THE 2005 FHWA CONFERENCE

Integral Abutment and Jointless Bridges (IAJB 2005)

March 16 18, 2005

Baltimore, Maryland

Organized by: Constructed Facilities Center

College of Engineering and Mineral Resources West Virginia University

Conference Sponsors: Federal Highway Administration USDOT

West Virginia Department of Highways - WVDOT

ii

iii

TABLE OF CONTENTS

Session I: Current Practices with Design Guidelines and Foundation Design Integral Abutment and Jointless Bridges V. Mistry 3 Integral Abutments and Jointless Bridges (IAJB) 2004 Survey Summary R. Maruri, S.Petro 12 The In-Service Behavior of Integral Abutment Bridges: Abutment-Pile Response R. Frosch, M. Wenning, V. Chovichien 30 New York State Department of Transportation's Experience with Integral Abutment Bridges A. Yannotti, S. Alampalli, H. White 41 Integral Abutment Design and Construction: The New England Experience D. Conboy, E. Stoothoff 50 VDOT Integral Bridge Design Guidelines K. Weakley 61 Session II: Case Studies Case Study: A Jointless Structure to Replace the Belt Parkway Bridge Over Ocean Parkway S. Jayakumaran, M. Bergmann, S. Ashraf, C. Norrish 73 Case Study Jointless Bridge Beltrami County State Aid Highway 33 Over Mississippi River in Ten Lake Township, Minnesota J. Wetmore, B. Peterson 84 Design and Construction of Dual 630-foot, Jointless, Three-span Continuous Multi-girder Bridges in St. Albans, West Virginia, United States, Carrying U.S. Route 60 over the Coal River J. Perkun, K. Michael 97 Integral Abutment Bridges with FRP Decks Case Studies V. Shekar, S. Aluri, H. GangaRao 113

iv

New Mexicos Practice and Experience in Using Continuous Spans for Jointless Bridges S. Maberry, J. Camp, J. Bowser 125 Integral Abutment Bridges Iowa and Colorado Experience D. Liu, R. Magliola, K. Dunker 136 Moose Creek Bridge Case Study of a prefabricated Integral Abutment Bridge in Canada I. Husain, B. Huh, J. Low, M. McCormick 148 Session III: Maintenance and Rehabilitation Field Data and FEM Modeling of the Orange-Wendell Bridge C. Bonczar, S. Brea, S. Civjan, J. DeJong, B. Crellin, D. Crovo 163 Integral Abutment Pile Behavior and Design Field Data and FEM Studies C. Bonczar, S. Brea, S. Civjan, J. DeJong, D. Crovo 174 Effects of Restraint Moments in Integral Abutment Bridges M. Arockiasamy, M. Sivakumar 185 Full-Scale Testing of an Integral Abutment Bridge S. Hassiotis, J. Lopez, R. Bermudez 199 Analysis and Design of Integral Abutment by LRFD Method Y. Deng, J. Farre, J. Chang, P. Penafiel 211 Behavior of Pile Supported Integral Abutments E. Burdette, S. Howard, E. Ingram, J. Deatherage, D. Goodpasture 222 Soil Structure Analysis of Integral Abutment Bridges P. Christou, M. Hoit, M. McVay 233 Behavior of Two-Span Integral Bridges Unsymmetrical About the Pier Line D. Knickerbocker, P. Basu, E. Wasserman 244 Session IV: Construction Practices Field Study of Integral Backwall with Elastic Inclusion E. Hoppe 257

v

Plastic Design of Steel HP-Piles for Integral Abutment Bridges P. Huckabee 270 Integral-Abutment Bridges: Geotechnical Problems and Solutions Using Geosynthetics and Ground Improvement J. Horvath 281 P-y Curves from Pressuremeter Testing at Kings Creek Bridge, WV Route 2, Hancock County, West Virginia W. Kutschke, B. Grajales 292 Effective Temperature and Longitudinal Movement in Integral Abutment Bridges R. Oesterle, J. Volz 302 Transverse Movement in Skewed Integral Abutment Bridges R. Oesterle, H. Lotfi 312 Soil-Structure Interaction of Jointless Bridges O. Kerokoski, A. Laaksonen 323 Author Index 337

vi

1

SESSION I: CURRENT PRACTICES WITH DESIGN GUIDELINES AND FOUNDATION DESIGN

2

3

Integral Abutment and Jointless Bridges

Vasant C. Mistry; Federal Highway Administration; Washington, D. C.

ABSTRACT The most frequently encountered corrosion problem involves leaking expansion joints and seals that permit salt-laden run-off water from the roadway surface to attack the girder ends, bearings and supporting reinforced concrete substructures. Because neither the materials used nor the pains taken to mitigate joint leakage can fully resolve these problems, other options such as, the construction of jointless bridges, the use of integral or semi-integral abutments, and moving the joints beyond the bridges should be sought. Since 1987, numerous States have adopted integral abutment bridges as structures of choice when condition allow. At least 40 States are now building some form of jointless bridges. While superstructures with deck-end joints still predominate, the trend appears to be moving toward integral. This paper presents some of the important features of integral abutment and jointless bridge design and some guidelines to achieve improved design. The intent of this paper is to enhance the awareness among the engineering community to use Integral Abutment and Jointless Bridges wherever possible. WHY JOINTLESS BRIDGES?

One of the most important aspects of design, which can affect structure life and maintenance costs, is the reduction or elimination of roadway expansion joints and associated expansion bearings. Unfortunately, this is too often overlooked or avoided. Joints and bearings are expensive to buy, install, maintain and repair and more costly to replace. The most frequently encountered corrosion problem involves leaking expansion joints and seals that permit salt-laden run-off water from the roadway surface to attack the girder ends, bearings and supporting reinforced concrete substructures. Elastomeric glands get filled with dirt, rocks and trash, and ultimately fail to function. Many of our most costly maintenance problems originated with leaky joints.

Bridge deck joints are subjected to continual wear and heavy impact from repeated live loads as well as continual stages of movement from expansion and contraction caused by temperature changes, and or creep and shrinkage or long term movement effects such as settlement and soil pressure. Joints are sometimes subjected to impact loadings that can exceed their design capacity. Retaining hardware for joints are damaged and loosened by snowplows and the relentless pounding of heavy traffic. Broken hardware can become a hazard to motorists,

4

and liability to owners. Deck joints are routinely one of the last items installed on a bridge and are sometimes not given the necessary attention it deserves to ensure the desired performance. While usually not a significant item based on cost, bridge deck joints can have a significant impact on a bridge performance. A wide variety of joints have been developed over the years to accommodate a wide range of movements, and promises of long lasting, durable, effective joints have led States to try many of them. Some joint types perform better than others but all joints can cause maintenance problems.

Bearings also are expensive to buy and install and more costly to replace. Over time steel bearings tip over and seize up due to loss of lubrication or buildup of corrosion. Elastomeric bearings can split and rupture due to unanticipated movements or ratchet out of position.

Because of the underlying problems of installing, maintaining and repairing deck joints and bearings, many States have been eliminating joints and associated bearings where possible and are finding out that jointless bridges can perform well without the continual maintenance issues inherent in joints. When deck joints are not provided, the thermal movements induced in bridge superstructures by temperature changes, creep and shrinkage must be accommodated by other means. Typically, provisions are made for movement at the ends of the bridge by one of two methods: integral or semi-integral abutments, along with a joint in the pavement or at the end of a reinforced concrete approach slab. Specific guidelines for designing and detailing jointless bridges have not yet been developed by AASHTO so the States have been relying on established experience

A 1985 FHWA report on tolerable movement of highway bridges examined 580 abutments in 314 bridges in the United States and Canada. Over 75 percent of these abutments experienced movement, contrary to their designers intent, typically much greater movement vertically than horizontally. The following paragraph is from the report.

The magnitude of the vertical movements tended to be substantially greater

than the horizontal movements. This can be explained, in part, by the fact that in many instances the abutments moved inward until they became jammed against the beams or girders, which acted as struts, thus preventing further horizontal movements. For those sill type abutments that had no backwalls, the horizontal movements were often substantially larger, with abutments moving inward until the beams were, in effect, extruded out behind the abutments.

The use of expansion joints and bearings to accommodate for thermal

movements does not avoid maintenance problems; rather, the provision to these items can often facilitate such problems.

5

In this 40-year national experience, many savings have been realized in initial construction costs by eliminating joints and bearings and in long-term maintenance expenses from the elimination of joint replacement and the repair of both super and substructures. Designers should always consider the possibilities of minimum or no joint construction to provide the most durable and cost-effective structure. Steel superstructure bridges up to 400 ft. long and concrete superstructure bridges up to 800 ft. long have been build with no joints, even at the abutments.

