffic underneath
(Phipps and Spruill 1990). Erection girders were supported at their ends by steel falsework resting
on the footings at each pier (Phipps and Spruill 1990, p130). After completion of a span the
erection girders were set forward to the next span and adjusted. By then, the precast segments for
this span had already been supplied and would be lifted in place by crane. Fine adjustments of the
segments on the erection girders were possible by means of variable individual supports. Finally,
post-tensioning would be performed to link all the segments together to form a complete span. The
structure became self-supporting after casting of the closure joints with the pier table segments
had been done and the longitudinal prestressing force was induced. With the method described, an
erection speed of one span per about 3.5 days could be achieved (Phipps and Spruill 1990).
Erection girder need not rest on the ground, but can also be supported by already existing
substructure or superstructure, e.g. the piers of a span that is to be constructed. Project specific
design of substructure and superstructure and considerations as e.g. for traffic clearances set the
boundaries for erection with erection girders

34
 Figure 8: Span-By-Span Erection

4.3 Consideration of Construction loads and stresses

The following sections deal with the relationship between construction loads and the
stresses that these induce in structures and the structures themselves while they are under
construction and still awaiting completion. The central issue for all considerations is structural
safety, meaning failure against structural failure. The generic concept of resistance R that is greater
than the most unfavorable combination of load S that induces stresses has been introduced in
Section 3.5.2.

Codes require that all construction influences will be properly taken into account during
design. Even a professional code applicable for bridges (ACI 1995, p51) in its Section 5.3 only
points out that Consideration should be given to temporary loads caused by the sequence of
construction stages, forming, falsework, or construction equipment and the stresses created by
lifting and placing precast members. It assigns the responsibility for the construction scheme,
which imposes stresses on the structural members, to the contractor. It is further pointed out that
the stability of precast members and prestressing should be taken into account. Environmental
loads should be considered during construction using an appropriate return period or reduced
severity. Lower load factors may be used to account for the acceptability of higher temporary stress
levels (ACI 1995, p51). The aforementioned load factors are provided in tables in the same code.

How the actual process has to be accomplished in detail remains the structural engineers
task. Close cooperation of designer and contractor in development of the construction sequence
contributes to quality.

35
For better understanding of the relationship between structures under construction and actions
influencing them it is useful to get a clear definition of the technical terms related to this topic.
The nature of actions has been explained in Section 3.5.3 as any loads or restrained deformations
that can cause stresses within the structural system. The term construction loads should in this
context be understood as the broader sense of any actions that occur during construction of the
structure prior to normal service conditions.

Then again, the concept of construction loads needs to be extended by consideration of the
uncompleted structure during construction, which may not have reached full resistance to imposed
actions.

The term erection method denotes the physical means of putting the foundations, bridge
substructure, and especially its superstructure into place. Every type of erection method requires
certain equipment and site installations to carry out the work tasks. The erection methods goes
along with specific limitations imposed on the flow of work tasks to be scheduled, e.g. that bridge
piers have to be finished prior to begin of incremental launching of a bridge superstructure.

The construction sequence is the specific succession of work tasks for one particular
project developed under consideration of the erection method chosen for its conditions and
restraints for economic construction Construction stages are notable steps within the progression
of work tasks from the initial site operations startup to the finished structure. These steps can be
distinguished by the appearance of the structural system, loading conditions, or other factors.

Load steps can be defined as specific sets of loads on a structure in its current construction
stages. These load steps or load cases are combinations of actions that are anticipated to occur at
the same time with a certain probability, and are incorporated into analytical calculations with
partial factors of safety.

4.4. Types of Construction Loads and Influences

Generally, construction loads are by nature of relatively short duration in comparison with
the overall planned service life of a structure. Construction loads may influence a structure over a
brief time only, e.g. from equipment or material for a new segment that is temporarily stored on
an already completed part of the superstructure of a bridge. Construction loads can affect the
structure in very unfavorable conditions, e.g. when a crane is located at the tip of a cantilever to

