Design Guide for Bridges for Service Life (2013)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

343 8.1 introduction A jointless bridge has a continuous deck with no expansion joints over the super- structure, abutments, and piers. Jointless bridges are commonly referred to as integral bridges. In this type of bridge structure, all movement due to thermal, creep, and shrinkage strain is accommodated either within the system itself or at the ends of the approach slabs where the slabs abut the roadway pavement. Because there are no joints, ride quality is improved, and maintenance can be greatly reduced. Leaking deck joints have been a major cause of bridge deterioration and reduced service life, especially where roadway drainage carrying deicing chemicals can spill onto the bridge elements below. Elimination of bridge-deck expansion joints is there- fore an important consideration in bridge system selection to provide long-term service life, as discussed in Chapter 2. This chapter summarizes the design, construction, and maintenance provisions related to the use of jointless bridges. 8.2 hiStory oF jointLeSS bridgeS A detailed history of jointless bridges is provided by Burke, Jr. (2009). The following is a brief summary of that history. The Ohio highway department was the first to rou- tinely use continuous construction for multispan bridges, beginning in 1930, although expansion joints were present at the abutments. The next step—elimination of deck joints at the abutments—was undertaken by the Ohio Department of Transportation in 1938 with construction of the Teens Run Bridge near Eureka, Ohio. The five-span, continuously reinforced, concrete slab bridge was the first integral bridge in the United States. 8 JOiNTLESS BRiDGES

344 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Use of integral bridges continues to increase both in the United States and abroad. Several countries adopting the practice are Japan (1996) and South Korea (2002), with the United Kingdom recently using integral bridges in routine applications. A survey conducted in 1987 indicated 20 of the 30 respondent transportation departments were using integral construction details for continuous bridges. One point of contention is the length limit for which integral systems can be applied. North Dakota, South Dakota, and Tennessee have used continuous integral bridges with steel main members at spans in the 300-ft range for many years. Similar bridges with concrete main members have been constructed at lengths of 500 to 800 ft in Kansas, California, Colorado, and Tennessee. The Tennessee DOT is in the lead when it comes to span length. The Long Island Bridge of Kingsport, built in 1980, has 29 continuous spans without deck expan- sion joints. Additionally, Tennessee recently completed the seven-span Happy Hollow Creek Bridge, which is a curved, prestressed concrete bridge with a total length of 1,175 ft. Despite its extreme length, the structure falls within the Tennessee DOT’s policy for integral bridges. Seamless bridges, another type of jointless bridge introduced by SHRP 2 Project R19A for practice in the United States, allow elimination of expansion joints even at the end of the approach slab. The seamless bridge system was first introduced by Bridge et al. (2000) in Australia for use with continuously reinforced concrete pavement for the approach roadways. Most commonly used pavements in the United States, however, are jointed plain concrete (JPCP) and flexible pavements, which require a modified application. Seamless bridges do not have any joints, even at the ends of an approach slab (hence seamless). Instead, a pavement transition zone is used to dissipate the thermal displacements of the bridge. The transition zones can be rather lengthy rela- tive to the bridge. The benefit, however, is that movements at the end of the transition zone are very small. 8.3 tyPeS oF jointLeSS bridgeS Three main types of jointless bridges are described in this chapter: integral and semi- integral jointless bridges, which are commonly used in practice, and a new class of jointless bridges referred to as seamless jointless bridges. The main characteristic of the seamless system is that expansion joints are eliminated altogether, and the bridge deck is connected to the approach road pavement without a joint. Figure 8.1 is a rendering of a typical layout of a jointless bridge, shown with the superstructure cast in an integral abutment. 8.3.1 integral Bridges Integral bridges have the superstructure constructed monolithically with the abutments, encasing the ends of the superstructure within the backwall. The main characteristics of integral bridges are their jointless construction and flexible abutment foundations. The system is structurally continuous, and the abutment foundation is flexible longitu- dinally. The movement of the superstructure is accommodated by the foundation.