The impact on the total project cost and quality is best illustrated by the figure shown on the right. As is seen, the decisions made at the design stage account for over 80 percent of the influence on both cost (first and life-cycle) and quality (service life performance) of the structure. Decisions made in the initial stages of design establish a program that is difficult and costly to change once detailed design or construction begins.

The following quote is very appropriate for bridge engineering:

Quality is never an accident. It is always the result of high intention, sincere effort, intelligent direction, and skillful execution. It represents the wise choice of many alternatives. This is especially true when the Engineer begins the task of planning, designing and detailing a bridge structure. The variables are many, each of which has a different, first and life cycle, cost factor. The question to be asked continuously through the entire process is what value is added if minimum cost is not selected? Another question to be asked is what futures should be incorporated in the structure to reduce the first and life cycle cost and enhance the quality? Most of the variables are controlled by the designer. These decisions influence the cost and quality of the project; for better or for worse! WHAT IS AN INTEGRAL ABUTMENT BRIDGE?

Integral abutment bridges are designed without any expansion joints in the bridge deck as shown by the figures on the right. They are generally designed with the stiffness and flexibilities spread throughout the structure/soil system so that all supports accommodate the thermal and braking

6

loads. They are single or multiple span bridges having their superstructure cast integrally with their substructure. Generally, these bridges include capped pile stub abutments. Piers for integral abutment bridges may be constructed either integrally with or independently of the superstructure. Semi-integral bridges are defined as single or multiple span continuous bridges with rigid, non-integral foundations and movement systems primarily composed of integral end diaphragms, compressible backfill, and movable bearings in a horizontal joint at the superstructure-abutment interface. WHY INTEGRAL ABUTMENTS?

As stated earlier, integral abutment and jointless bridges cost less to construct

and require less maintenance then equivalent bridges with expansion joints. In addition to reducing first costs and future maintenance costs, integral abutments also provide for additional efficiencies in the overall structure design. Integral abutment bridges have numerous attributes and few limitations. Some of the more important attributes are summarized below. Simple Design- Where abutments and piers of a continuous bridges are each supported by a single row of piles attached to the superstructures, or where self-supporting piers are separated from the superstructure by movable bearings, an integral bridge may, for analysis and design purposes, be considered a continuous frame with a single horizontal member and two or more vertical members. Jointless construction - Jointless construction is the primary attribute of the integral abutment bridges. The advantages of jointless construction are numerous as has been stated earlier. Resistance to pressure - The jointless construction of integral bridges distributes longitudinal pavement pressures over a total superstructure area substantially greater than that of the approach pavement cross-section. Rapid construction - Only one row of vertical piles is used, meaning fewer piles. The back wall can be cast simultaneously. Fewer parts are required. Expansion joints and bearings are not needed. The normal delays and the costs associated with bearings and joints installation, adjustment, and anchorages are eliminated.

7

Ease in constructing embankments - Most of the embankment is done by large earth moving and compaction equipment requiring only little use of hand operated compaction equipment. No cofferdams - Integral abutments are generally built with capped pile piers or drilled shaft piers that do not require cofferdams. Vertical piles (no battered piles) - At abutment a single row of vertical piles is used. Simple forms - Since pier and abutment pile caps are usually of simple rectangle shape they require simple forms. Few construction joints are required in the integral abutment bridges, which results in rapid construction. Reduced removal of existing elements - Integral abutment bridges can be built around the existing foundations without requiring the complete removal of existing substructures. Simple beam seats - Preparation of load surface for beam seat can be simplified or eliminated in integral bridge construction. Greater end span ratio ranges - Integral abutment bridges are more resistant to uplift. The integral abutment weight acts as a counterweight. Thus, a smaller end span to interior span ratio can be used without providing for expensive hold-downs to expansion bearings. Simplified widening and replacement - Integral bridges with straight capped-pile substructures are convenient to widen and easy to replace. Their piling can be recapped and reused, or if necessary, they can be withdrawn or left in place. There are no expansion joints to match and no difficult temperature setting to make. The integral abutment bridge acts as a whole unit. Lower construction costs and future maintenance costs. Improved ride quality - Smooth jointless construction improves vehicular riding quality and diminishes vehicular impact stress levels. Design efficiency - Design efficiencies are achieved in substructure design. Longitudinal and transverse loads acting upon the superstructure may be distributed over more number of supports.

8

For example, the longitudinal load distribution for the bent supporting a two span bridge is reduced 67 percent when abutments are made integral instead of expansion. Depending upon the type of bearings planned for expansion abutments, transverse loadings on the same bent can be reduced by 67 percent as well. Added redundancy and capacity for catastrophic events - Integral abutments provide added redundancy and capacity for catastrophic events. Joints introduce a potential collapse mechanism into the overall bridge structure. Integral abutments eliminate the most common cause of damage to bridges in seismic events, loss of girder support. Integral abutments have consistently performed well in actual seismic events and significantly reduced or avoided problems such as back wall and bearing damage, associated with seat type jointed abutments. Jointless design is preferable for highly seismic regions. Improve Load distribution - Loads are given broader distribution through the continuous and full-depth end diaphragm. Enhance protection for weathering steel girders Tolerance problems are reduced or eliminated - The close tolerances required with expansion bearings and joints are eliminated or reduced with the use of integral abutments. RECOMMENDED BEST PRACTICES

The following best practices are believed to contain the key elements to ensure quality improvements in designing and constructing Integral Abutment and Jointless Bridges. Develop design criteria or office practices for designing integral abutment

and jointless bridges. In extending the remaining service lives of existing bridges, develop

criteria for evaluating and retrofitting bridges with joints to integral or semi-integral structures.

Establish an annual workshop between joint specialists of various State to

exchange information in the areas of design, construction and maintenance of joints and jointless bridges since there is continuing innovation and changing technology. This will help leverage the expertise of limited manpower in all the States and allow more effective communication of What works and what does not.

The decision to install an approach slab should be made by the Bridges

and Structures Office, with consultation from the Geotechnical group.

9

The decision should be based upon long-term performance and life cycle costs, rather than just first costs to the project.

Standardize practice of using sleeper slabs at the end of all approach slabs.

An irregular crack and pavement settlement typically develops at the interface of the approach slab and the approach pavement. Develop a method to control and seal this cracking, and if not already provided, develop a method to channel the water coming through this crack away from the pavement without allowing material to be washed away.

RECOMMENDED DESIGN DETAILS FOR INTEGRAL ABUTMENTS Use embankment and stub-type abutments. Use single row of flexible piles and orient piles for weak axis bending. Use steel piles for maximum ductility and durability. Embed piles at least two pile sizes into the pile cap to achieve pile fixedly

to abutment. Provide abutment stem wide enough to allow for some misalignment of

piles. Provide an earth bench near superstructure to minimize abutment depth

and wingwall lengths. Provide minimum penetration of abutment into embankment. Make wingwalls as small as practicable to minimize the amount of

structure and earth that have to move with the abutment during thermal expansion of the deck.

For shallow superstructures, use cantilevered turn-back wingwalls (parallel to center line of roadway) instead of transverse wingwalls.

Provide loose backfill beneath cantilevered wingwalls. Provide well-drained granular backfill to accommodate the imposed

expansion and contraction. Provide under-drains under and around abutment and around wingwalls. Encase stringers completely by end-diaphragm concrete. Paint ends of girders. Caulk interface between beam and backwall. Provide holes in steel beam-ends to thread through longitudinal abutment

reinforcement. Provide temporary support bolts anchored into the pile cap to support

beams in lieu of cast bridge seats. Tie approach slabs to abutments with hinge type reinforcing. Use generous shrinkage reinforcement in the deck slab above the

abutment. Pile length should not be less than 10 ft. to provide sufficient flexibility. Provide prebored holes to a depth of 10 feet for piles if necessary for

dense and/or cohesive soils to allow for flexing as the superstructure translates.

10

Provide pavement joints to allow bridge cyclic movements and pavement growth.

Focus on entire bridge and not just its abutments. Provide symmetry on integral bridges to minimize potential longitudinal

forces on piers and to equalize longitudinal pressure on abutments. Provide two layers of polyethylene sheets or a fabric under the approach

slab to minimize friction against horizontal movement. Limit use of integral abutment to bridges with skew less than 30 degree to

minimize the magnitude and lateral eccentricity of potential longitudinal forces.

SUMMARY

There are many advantages to jointless bridges as many are performing well in service. There are long-term benefits to adopting integral bridge design concepts and therefore there should be greater use of integral bridge construction. Due to limited funding sources for bridge maintenance, it is desirable to establish strategies for eliminating joints as much as possible and converting/retrofitting bridges with troublesome joints to jointless design.