36
place segments. Thus resulting stresses in the structure can even exceed stresses due to permanent
and dynamic loads under service.
Real loads are generally distinguished into two classes, dead loads and live loads. During
construction the structure has to carry its own weight, its dead load, and the superimposed dead
loads of bridge parts that are not structurally important but necessary for service, as e.g. the bridge
furniture, the so-called accessories.
Many different live loads influence the structure. In most of all cases, live loads are
idealized either as uniformly loaded areas on the superstructure or as singular loads from larger
pieces of equipment. Live loads can result from erection equipment, e.g. the launching nose in
incremental launching and lifting devices placed on the structure such as cranes and launching
girders. Forces are also imposed on the structure through restraints from fixed bearings during
construction, e.g. on piers for cantilevering. Along with these structural details, the boundary
conditions can still change; e.g. considerable settlements can occur when the soil is initially loaded.
After having considered the loads and restraints, i.e. all actions, two elements remain to be
discussed. The geometry of the structural system itself is developing continuously with progress
of construction depending in a manner that depends on the erection method and needs to be
considered. Incomplete structures, e.g. the cantilever beam prior to midspan closure for continuity
are inherently weaker than in the finished state because of less structural redundancy. All these
different construction stages need to be analyzed as a combination of load cases and the resistance
of the structure at that construction stage for structural safety.
Apart from that structural resistance the material resistance may also be weaker than in the
final state. Especially for cast-in-place segmental bridges the still young concrete usually has not
developed its full specified strength when it is being prestressed and loaded with more segments
for quick erection. The structural resistance and material resistance can also be understood as the
two components of structural safety, namely strength and stability, as outlined in Section 3.1.3.
Summarizing, stresses induced by construction loads may be higher than those from service loads
as the incomplete structural system is mostly different and weaker than finished structures,
concrete has not gained full strength, and the boundary conditions may be different from the
service state. In other words, the great importance of construction stages lies in the criticality that
results from the still low structural and material resistance, while loads may be actually more
adverse and boundary conditions different.

37
4.5. Segment Casting Cycle
Based on Mathivat Construction of the superstructure of Wilson Creek Bridge closely
followed the duration for the individual steps given by Mathivat (1983), with the difference of
allowing three days instead of two for installation of reinforcement and prestressing tendons and
reducing curing duration to two days. With respect to the overall duration of one casting cycle,
British engineer Fletcher (1984, p17) gives information that it generally settles down to one
pour [i.e. segment placement] per shutter [form traveler] per week. He also names important
factors that influence casting cycle duration:

A learning effect takes place due to the repetitive nature of the segmental work tasks;

Speed of reinforcement and tendon installation and speed of concrete placement depend on the
size of the segment;

Form traveler adjustment takes longer when changes in segment geometry, e.g. variation in the
box girder depth or slab width occur;

Form traveler adjustment takes longer for incorporation of special details, e.g. diaphragms and
tendon anchorages, due to additional formwork and reinforcement installation;

Adverse conditions, such as inclement weather (e.g. low temperatures) can delay sufficient
gain of compressive strength of the concrete, which provides the lower limit for the casting
cycle duration. The quality of each individual segment in the cantilevering chain of cast-in-
place segments is critical to the structural integrity of the entire superstructure.

The casting cycle for Wilson Creek Bridge can be broken down into eight major steps as
shown in Figure 6. In Step 1 the form traveler that is shown in Figure 7 is detached from its rails,
which are then moved forward, anchored firmly with hydraulic tie-downs, and the form traveler is
set back onto the rails. In Step 2, the form traveler with its exterior forms is advanced with
hydraulic jacks, brought into alignment according to the calculated camber data, and is anchored
in the new position over the end of the cantilever, as shown in Figure 8. Once the form traveler
has been set up, a work crew can manually install reinforcement in the bottom slab and webs in
Step 3, as the form traveler framework and bracing does not provide space to lift preassembled
reinforcement cages into place. Also, a carpenter builds a custom-made plywood bulkhead that
will close the formwork at its front face during casting and curing. Step 4 comprises of advancing
the interior formwork as well, which would previously have obstructed the reinforcement