345 Chapter 8. JOiNTLESS BRiDGES Figure 8.1. Elements of jointless integral bridges. Figure 8.1. Elements of jointless integral bridges. Figure 8.1 shows, schematically, the main elements of an integral bridge system, which consist of bridge deck, girders, integral cast abutments, and approach slabs. The bridge movement is accommodated at the ends of the approach slabs. Sleeper slabs are commonly used to provide vertical support for the ends of the approach slab where the slabs abut the roadway pavement (not shown in Figure 8.1). In addition, jointless integral bridges can be continuous multispan structures with intermediate piers (also not shown in Figure 8.1). Various details are described in greater detail in Section 8.7; see Section 8.7.3 for further discussion of sleeper slabs. 8.3.2 Semi-integral Bridges Semi-integral bridges are defined as having an end diaphragm that serves as the abut- ment backwall and that is cast encasing the superstructure ends. In this system, the superstructure rests on expansion bearings, and the end diaphragm is not restrained longitudinally with respect to the pile cap or abutment stem. The deck may be slid- ing or cast monolithically with the backwall, but it does not have a joint above the abutment. The foundation is rigid longitudinally, where superstructure movement is accommodated through bearings. The main elements of a semi-integral bridge system consist of bridge deck, girders, abutment stem and bearing seat, integral cast diaphragm backwall, approach slab, and sleeper slab. The bridge movement is accommodated at the ends of the approach slabs. Greater detail is provided in Section 8.7. 8.3.3 Seamless Bridges The seamless bridge system is characterized by eliminating the need for expansion joints, even at the ends of the approach slabs, while limiting the longitudinal expan- sion and contraction of the bridge superstructure. Imposing a limitation on longitudi- nal expansion and contraction of the bridge superstructure results in development of longitudinal forces that need to be resisted with appropriate design features. Longitu- dinal expansion and contraction are limited, but not eliminated. This philosophy is

346 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE used to reduce the level of longitudinal forces that can be developed, while making the gap between the end of the transition zone and the start of pavement manageable (to less than about 0.25 in.). The foundation requirements are very similar to those of integral abutments. A seamless bridge system for jointed pavement types used in the United States was developed by SHRP 2 Project R19A and is described in Appendix E. Additional information on this system can be found in Ala (2011) and in Ala and Azizinamini (in press a and b). Figure 8.2 shows, schematically, the main elements of the seamless bridge system, which consist of the bridge deck, transition zone, and roadway pavement. The bridge movement is accommodated within the transition zone, and the movement at the end of the transition zone is relatively small. This eliminates having expansion joints at the end of transition zone, where the roadway pavement starts. The thickness of the transition zone and approach slab near the abutment is increased to account for lack of support from the soil below. The assumption is that the transition zone near the abutment has no support and resists the imposed loads by flexure. 8.3.4 Advantages of Jointless Bridges Henry Derthick, former engineer of structures at the Tennessee DOT, once stated, “The only good joint is no joint.” In keeping with this statement, known advantages of the jointless bridge systems include • Lower initial cost; • Lower maintenance cost; • Prevention of leakage of moisture to bridge elements below deck, resulting in longer service life; • Improved ride quality; Bridge Approach Abutment Transition Zone JPCP Figure 8.2. Seamless bridge system.