The National Bridge Inventory database notes that eighty percent of the bridges in the United States have a total length of 180-ft. or less. These bridges are well within the limit of total length for integral abutment and jointless bridges. Where jointless bridges are not feasible, installation of bridge deck joints should be done with greater care and closer tolerances than normal bridge construction to achieve good performance. Since 1987, numerous States have adopted integral abutment bridges as structures of choice when conditions allow. At least 40 States are now building integral and/or semi-integral abutment type of bridges. Preference range from Washington State and Nebraska, where 80-90 percent of structures are semi-integral; to California and Ohio, which prefer integral, but use mix, depending upon the application; to Tennessee, which builds a mix of both integral and semi-integral, but builds integral wherever possible.

While superstructures with deck-end joints still predominate, the trend appears

to be moving toward integral. Although no general agreement regarding a maximum safe-length for integral abutment and jointless bridges exists among the state DOTs, the study has shown that design practices followed by the most DOTs are conservative and longer jointless bridges could be constructed.

There are several activities underway that will affect the way States are

designing jointless bridges in the future. These include a joint AASHTO/NCHRP task force responsible for initiating and drafting AASHTO design guide specifications and synthesis report on current practices for integral and semi-integral abutment bridges, FHWA-sponsored research study on Jointless

11

Bridges, update of LRFD specs to address jointless bridge design issues, and future workshops. An excellent reference document on current issues regarding jointless bridges is the FHWA Region 3 Workshop manual on Integral Abutment Bridges, November 1996.

Continuity and elimination of joints, besides providing a more maintenance

free durable structure, can lead the way to more innovative and aesthetically pleasing solutions to bridge design. As bridge designers we should never take the easy way out, but consider the needs of our customer, the motoring public first. Providing a joint free and maintenance free bridge should be our ultimate goal. The best joint is no joint. REFERENCES 1. Wasserman, Edward P. And Walker, John H., Integral Abutments for Steel Bridges, October

1996. 2. Burk, Martin P., Jr., An Introduction to the Design and Construction of Integral Bridges,

FHWA, West Virginia DOT and West Virginia University, Workshop on Integral Bridges, November 13-15, 1996.

12

Integral Abutments and Jointless Bridges (IAJB) 2004 Survey Summary

Rodolfo F. Maruri, P.E.1 and Samer H. Petro, P.E.2

ABSTRACT

Integral Abutment and Jointless Bridges (IAJB) have been used for decades and the criteria for using them and detailing has varied from state to state. The main advantage of IAJB is the elimination of joints, which after they start leaking, account for 70% of the deterioration that occurs at the end of girders, piers and abutment seats. FHWA promotes the usage of Integral Abutment and Jointless bridges, where appropriate, as one method of building bridges that will last 75-100 years with minimal maintenance.

In 1995 and 1996, Federal Highway Administration (FHWA) in conjunction

with the Constructed Facilities Center (CFC) at West Virginia University (WVU) conducted a survey and workshop about Integral Abutment Bridges [1]. In 2004, another survey [2] was developed by FHWA and the CFC at WVU, using similar questions as the 1995 survey and incorporating additional questions, to obtain a status of usage and design for Integral Abutments and Jointless bridges. The survey was distributed by AASHTO subcommittee on Bridge and Structures to all 50 states Department of Transportation (DOT), District of Columbia DOT, Puerto Rico Highway and Transportation Authority and Federal Lands Highways Division (referred to as states in the paper).

This paper summarizes the responses received to date from the states. The

survey was divided into different topic areas which included General Issues, Design and Details, Foundation, Abutment/Backfill, Approach Slabs, Retrofit (Jointed to Jointless), and Other Issues. Integral Abutments, as defined in the survey and in this paper, refers to the monolithic construction of the abutment with the deck in order to eliminate the joints at the end of the bridge. This includes the use of Full, Semi Integral Abutments and Deck Extensions. Jointless bridges refers to the elimination of joints at the piers through the usage of integral pier caps, continuous spans and continuous for live load construction.

The purpose of the survey was to obtain a snapshot about the usage of integral

abutments and jointless bridges from the states, their policy, their design criteria and other issues. The results of the survey are presented in this paper and will be used to disseminate information between states and help FHWA encourage the usage of IAJB.

1 Rodolfo F. Maruri, P.E., Federal Highway Administration, Richmond, Virginia. 2 Samer H. Petro, P.E., Gannett Fleming, Inc., Morgantown, West Virginia.

13

INTRODUCTION

Integral abutments and jointless bridges (IAJB), when properly designed and constructed, perform better than bridges with expansion joints because they minimize maintenance, extend service life of bridge components including bearings, abutment and pier seats, paint system and superstructure. Although integral abutments have been designed and constructed successfully for decades, the design and analysis of these structures have relied primarily on a fragmented body of technical references, and design and construction details have varied from state to state.

The Federal Highway Administration (FHWA) in conjunction with the

Constructed Facilities Center (CFC) of West Virginia University (WVU) conducted a survey of integral abutment design and construction as part of a three-day workshop on integral abutment and jointless bridges scheduled for March 16-18, 2005 in Baltimore, Maryland. The IAJB 2004 survey was sent to all 50 States Department of Transportation (DOT), DC DOT, Puerto Rico Highway and Transportation Authority and the Federal Lands Highway Division (referred to as states in the paper). The survey was conducted with the intention that bridge designers and owners will use the information to promote usage and design practices for integral abutment and jointless bridges.

The IAJB 2004 survey include questions about the number of integral

abutments designed, built and in service, the criteria used for design and construction, including maximum span lengths, total length, skews and curvature and problems experienced with integral abutment bridges. In addition, the survey questioned the states about their design considerations such as thermal movement, passive earth pressure, approach slabs, foundation and pile design and retrofitting of non-integral abutments to integral abutments.

For consistency in the terminology used, the survey defined and provided a

sketch of full integral abutments, semi integral abutments, and deck extensions. Full Integral Abutment was described as a capped pile stub type abutment with or without a hinge between superstructure and foundation cap, semi-integral abutment was described as a rigid, non-integral foundation with movement system primarily composed of internal end diaphragms and movable bearings in a horizontal joint at the superstructure-abutment interface and a deck extensions was described as extension the deck over the top of the backwall and place joint behind the abutment backwall to prevent deterioration of the end of superstructure beams [2].

Of the fifty-three (53) states surveyed, thirty-nine (39) states (74%) responded (Figure 1). In addition, other states indicated that they will submit the responses to the survey in the future.

14

Figure 1: States that Responded to the IAJB 2004 Survey According to the responses, there are approximately 13000 integral abutment

(IA) bridges, of which approximately 9000 are full integral abutment bridges, approximately 4000 are semi-integral abutment bridges and approximately 3900 deck extension bridges in-service (Table 1). The increase in the number of integral abutments from the numbers reported in the 1995 survey [1] can be attributed to the acceptability of the benefits of integral abutments, familiarity with design and construction issues and a larger sample of responding states (39 respondents in 2004 versus 18 in 1995). The numbers reported are approximate since the National Bridge Inventory (NBI) data, which is kept by all the states with information about their bridges, does not differentiate between the different types of abutments and most states do not have other methods for maintaining an inventory bridges and/or integral abutments.

The following sections in the paper, discuss the survey answers, analysis and

compilation of answers and conclusions based on responses provided.

GENERAL ISSUES This section of the survey questioned the states about their use of integral

abutments since the last workshop [1], the number of integral abutment bridges in service, the states policy for design of jointless bridge construction and the criteria used for integral abutments and jointless bridges. A breakdown of integral bridges designed and built since 1995 and the total number of in-service integral abutment bridges is shown in Table 1.

15

Table 1: Number of IAJB Designed and Built Since 1995 and In-Service

DESIGNED since 1995 BUILT

since 1995 IN SERVICE

(TOTAL) Integral Abutment ~ 7000 ~ 8900 ~ 13000

Full Integral ~ 5700 ~ 6400 ~ 9000 Semi Integral ~ 1600 ~ 1600 ~ 4000

Deck Extension ~ 1100 ~ 1100 ~ 3900

The survey responses indicate an increase in the number of integral abutments

of over 200% in the last 10 years. As in 1995, Tennessee continues to have over 2000 integral abutments bridges, but Missouri reports having 4000 integral abutment bridges, which represent the largest amount of integral bridges. An increase in the number of integral abutments, since 1995, is most evident in the northern states where Illinois, Iowa, Kansas and Washington all reported having over 1000 in-service. In addition, Michigan, Minnesota, New Hampshire, North Dakota, South Dakota, Oregon, Wyoming and Wisconsin, reported having between 100-500 integral abutment bridges in-service. Unlike the northern states, the southern states like Florida, Alabama and Texas do not use integral abutment and reported having one or less bridges with integral abutments.