38
installation. The bulkhead, interior formwork, and exterior top slab formwork can be seen in Figure
9. To keep both forms together, wall ties are installed. Finally, the reinforcement for the top slab
is put in place. In Step 5 the reinforcement crew installs and couples ducts for the longitudinal and
transverse prestressing tendons and inserts the tendons into the ducts. In Step 6 the concreting crew
places concrete into the bottom slab, the webs, and the top slab of the new segment. Concrete is
placed through tremie pipes that run through openings in the formwork panels. The individual
concrete lifts are consolidated using internal vibrators. During Step 7 the concrete is curing to
achieve sufficient compressive strength and durability, so that it can withstand the construction
loads and the in-service loads as well as environmental influences. Finally, transverse tendons in
the top slab and the longitudinal cantilever tendons are stressed and their ducts are filled with grout
in Step 8, the forms will subsequently be stripped from the segment, and a new casting cycle can
begin.
The above mentioned casting cycle was repeated a dozen times on both cantilever arms, as
shown in Figure 10, until midspan had been reached and the closure segment was placed, as shown
in Figure
Only one set of two form travelers was used in this project for economical reasons. Thus,
cantilevering was performed subsequently about all pier tables in the order of Pier 2 Pier 3 Pier
4 Pier 1 to construct the bridge superstructure. However, work tasks were not performed
concurrently on both cantilever arms, as mentioned above. Figure 12 provides a linear schedule
for the beginning cantilevering operation about Pier 2. For both cantilever arms, which were
referred to as the surveying Downstation (in the direction of Abutment A) and Upstation (in
the direction of Abutment B), it clearly shows the curing time for newly placed segments before
the casting cycle begins anew. Construction of each segment is shown as a continuous operation
to keep this schematic linear schedule easy to read; in reality each segment involves many
individual tasks like formwork setup, reinforcement installation, and concrete placement, as
described above. Note also the planned lag of two days between repetitions of identical operations
for pairs of equivalent segments on both sides, like Segments 1 and 2, Segments 3 and 4, etc. This
time lag allowed optimized allocation of work forces by reducing lead-time for each crew, as the
crews for formwork setup, reinforcement installation, and concrete placement alternated between
the cantilever arms. Some reduction of the duration of work activities could be observed because
of (a) the reduction of segment dimensions, i.e. less volume of concrete per segment towards
midspan and (b) due to a learning effect from the repetitive nature of the casting cycle (Sean Bush,

39
personal communication, September 17, 1999). A schematic of the cantilevering operations that
combines the information from Table 1 and

40
 CONCLUSION

This report provides an introduction to constructability concepts in segmental bridge

superstructure erection. The various influences that arise from the chosen erection method, the
bridge geometry, and the materials behavior and their complex interactions were outlined focusing
on Balanced Cantilever Construction. Short-term effects and long-term effects were introduced
and put in relation with loading conditions and environmental influences in a comprehensive
diagram.

Construction loads depend on the actual erection method and may create higher stresses in
the structure than any loads anticipated for the bridge under service conditions could cause.
Furthermore, construction loads affect the structure in its weak stages prior to completion the
structural system has not yet reached continuity and additional redundancy from being statically
indeterminate in its final configuration. During construction the segments may have reached only
a minimum level of compressive strength that may be less than the specified 28-day compressive
strength when successor segments are added. Consideration of all construction stages with their
respective geometry and boundary conditions, structural details, time-dependent material
properties, and construction loads therefore is essential for adequately analyzing the structure and
designing against failures.

41
 Providing the case study of the Wilson Creek Bridge in Virginia, which was erected using
Balanced Cantilever Construction, enhanced the concepts outlined in this paper and gave insight
into how these concepts are applied in real engineering practice. The segment casting cycle with
its individual steps was explained and illustrated in detail. Finally, two constructability changes
that the contractor implemented on the Wilson Creek Bridge to optimize operations and reduce
schedule duration were presented. In particular, a redesign of the segment length in conjunction
with a planned asymmetry of the pier tables considerably reduced overturning moments from
cantilevering operations of the superstructure.
Future research could include examination of other erection methods, such as e.g.
Incremental Launching and analyze these with respect to constructability issues. Providing a
collection of several different construction methods, each illustrated with case studies of real-life
examples would be a valuable source of information in teaching future construction engineers.
This can ultimately lead to yet safer and economical construction of bridges to serve the public.

42
 REFERENCES

1. Podolny, W., Muller, J. M. (1982). Construction and Design of Prestressed Concrete

Segmental Bridges. A Wiley-Interscience Publication, John Wiley & Sons, Inc., New York,

NY.

2. Fletcher, M. S. (1984). In-Situ Free Cantilever Concrete Bridges. Highways and

Transportation
3. Technical, Longman Group UK Limited, Burnt Mill, Harlow, Great Britain.

4. Lucko, G. (1999). Means and Methods Analysis of a Cast-In-Place Balanced Cantilever

Segmental Bridge: The Wilson Creek Bridge Case Study. Thesis submitted to the Faculty

of Virginia Polytechnic Institute and State University in partial fulfillment of the

requirements for the degree of Master of Science, Blacksburg, VA.

5. Mathivat, J. (1983). The Cantilever Construction of Prestressed Concrete Bridges. A

Wiley-Interscience Publication, John Wiley & Sons, Inc., New York, NY

43