As illustrated in Figure 2, fifty-one percent (51%) of the responding states

indicated that they designed and built over 50 integral abutment (IA) bridges since 1995, of which 21% built 101-500 integral abutment bridges, 5% built 501-1000 integral abutment bridges and 5% built over 1000 integral abutment bridges.

3%

8%

10%

23%

18%

5% 5%

3%

18%

10%

8%

21%

5% 5%

15%

21%

0%

10%

20%

None 1 - 10 11 - 20 21 - 50 51 - 100 101 - 500 501 - 1000 Over 1000

NUMBER OF BRIDGES DESIGNED USING INTEGRAL ABUTMENTS (RANGE)

PER

CEN

T O

F ST

ATE

S

Integral Abutment (Designed)

Integral Abutment (Built)

Figure 2: Percent of States that reported designing and building Integral Abutments within

the range specified (since 1995).

16

Analysis of responses shows a similar distribution with the in-service integral abutments bridges as in Figure 2. For in-service integral abutment bridges, fifty-nine percent (59%) of the responding states indicated having over 50 integral abutment bridges in-service, of which 31% have 101-500 integral abutment bridges, 3% have 501-1000 integral abutment bridges, and 15% have over 1000 integral abutment bridges. The 2004 survey revealed that approximately 85% of the states (about 33 states) constructed integral abutment bridges equally with either simple spans or multiple spans.

The responses to General Question 4, regarding the states criteria for using

integral abutments, show that a majority of the states do not limit the maximum span within the bridge, but do limit the total length of the bridge and the skew of the bridge. Table 2 summarizes the criteria range provided by the states for prestressed concrete girder and steel bridges for maximum span, total length of bridge, maximum skew of bridge and maximum curvature.

Table 2: Range of Design Criteria Used For Selection of Integral Abutments.

PRESTRESSED CONCRETE GIRDERS

RANGE

STEEL GIRDERS RANGE

MAXIMUM SPAN MAXIMUM SPAN Full Integral 60 200 Full Integral 65 - 300 Semi Integral 90 200 Semi Integral 65 - 200 Deck extensions 90 200 Deck extensions 80 - 200 Integral Piers 120 200 Integral Piers 100 - 300 TOTAL LENGTH TOTAL LENGTH Full Integral 150 1175 Full Integral 150 - 650 Semi Integral 90 3280 Semi Integral 90 - 500 Deck extensions 200 750 Deck extensions 200 - 450 Integral Piers 300 400 Integral Piers 150 - 1000 MAXIMUM SKEW MAXIMUM SKEW Full Integral 15 70 Full Integral 15 - 70 Semi Integral 20 45 Semi Integral 30 - 40 Deck extensions 20 45 Deck extensions 20 - 45 Integral Piers 15 80 Integral Piers 15-No Limit MAXIMUM CURVATURE

MAXIMUM CURVATURE

Full Integral 0 10 Full Integral 0 - 10 Semi Integral 0 10 Semi Integral 0 - 10 Deck extensions 0 10 Deck extensions 0 - 10 Integral Piers 3 - No Limit Integral Piers 0 - No Limit

17

Based on the experience of many states, their comments and their established design criteria, it can be concluded that a majority of new bridges could be built using integral abutments. Colorado, Iowa and Tennessee indicated that they built the majority of their new bridges using integral abutments.

The utilization of integral abutments with curved bridges is not widely

accepted based on survey responses. Four states reported that they allow the use of curved girder bridges with integral abutments and three (3) more allow the construction of curved bridges with straight girders and integral abutments. An alternative mentioned to account the forces in curved bridges and/or long bridges is the use of integral abutments with an expansion joint elsewhere on the bridge.

The IAJB 2004 survey shows that although progress has been made in the construction of integral abutments since 1995, there is still a lot of variability in the usage and criteria used for selection of integral abutments. The non-uniformity of selection criteria for integral abutments indicates that this is an area where standardization is warranted.

DESIGN AND DETAILS This section of the IAJB 2004 survey questioned the states regarding their changes to the design procedures or details, future plans for jointless bridges, policy regarding the use of integral abutments, forces and loads used to design integral abutments and other design issues. The following presents the questions asked in the survey and their respective responses. Question 1 in this section asked whether the design procedures or details changed since August of 1995 with regard to loads, substructure design, backfill, approach slabs and jointless retrofit of bridges. The survey results indicated that less than 25% of the states that responded changed their design procedures regarding primary and secondary loads, substructure design, abutment/backfill and approach slab since 1995. However, this percentage increased with regard to changing the details associated with the same issues of integral abutments. The largest change occurred in the details for foundation/substructures and approach slabs (38% and 36% respectively). Figure 3 illustrates the percentage of states that responded in each of the issues noted. Some of the changes reported in the survey include Iowa specifying a 10-foot prebored hole filled with bentonite for each pile (8-foot prebored hole used prior to 2002), Virginias accountability of lateral forces on skewed integral abutment bridges, and Connecticuts incorporation of approach slabs in all bridges to minimize bump/settlement problems at the bridge/approach fill interface. In addition to these detailing changes, several states noted that they have changed their design to incorporate the Load Factor Design (LFD) specifications and/or the Load and Resistance Factor Design (LRFD) specifications.

18

24% 23%

21%

15%

18%

8%

10%

13%

36%

28%

36%

13%

0%

10%

20%

30%

40%

Primary LoadConsiderations

Secondary LoadConsiderations

Substructure/foundation Abutment/Backfill Approach Slab Jointless retrofit

PER

CEN

T O

F ST

ATE

S

DESIGN PROCEDURES

DETAILS

Figure 3: Percent of States that Reported Changing their Design Procedures and Details since 1995.

Design and Details Question 2 and 3 asked about the states future plans for jointless bridge construction, including the future use of integral abutments, continuous spans, retrofit of existing bridges, and policy about elimination of joints. The survey revealed that over ninety percent (90%) of the states have a policy to eliminate as many joints as possible and construct jointless simple and continuous span bridges whenever possible. However, only 77% indicated that they will design integral (fully and semi) abutments whenever possible and 79% noted that they will design bridges as jointless whenever they meet the design criteria for jointless bridges (Figure 4). The difference in the percentages between eliminating as many joints as possible (92%) and using integral abutments whenever possible (77%) can be attributed to states that do not extensively use chemicals for deicing of bridges in the winter and therefore do not have a policy of incorporating integral abutments in their bridge design (Figure 5). Noteworthy comments included Oregons comment about problems with multiple-span jointless bridges; Arizonas comment about having problems with integral abutment approach slabs which is the reason Arizona does not use integral abutments anymore; Vermont noting that they do not use integral abutment extensively because of scourability issues; and Washington State noting that they preferred using semi-integral type abutments because they are more economical since it avoids the transfer of seismic forces into the substructure

19

79%

92%

54%

90%

77%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Design integral (fully andsemi) abutments whenever

possible.

Design bridges as jointlesswhenever they meet the

design criteria for jointlessbridges.

Eliminate as many deck jointsas possible.

Retrofit existing bridges andeliminate deck joints where

possible

Design (jointless) simple andcontinuous span bridges

PER

CEN

T O

F ST

ATE

S

Figure 4: Percent of States that Answered with Regard to Their FUTURE Plans for

Jointless Bridge Construction.

Figure 5: Future Plans for IAJB Design and Construction

20

Design and Details Question 4 dealt with the forces, including passive and active earth pressure, temperature, creep, shrinkage, settlement, additional loads due to skew layout, additional forces due to curvature and other forces that states account for in the design of integral abutments. The survey revealed that 72% of the states account for temperature related forces (Figure 6). In addition, states also noted that they account for temperature (temperature gradient, thermal expansion and contraction in longitudinal and transverse direction) in their design (Design and Details Question 5), but the procedure for accounting for the thermal expansion and contraction varied widely.

The survey results also indicate that 59% of the states surveyed accounted for passive earth pressures, but only 21% of the states allow for curved bridges with integral abutments and account for the additional forces due to the curvature of the bridge (Figure 6).

Noteworthy comments about design of integral abutments include Illinois practice to designed only for vertical loads, North Dakotas practice to use 1000 lb/ft2 to account for various loads (passive pressure, thermal, creep and shrinkage loads) and Iowas use of the a simple, fixed-head pile model which does not consider passive or active pressure and is based on research conducted by Greimann and Abendroth at Iowa State University during the 1980s.

59%

33%

44%41%

21%

15%

72%

28%

0%

10%

20%

30%

40%

50%

60%

70%

Passive EarthPressure

Temperature. Creep Shrinkage Settlement Additional forcesdue to skew

Additional forcesdue to curvature

Other. Describebelow in

commentssection.

PER

CEN

T O

F ST

ATE

S

Figure 6: Percent of States That Account for the Forces Listed in the Design of Integral

Abutments. Thirty-three percent (33%) of the responding states account for creep effects when designing integral abutment bridges (Design and Details Question 4 and 6), while Georgia, Illinois, Iowa and other states indicated that they do not account for creep movements.

21

As expected, the majority of the states responded that they use computer software to design integral abutment bridges (78%). However, the program and/or method used varied widely. Several states, including California, Illinois and North Dakota indicated that they use hand calculations and charts, while other states noted that they have developed their own in-house spreadsheet, using Excel and MathCAD, to design integral abutment bridges. Structural programs and finite element software like STAAD, STRUDL, and RISA are used by Pennsylvania, Rhode Island and North Carolina to design integral abutments while Tennessee, New Hampshire, Virginia and New Jersey use COM624P and/or L-pile for pile design. FOUNDATIONS The monolithic construction of the deck with integral abutment (backwall) requires special design for the backwall and supporting piles of integral abutments and jointless bridges. The design of the foundation for integral abutments needs to account for the expansion and contraction of the bridge due to thermal movement. The resulting soil pressures due to thermal expansion and restraining effects due to jointless construction of the bridge have been recognized as the controlling load for design of integral abutments and piles. Designing and detailing of integral abutments to handle these forces is critical for the proper performance of integral abutments. The 2004 IAJB survey questions where chosen to obtain an understanding about how states are designing foundations for integral abutments, including criteria used to select foundation type, type of pile, orientation of pile, pile design considerations, pressure used in the design of integral abutments and special details utilized to reduce the pressures at the integral abutment. The survey responses (Foundation, Questions 1 and 2) indicate that full-integral abutment with steel bearing piles is the most commonly type of integral abutments (~ 70%). However, several states noted that they are currently designing and/or creating standards for semi-integral abutments. The comments provided indicated that semi-integral abutments are commonly used with the uncharacteristic designs that incorporate larger skews, higher abutment walls and unique soil conditions. Washington State noted they preferred using semi-integral abutments because they are more economical since they avoid transferring seismic forces into the substructure. New Hampshire indicated that they use deck extensions extensively since the foundation design is not an issue. The use of deck extensions is predominant in the northeast region (New Hampshire, New York, Connecticut and Maine) as is evident in the large number of in-service deck extensions in this region.

22

Nevada and Hawaii indicated that in addition to steel bearing piles (H piles and pipe piles), friction piles and spread footings, they are using drilled shafts for foundations of integral abutments. Noteworthy, even though steel bearing piles were the most common type of pile used for integral abutments, there was no consensus on the typical orientation of the pile (Foundation, Question 4). Thirty three percent (33%) of the responding states orient the piles with the strong axis parallel to the centerline of bearing, 46% orient the piles with the weak axis parallel to the centerline of bearing, 8% (3 states) leave it to the discretion of the Engineer and the remaining 13% did not provide a comment or noted that the question was not applicable because of their use of symmetric piles (Figure 7). The non-uniformity of pile orientation seems to indicate that this is an area where further standardization is warranted.

33%

46%

8%

13%

0%

10%

20%

30%

40%

50%

Strong Axis Parallel to CL ofBearings

Weak Axis Parallel to CL ofBearings

Designer's Option Not Applicable (Symmetric Pile)or No Answer Provided

ORIENTATION OF PILES

PER

CEN

T O

F ST

ATE

S

Figure 7: Typical Pile Orientation Use for FULL Integral Abutments Responses to Foundation Question 3 indicate that a number of states have developed office practices that allow designers to detail integral abutments without doing complicated analysis. These states use the office practices in conjunction with geotechnical recommendations based on soil parameters to decide the type of foundation used. Based on the comments provided, there is no evidence of problems relating to the type of foundation used for integral abutments. The use of MSE wall has increased dramatically over the years and as such the use of integral abutments where the MSE walls serves as a component of the integral abutments has increased correspondingly. Foundations-Question 5, questioned about the states policy regarding the combination of integral

23

abutments and MSE walls. Based on the survey responses, the preferred detail is to offset the MSE wall from the integral abutment and footing between two (2) feet to five (5) feet. According to comments, the offset provides space for construction of MSE wall and offsetting of MSE straps around abutment piles. In addition to offsetting of the integral abutment behind the MSE wall, several states noted that they have special requirements for the placement of piles in the MSE backfill including the use of sleeves filled with sand. The detailing of MSE abutments with integral abutments is inconsistent based on the responses received and is another area where guidelines based on all available research would be beneficial to states that are currently using this type of construction and/or plan to use it. The soil pressure used for the design of integral abutments and its piles has been the subject of controversy and much research. The survey, Foundation-Question 6, shows that there is still no consistent design method used with regard to soil pressures. The majority of the respondents indicated that they use passive pressure (33%) and/or a combination of passive and active pressures (18%). Active pressures, however, is used by a minority of respondents (8%) and other combination of pressure and/or methods was used by 26% of the states responding (Figure 8). The survey was not specific enough to make any conclusions about the variability of pressures used in the design of integral abutments.

18%

8%

26%

33%

0%

10%

20%

30%

Combination Active Pressure Passive Pressure Other PRESSURE USED FOR DESIGN

PER

CEN

T O

F ST

ATE

S

Figure 8: Typical Soil Pressures Used for Design of Substructure

24

The limits or capacities used for piles provide another opportunity for standardization. Based on the comments provided (Foundation-Questions 7, 8 and 9), states use AASHTO in combination with statewide practices that limit lateral deflection of pile, computer programs and other methods to determine the capacity of piles. The pile capacity is based on the axial capacity of the pile (using 0.25*fy as stipulated in AASHTO Standard Specification, section 4.5.7.3 or other) [41% of states], or a combination of axial/bending capacity based on beam-column analysis and frame analysis [51% of states]. In addition to accounting for bending due to expansion/contraction of superstructure, 26% of the states also account for the bending due to superstructure rotation in the horizontal plane (skew bridges) (Figure 9).

41%

51%

26%

0%

10%

20%

30%

40%

50%

Axial Forces (No Bending) Bending Forces Due toExpansion/Contraction of Superstructure

Bending Forces Due to SuperstructureRotation in the Horizontal Plane (skew

bridges)

TYPE OF FORCE USED TO DETERMINE CAPACITY OF PILE

PER

CEN

T O

F ST

ATE

S

Figures 9: Forces Used to Determine Capacity of Piles for Integral Abutments.

ABUTMENT/BACKFILL

The handling of the backfill behind the integral abutments can have a significant effect on the performance of integral abutments and as a result has been discussed and researched over the past decades. A review of the answers and comments provided in the IAJB 2004 survey (Abutment/Backfill Question 1 and 2), show that most states require the fill behind the integral abutment to be compacted (69%) as compared to 15% for uncompacted fills. Interestingly, in addition to using compacted fills there are a number of states that require the use of expanded polystyrene (EPS), other compressible materials behind abutment, lightweight fills and additional inspection during construction in order to reduce

25

and/or control the earth pressures exerted on integral abutments during expansion cycles (Figure 10). The other survey questions in this section inquire about whether the states specify the minimum length of approach fill required behind the integral abutment (Abutment/Backfill - Question 3) and whether the states limit the maximum height of integral abutments (Abutment/Backfill Question 4). In both questions, the analysis of the responses indicated that the states specified the length of approach fills and/or limited the height of the integral abutments 31% of the time, but the majority of the states did not limit or specified these parameters. The comments for Abutment/Backfill Question 4 inferred that the limit for the height applied only to the full integral abutment. Washington State indicated that they have used a 30-foot high semi-integral abutment.

13%10% 10%8%8%

13%

69%

0%

10%

20%

30%

40%

50%

60%

70%

Require compactedbackfill

Requireuncompacted backfill

Use ExpandedPolystyrene (EPS)

Use othercompressible

material behindabutment.

Use Lightweight fill Require additionalinspection during

construction

Other. Describe incomments section

below.

PER

CEN

T O

F ST

ATE

S

Figure 10: Percent that Responded to Listed Requirement for Approach Backfill.

APPROACH SLABS

Some of the most common problems associated with integral abutments are the settlement and the cracking of approach slabs. Fortunately, these problems do not cause a significant disruption of traffic or a decrease of the service life of the bridge. The questions in this section were designed to find out the design procedure and details used for approach slabs (Approach Slabs Questions 1 and 2), the problems experienced with approach slabs (Approach Slabs Question 3) and the criteria about the usage of approach slabs behind integral abutments (Approach Slabs Question 4).

26

The comments provided in Approach Slabs - Questions 1 and 2, indicate that there is no consistency in the detailing of approach slabs. Thirty-one percent (31%) of respondents indicated that they use a sleeper pad at the end of approach slab, 26% indicated that they float the slab on the approach fills and 30% indicated that they do both. Many states indicated that they have or are using corbels on the abutment backwall for the support of the approach slab, while other states indicated that they use reinforcing projecting from the abutment backwall to tie the approach slab to the abutment backwall, and other states are using a combination. Based on the responses received, it is evident that the detailing of approach slabs, including the connection to the abutment backwall and the interface between the approach slab and approach fills is an area where standardization and guidelines would be beneficial. Review of comments provided for Approach Slabs Question 3, indicates that approach slab settlement, cracking, and bump at the interface between the approach slab and approach fill are the major problems with approach slabs. The answers and comments provided with this question are consistent with the answers provided in Other Issues Question 1 (Figure 11). In order to mitigate some of the problems with approach slabs, several states are using buried approach slabs and/or select fills under the approach slab while other states have filled voids under approach slab with grout, resurfaced approach slabs with asphalt, and/or used an overlay. Surprisingly, a state noted that the reasons they do not use integral abutment bridges anymore are because of the bump formed at the end of the approach slab and settlement problems under approach slabs due to poor drainage.

RETROFIT Retrofit of jointed decks to jointless decks has been increasing as the condition of the decks deteriorates and a complete deck replacement is needed. Forty-nine percent (49%) of the respondents indicated that they have a policy to retrofit existing bridges whenever possible. Virginia has used a poor-man continuity detail, obtained from Utah DOT, for complete and partial re-decking projects with great success. For re-decking projects, Virginia has found that the main cost of using a continuous deck with simple spans is the need to retrofit the existing bearings. As the age of the bridges in the United States continues to increase and concrete decks need to be replaced to improve rideability and the condition rating of the decks, consideration should be given to retrofitting existing simple span to continuous (jointless) deck. Based on the responses in the survey, this is an area where further guidelines would be beneficial to those states that have not yet established a retrofit policy.

27

Conversion of non-integral abutment bridges to integral or semi-integral abutment bridges is not widely used and only 39% of the responding states indicated that they have a policy to investigate its feasibility. New Mexico notes that this is very common retrofit and that the selection of bearings is critical; Missouri notes that they consider retrofitting only on short spans and small skew; Virginia notes that they try to incorporate this retrofit with major superstructure replacement projects; and South Dakota notes that they incorporate this type of retrofit with a major renovation such as a deck replacement or if there are severe problems at the abutment.

OTHER ISSUES The last section of the survey inquired about the problems states are having with integral abutments (Other Issues Question 1), special details/design procedure for bearing in integral abutments (Other Issues Question 2) and list of recent research in the area of integral/jointless bridges (Other Issues Question 3). Details and design procedure for bearings and recent research in the area of integral/jointless bridges will be provided during the IAJB 2005 Conference. A summary of the problems reported with Other Issues Question 1 is shown in Figure 11. The compilation of Other Issues Question 2 revealed that 36% of the states use the same procedure for design of integral abutment bearings as for regular abutment bearings, 13% design the bearings based on thermal movements and 38% use other methods to design and detail the bearings of integral abutments.

15%

26%

10%

8%

3%

46%

28%

0%

10%

20%

30%

40%

50%

Cracking of integralabut. backwall

Cracking of deck atintegral abutment

Cracking of approachslabs

Cracking of wingwall Detailing Detrimental rotationof integral abutment

backwall

Setlement ofapproach slabs

PER

CEN

T O

F ST

ATE

S

Figure 11: Problems Experienced with Integral Abutments.

28

CONCLUSIONS / RECOMMENDATIONS The IAJB 2004 survey revealed that design practices and details vary greatly from state to state. For example, maximum span limits, curvature and skew effects, thermal movement limits and creep effects are just a few criteria that differ considerably between states according to the surveys received. Therefore, the papers authors believe that standardization and/or guidelines are warranted. The cost associated with proper maintenance of joints, subsequent deterioration of bridge components when joints do not perform satisfactorily and FHWAs goal to built bridges with 75-100 years service life with minimal maintenance, are some of the reasons that the authors consider that integral abutment and jointless bridges should be the construction of choice, whenever feasible. Based on the compilation of the surveys, the following conclusions and recommendations are made:

1. Develop guidelines that incorporate the research done in the area of integral abutments. These guidelines should include problems experienced by other states, as well as design guidelines and examples. Examples of areas identified in the 2004 IAJB survey where non-uniformity in design and detailing are apparent include,

a. Criteria used for selection of integral abutments. b. Forces and pressures used to design integral abutment and integral

abutment piles. c. Orientation of integral abutment piles. d. Design of integral abutments with curved bridges. e. Detailing of approach slab at bridge interface and approach fill

interface.

2. Promote the issuance of a national policy for the use of integral abutments, especially in states that indicated that they do not have a policy regarding the future use of integral abutments.

3. Develop guidelines and/or additional information to increase the use of continuous jointless and/or continuous decks with simple span superstructures, whenever appropriate.

4. Develop guidance and/or additional information on the use of deck extensions to eliminate joints at abutments. These guidelines should incorporate criteria on when to use them, problems experienced by other states, and design guidelines.

5. Develop guidelines and/or additional information for detailing of integral abutments around MSE walls.

29

6. Develop guidelines and/or additional information for handling of backfill behind integral abutments.

7. Develop guidelines and/or additional information for detailing of approach slabs to minimize cracking and mitigate problems at the approach slab and roadway fill interface.

8. Develop guidelines and/or additional information for the retrofitting of existing bridges to eliminate joints at piers and abutments.

9. Create a new survey incorporating comments given in the IAJB 2004 Survey and clarifying areas which were unclear according to submitters.

10. Re-issuance as a Technical Advisory by Federal Highway Administration and/or policy guidelines for Integral Abutments.

These recommendations and conclusions presented within this paper are the opinions of the authors, based on the IAJB 2004 survey, and are not necessarily endorsed by Federal Highway Administration. However, a goal of the conference is for the CFC at WVU to provide recommendations to Federal Highway Administration for its consideration and issuance as a Technical Advisory and/or policy guideline. ACKNOWLEDGEMENT

The authors would like to thank all the individuals who completed the survey and returned it to the CFC at WVU. Their work is the basis for the survey summary and this paper. In addition, we would like to acknowledge the support of Malcolm T. Kerley, chairman of AASHTO sub-committee on Structures and Bridges, for sending the IAJB 2004 survey to all the states Bridge Engineers. REFERENCES 1. GangaRao, H., Thippeswamy, H., Dickson, B. Franco, J., 1996. Survey and Design of

Integral Abutment Bridges, Constructed Facilities Center at West Virginia University, Morgantown, West Virginia.

2. Maruri, R., Petro, S., GangaRao, H., 2004. IAJB 2004 Survey, Federal Highway

Administration and Constructed Facilities Center at West Virginia University, Morgantown, West Virginia.

30

The In-Service Behavior of Integral Abutment Bridges: Abutment-Pile Response

Robert J. Frosch, Purdue University

Michael Wenning, American Consulting, Inc. Voraniti Chovichien, Thai Engineering Consultants Co, Inc.

ABSTRACT

Integral bridges have been used for many years across many regions of the country. However, empirical guidelines have often limited their use. While removal of limits imposed by these guidelines may be warranted, there are many questions regarding the behavior of these structures that remain unanswered. In particular, the interaction of the abutment, pile and soil remains uncertain. In Indiana, the decision to explore extension of the limits has resulted in a study to ascertain the in-service behavior of integral abutment bridges. Through several field instrumentations, new light is being shed on the behavior and performance of these bridges. The behavior of integral abutment bridges is concentrated in the response of the abutment-pile-soil system. Therefore, this response is the focus of this paper. INTRODUCTION

The Indiana Department of Transportation (INDOT) has used empirical limits for integral bridge construction. These limits are similar to those used by many departments; namely bridges less than 300 ft in total length with skews no greater than 30 degrees [1]. Like most departments, INDOT has observed many advantages to this type of construction and wishes to increase these limits. Among the leading advantages are simpler construction and reduced maintenance.

Many unanswered questions, however, have kept INDOT from opening the

limits on these designs. Among them are: How far can empirical details be stretched without additional analysis? Should H-piles be oriented in their strong or weak axis for best results? What methods of design should be proscribed when bridges fall outside

the empirical limits? What components of the bridge require additional design? How much does the mass of the bridge buffer the daily temperature

changes? Does the bridge skew enter into the design and if so how?

31

In May 1998, the Indiana Department of Transportation made a decision that is likely to change the course of bridge design in the state. They authorized the design of a jointless bridge 302 m (990 ft) long, over 3 times longer than any previously built in Indiana. This decision began a series of events that resulted in a research study at Purdue University [2,3]. While the overall goal of the Purdue research study was to provide answers to the various questions, the first phase of research was directed towards providing an understanding of the in-service behavior of these bridges.

SR 249 OVER US 12 THE FIRST BRIDGE The SR 249 Bridge is designed to carry traffic over US 12 and nine railroad

tracks into the Port of Indiana at the northernmost part of the state. The ten-span bridge has a 13-degree skew and a total length of 990 ft (302 m). Individual spans ranged from 87 to 115 ft (26.4 to 35.0 m). The intent of this project was to build a bridge using standard construction materials and details and monitor it to evaluate the assumptions that were made in design.

Due to the size of the bridge, alternate plans were produced for both a steel

girder and prestressed bulb-tee option. The continuous, composite prestressed bulb-tee option received the low bid and was constructed. The four girders were 5 ft deep, made of semi-lightweight concrete (130 pcf) and spaced at 10 ft-4 in. (3.15 m) on the 38ft-4 in. (11.7 m) wide bridge.

The girders sat on elastomeric bearing pads that were designed to deflect with

the anticipated temperature change. Per INDOT detailing standards, the diaphragm encapsulating the ends of the beams over the piers were cast the full width of the bridge seat, or 50 inches in this case. The bottom of the diaphragm rested on a layer of polystyrene that was intended to allow the superstructure to expand or contract without locking up at the piers. A keyway is provided to restrain the superstructure from excessive movement.

The design of the end bent presumed that the bridge length would expand and

contract with annual temperature variations. For this region, AASHTO temperature ranges of 45 and 60 degrees from construction temperatures were used for the concrete and steel options, respectively. This amounted to 1.6 in. and 2.3 in. of anticipated movement in each direction.

Once the movement was computed, a model was set up to calculate the

resisting force of the soil. Soil borings at the site revealed that the bridge would be founded on seams of peat and marl that extended about 45 ft below ground line. The geotechnical report advised against adding any weight to the existing spill slopes due to large anticipated settlements. As a result, the approaches for the bridge had to be constructed of expanded polystyrene fill. This created a situation that pulled the project out of its standard construction methods

32

Sub-Base

PrestressedBulb TeeGirder

HP14x89

Expanded Polystyrene

Fill

Approach Slab Deck

17

Retaining Wall

18Min. 18

Anchor Plateand Beam Seat

Sleeper Slab

Ground Level

Sub-Base

PrestressedBulb TeeGirder

HP14x89

Expanded Polystyrene

Fill

Approach Slab Deck

17

Retaining Wall

18Min. 18

Anchor Plateand Beam Seat

Sleeper Slab

Ground Level

category. The piles supporting the end bents would have to be designed as free standing for the first 19.68 ft. The computer program COM624P was used to model the spring coefficients for the various soil layers. The anticipated deflection at the pile head was inputted to obtain the maximum moment and point of fixity for the pile. The point of fixity was assumed to be the second location where the deflection diagram crossed the zero point. The portion of the pile extending into the cap was covered with polystyrene to obtain a pinned connection. Piles were then designed as columns with a height from the point of fixity to the bottom of the end bent.

Steel encased concrete (shell) piles, 14 in. diameter, would have been the first

choice of support for these soil conditions. However, when an analysis was performed, the thickness of the piles would have been excessive. Larger diameters were investigated but the same thickness was always required. Since the stiffness of the pile increases the force required to move it the predetermined distance, the moment increased linearly with the pile section modulus. It was determined that the shape of the pile would have to change to obtain a better ratio.

H piles were then investigated in both strong and weak axis orientation.

Ultimately a strong axis orientation was used to avoid the possibility of local flange buckling. A schematic of the end bent detail is provided in Figure 1.

Figure 1: SR 249 End Bent Detail

The bridge was instrumented with a combination of strain, tilt, crack and temperature meters. In addition to this bridge, INDOT has continued to build and instrument others to measure the response of other types of bridges, piles, skews and soil conditions.

FIELD STUDIES

Overall, four bridges in Indiana have been instrumented to observe the in-service behavior of integral abutment bridges as well as the behavior of the piles

33

supporting these structures. These bridges range in length from 150 to 990 ft providing a spectrum of behavioral data [2,3].

While the SR 249 Bridge provided excellent information regarding the

behavior of a relatively long integral structure, this structure was not typical in regards to the design of the end bent. Therefore, several other bridges including the SR18 over Mississinewa River Bridge (Figure 2) were also selected for instrumentation. There are several reasons that the SR18 Bridge in particular was selected.

1. The bridge was designed and constructed according to typical integral

abutment details. 2. The bridge exceeded the length limitation of INDOT and could provide

much needed data regarding bridge length. 3. The skew of the structure was small. Therefore the research could focus

on the effects of bridge length.

Figure 2: SR18 over Mississinewa River Bridge

To better understand the soil-pile-abutment-system, the five-span, continuous

prestressed, concrete bulb-tee integral bridge was instrumented. This bridge is located east of the city of Marion in Grant County, Indiana on the westbound lanes of State Road 18 crossing the Mississinewa River. The total bridge length is 367 ft with a skew angle of 8.

To evaluate the abutment movement, tiltmeters and convergence meters were

provided on the end bents (Bents 1 and 6). The convergence meter was oriented horizontally and operated perpendicular to the abutment. This meter measured the relative displacement between the end bent and the reference pile to determine the longitudinal abutment movement. The locations of the convergence meters, tiltmeters and pile strain gages are shown in Figure 3.

34

Tiltmeter

10

8 Slab

R/C Bridge Approach Slab

Prestressed Conc. Bulb-T Beam

End Bent Backfill

3-3

Strain Gages

1-3

14 CFT pile

Convergence Meter

Reference Pile

3

Drain

Ground Level

Strain GageTiltmeter

10

8 Slab

R/C Bridge Approach Slab

Prestressed Conc. Bulb-T Beam

End Bent Backfill

3-3

Strain Gages

1-3

14 CFT pile

Convergence Meter

Reference Pile

3

Drain

Ground Level

Strain Gage

Centerline ofAbutment

1-3

Parallel toRoadway

8

5 sp

aces

at 4

=20

Abutment

PilePlan

Elevation

N

90

GroundLevel

Strain GageCenterline ofAbutment

1-3

Parallel toRoadway

8

5 sp

aces

at 4

=20

Abutment

PilePlan

Elevation

N

90

GroundLevel

Centerline ofAbutment

1-3

Parallel toRoadway

8

5 sp

aces

at 4

=20

Abutment

PilePlan

Elevation

N

90

GroundLevel

Strain Gage

Strain gages were installed on piles, not only at ground level but also along the length of Pile 6 of the western bent (Figure 4) to evaluate the in-service, soil-structure response and to determine the response of the entire pile rather than only at the base of the abutment. All strain gages except the ones at ground level were installed prior to pile driving to provide the strain profile along the length of the pile enabling investigation of overall pile behavior. The strain gages at ground level were installed after driving. These gages on Pile 6 allow calculation of pile bending down the length of the pile and estimate of the deflected shape. Strain gages on the south face were installed to provide redundancy, locate the neutral axis, and evaluate out-of-plane movement of the pile.

Figure 3: End Bent Instrumentation (SR18)

Figure 4: Strain Gages along the Pile Length (SR18)

35

-100

102030405060708090

100110120

06/0

1/03

07/0

1/03

07/3

1/03

08/3

0/03

09/2

9/03

10/2

9/03

11/2

8/03

12/2

8/03

01/2

7/04

02/2

6/04

03/2

7/04

04/2

6/04

Air

Tem

pera

ture

(F)

Construction Day

Hottest Day

Coldest Day

Construction Temperature

DeckGirder

-100

102030405060708090

100110120

06/0

1/03

07/0

1/03

07/3

1/03

08/3

0/03

09/2

9/03

10/2

9/03

11/2

8/03

12/2

8/03

01/2

7/04

02/2

6/04

03/2

7/04

04/2

6/04

Air

Tem

pera

ture

(F)

Construction Day

Hottest Day

Coldest Day

Construction Temperature

DeckGirder

ABUTMENT BEHAVIOR

The temperature on the SR18 Bridge was measured by temperature gages located on a girder and in the deck as shown in Figure 5. The temperature measured by both gages was almost identical; however, the response of the deck was slightly slower than that of the girder.

Figure 5: Air Temperature

The rotation of the abutment was measured by tiltmeters located on the east and west faces of the end bents (Bents 1 and 6). The rotations of the abutments were filtered by taking the average of the data recorded between the time interval four hours before and four hours after the desired measurement time. The filtered rotations of both bents are plotted in Figure 6. The results indicate that both bents translated and hardly rotated. The date of deck casting is noted as the construction day. This day is significant in that it signifies the time at which the structure became integrally connected.

36

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

06/0

1/03

07/0

1/03

07/3

1/03

08/3

0/03

09/2

9/03

10/2

9/03

11/2

8/03

12/2

8/03

01/2

7/04

02/2

6/04

03/2

7/04

04/2

6/04

Long

itudi

nal M

ovem

ent (

in.)

Bent 6 Center

Bent 1 Center

Bent 6 SE

Inward

Outward

Construction Day

Hottest Day Coldest Day

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

06/0

1/03

07/0

1/03

07/3

1/03

08/3

0/03

09/2

9/03

10/2

9/03

11/2

8/03

12/2

8/03

01/2

7/04

02/2

6/04

03/2

7/04

04/2

6/04

Long

itudi

nal M

ovem

ent (

in.)

Bent 6 Center

Bent 1 Center

Bent 6 SE

Inward

Outward

Construction Day

Hottest Day Coldest Day

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

06/0

1/03

07/0

1/03

07/3

1/03

08/3

0/03

09/2

9/03

10/2

9/03

11/2

8/03

12/2

8/03

01/2

7/04

02/2

6/04

03/2

7/04

04/2

6/04

Rot

atio

n (d

egre

es)

Average Bent 1

Average Bent 6

Construction Day

Hottest Day Coldest Day

Bent 1

Bent 6 -1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

06/0

1/03

07/0

1/03

07/3

1/03

08/3

0/03

09/2

9/03

10/2

9/03

11/2

8/03

12/2

8/03

01/2

7/04

02/2

6/04

03/2

7/04

04/2

6/04

Rot

atio

n (d

egre

es)

Average Bent 1

Average Bent 6

Construction Day

Hottest Day Coldest Day

Bent 1

Bent 6

Figure 6: Abutment Rotation The movements of the abutments were measured by convergence meters, and

the displacements are plotted in Figure 7. The data was zeroed immediately prior to casting. These results indicate that the abutment movement corresponds well with temperature. For instance, as the temperature decreases (contraction phase), both abutments move toward each other as anticipated. Furthermore, the displacements were essentially identical indicating symmetrical behavior.

Figure 7: Abutment Displacement

37

The measured movement of Bent 1 was compared to the thermal movement calculated according to Equation 1 as shown in Figure 8. It can be seen that the calculated abutment movements are greater than the measured values. This difference is most likely due to backfill restraint, pile resistance and friction from the approach slab.

L = (T)L (1)

where: = thermal coefficient of concrete, taken as 6.0x10-6 /F;

T = temperature change L = half of the total span length, 367 ft/2 = 183.5 ft.

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

06/0

1/03

07/0

1/03

07/3

1/03

08/3

0/03

09/2

9/03

10/2

9/03

11/2

8/03

12/2

8/03

01/2

7/04

02/2

6/04

03/2

7/04

04/2

6/04

Lon

gitu

dina

l Mov

emen

t (in

.)

Bent 1 Center

Inward

Outward

Construction Day

Hottest Day

Coldest Day

Calculated

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

06/0

1/03

07/0

1/03

07/3

1/03

08/3

0/03

09/2

9/03

10/2

9/03

11/2

8/03

12/2

8/03

01/2

7/04

02/2

6/04

03/2

7/04

04/2

6/04

Lon

gitu

dina

l Mov

emen

t (in

.)

Bent 1 Center

Inward

Outward

Construction Day

Hottest Day

Coldest Day

Calculated

Figure 8: Calculated vs. Measured Movements

PILE BEHAVIOR

Stresses and strains along the pile length over various temperature change ranges, T, were determined by grouping the strain according to the temperature range (Figure 9). The average strains of each temperature range were calculated. The increment of the temperature change range is 10 F 5% except for T equal to 0 F. At T = 0 F, the range considered was from -1 to 1 F. It is noted that the construction temperature was considered as 60 F, and all temperature changes are referenced from this temperature.

38

Figure 9: Pile Stresses with Depth

Deflections along the pile depth were computed by integrating the moment of the area under the curvature diagram considering the deflection measured at the pile top as measured by the convergence meter located at the center of the eastern bent. The deflected shape of Pile 6 over various temperature change ranges was estimated as shown in Figure 10. The estimated deflected shapes correspond very well to the temperature change, T. Double curvature bending occurs with the inflection point located between a depth of 4 and 8 ft. It should be noted that the deflection at the bottom of the pile was not directly measured. This value was assumed for calculation of the displacement shape and was considered reasonable.

Figure 10: Pile Displacement with Depth

39

SUMMARY AND CONCLUSIONS

Based on the field study, the following conclusions have been reached.

1. The abutment responds to temperature changes, and its movement can be estimated conservatively using the theoretical thermal expansion/contraction of the superstructure, L = (T)L. The actual displacement is expected to be slightly less due to backfill restraint, pile resistance and approach slab friction.

2. The abutment primarily translates or slides longitudinally in response to

thermal expansion and contraction of the bridge. Only minor rotations of the abutment occur and for analysis purposes can be ignored.

3. Piles integrally connected with the abutment bend in double curvature.

Lateral displacements in the soil correspond directly with temperature changes. Measures to eliminate the integral abutment-pile connection can be used such as in the SR249 structure to provide for a pinned connection. This connection eliminated the double curvature response.

4. For satisfactory bridge performance, the structure must be detailed and

constructed properly. a. Piles must be constructed and oriented as designed. b. Intermediate piers should be designed to accommodate lateral

displacement or the connection must be detailed to minimize lateral force transfer. If the piers are not designed for the lateral displacement, locking of the superstructure to the intermediate piers must be prevented through isolation.

The overall goal of the research program was to provide answers to questions

that have limited the design of integral abutment bridges. While the scope of this paper is limited to the measured response, two additional phases of research have been conducted including an analytical and laboratory experimental study. The comprehensive view provided through these three phases of research has provided answers to most of the questions originally posed. An ongoing study is completing the investigation into bridge skew. While all three phases of research have been essential to providing answers, the measured in-service response was indispensable for not only the calibration of analytical models, but for a true understanding of behavior.

40

REFERENCES

1. Indiana Department of Transportation. 1992. Bridge Design Memorandum #233 Revised, Inter-Department Communication, Indianapolis, IN, 5 pp.

2. Durbin, K.O. 2001. Investigation of the Behavior of an Integral Abutment Bridge, Masters Thesis, Purdue University, West Lafayette, IN, 2001, 138 pp.

3. Chovichien, V. 2004. The Behavior and Design of Piles for Integral Abutment Bridges, Doctoral Dissertation, Purdue University, West Lafayette, IN, 489 pp.

ACKNOWLEDGEMENTS

The authors would like to gratefully acknowledge the Indiana Department of Transportation. Funding for the research program conducted by Purdue University was provided through the Joint Transportation Research Program (JTRP) through Project No. SPR-2393. Thanks are also extended to Katrinna Durbin and David Fedroff for their contributions to this research study.

41

New York State Department of Transportation's Experience with Integral Abutment Bridges

Arthur P. Yannotti, P.E. New York State, Department of Transportation

Sreenivas Alampalli, PhD, P.E. New York State Department of Transportation Harry L. White 2nd, P.E. New York State Department of Transportation

ABSTRACT The New York State Department of Transportation (NYSDOT) has been using integral abutment bridges since the late 1970's. Since that time, the design methodology and details have been modified several times to improve performance. Semi-Integral abutments were introduced in 1998. Approximately 450 integral and semi-integral abutment bridges have been constructed in New York and thus far, their in-service performance has been excellent. They are the preferred abutment type for NYSDOT. This paper examines the evolution of the design and construction practices and explains the reasons for the modifications. HISTORY

Traditionally, bridges in New York were constructed with a joint in the deck slab at abutments to accommodate thermal movements. By the 1970's, multiple span bridges were commonly designed as continuous spans, thus eliminating the deck joints over the piers. Deck joints were of a number of types, including various types of neoprene seals, finger joints and open trough systems. However, all types were prone to leak and allow water containing road salt to drain onto the underlying superstructure beams, bearings, abutment backwalls and bridge seats.

In order to eliminate the "leaking joint" problem, the first integral

abutment bridges constructed by NYSDOT were built in the late 1970's. These were typically single span steel bridges with a span length under 100 feet. A single row of steel H piles with the strong axis parallel to the girders was used so that bending occurred about the weak axis. The steel piles were extended into a short abutment stem. The steel girders were erected on the steel piles and attached to them by welding to a cap plate that was welded to the top of the piles. The girders and piles were then encased in concrete as the deck slab was placed creating an integral type abutment. Design assumptions included equal distribution of the vertical load to the piles. Bending moments in the piles were ignored, and the abutments and wingwalls were designed for full passive earth pressure. A section of an early typical integral abutment with a steel superstructure is shown in Figure 1.

42

Figure 1. Early Steel Superstructure Integral Abutment

Superstructures with adjacent prestressed concrete box beams are commonly used in New York. Early attempts were made in the 1980's to adapt integral a