347 Chapter 8. JOiNTLESS BRiDGES • Easier and faster construction; • Easier inspection; • Simplified bridge details; • Elimination of bearings (except for semi-integral bridges); • Ideally suitable for bridges with skew and curvature or located in high seismic areas; and • Enhanced buoyancy resistance of the bridge. Because of these advantages, many DOTs have started using jointless bridges; however, the design provisions vary significantly from one state to another. 8.3.5 Cost-Effectiveness of Jointless Bridges Jointless bridges have a significant cost savings advantage compared with traditional bridges with expansion joints. Cost savings are realized both during initial construc- tion and throughout the life of the bridge with reduced maintenance. This is particu- larly true for bridges with integral abutments. Most components of typical bridges with joints and jointless bridges are similar in construction and cost (e.g., deck, beams, cross frames). Thus, a comparison is made relative to the different components that distinguish each type of construction (e.g., the costs of the abutments and expansion joints). In addition, because unit pricing of each item is consistent neither from region to region nor over time, a qualitative comparison is made using relative costs. It is recognized that different states and municipalities have different specifications and construction techniques; however, initial construction of a typical abutment with an expansion joint will most often include the following: • Excavation; • Two rows of piling (in some cases); • Concrete cap, stem, diaphragm, and backwall with reinforcing (three pours); • Elastomeric bearings, per beam; • Precision-cast bridge seats for bearings; • Expansion joint; and • Porous backfill. Similarly, typical construction of an integral bridge includes • Excavation; • One row of piling; • Concrete cap, integral backwall, and diaphragm (two pours); • Sleeper slab with reinforcement; and • Porous backfill.

348 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE The important differences between an integral abutment and a traditional abut- ment with a jointed deck include a lack of expansion joint, no bearings, reduced num- ber of required piles, reduced number of concrete pours, and inclusion of a sleeper slab. Taking these differences into consideration, the cost savings are readily apparent. The sleeper slab adds a few cubic yards of concrete and an extra detail to the cost, but removal of the expansion joint, removal of the second row of piling (overall reduced number of piles), reduction in the required amount of concrete and reinforcing in the stem–backwall unit, and elimination of beam bearings at the abutment greatly reduce the cost relative to adding the sleeper slab. The overall reduction in initial construction cost can be over 40% for each abutment. (The percentage difference was estimated using Ohio 2010 bid planning costs for a typical 32-ft-wide bridge.) The life-cycle cost of the two types of abutments differs, as well. A service life of 100 years is used for the comparison, although it is acknowledged that differences in estimated service life can affect the parameters. For standard jointed bridges, common armored expansion joints typically require gland replacement on the order of 8 to 12 years, depending on condition severity. Additionally, the entire joint including armor will need to be replaced along with the deck at least once (based on an estimated deck life of 30 to 50 years). Again, this estimate depends on the severity of the conditions. It is also expected that the bearings will need replacing at least once over the life of the bridge, the cost of which includes jacking of the bridge. Similar to expansion joints, the sleeper slab joint seal, if used, will require replace- ment on the same, or at least similar, schedule. Likewise, the deck will need replacing on a similar schedule. Deck replacement of an integral bridge requires additional con- sideration of certain construction items, but it does not require a significant increase in construction cost compared with traditional deck replacement. For comparative purposes, consider that a typical bridge abutment with expansion joints will require • Expansion gland replacement every ~10 years, • Deck replacement every ~50 years, • Expansion joint replacement every ~50 years, and • Bearing replacement every ~50 years. A typical jointless bridge abutment will require • Sleeper slab joint seal (if used) replacement every ~10 years and • Deck replacement every ~50 years. Cost is similar for deck replacement and gland and seal replacement; however, there is a significant increase in cost when replacing the expansion joint. A qualitative cost comparison is presented in Figure 8.3 and Figure 8.4. Both fig- ures consider the difference in the initial cost at Year Zero and the accumulated cost difference over the life of the bridge. The figures show the difference in costs; that is, similar costs have been removed from the equation (i.e., the cost of replacing the deck itself is removed from the equation because it is similar for both types of bridge).

349 Chapter 8. JOiNTLESS BRiDGES 0 20 40 60 80 100 Relative Total Cost Life of Bridge (yr.) Jointless With Exp. Joints Joint Repair, 10-yr cycle Cost difference Expansion Joint Replacement, 50-yr cycle Abutment Construction Abutment Construction Joint Maintenance Total Difference Relative Cost Jointless With Exp. Joints Figure 8.3 shows the estimated cost comparison through the 100-year life of the struc- ture. Figure 8.4 shows the cost comparison differentiating the initial difference in the construction costs, the lifetime maintenance costs, and the overall total cost difference over the life of the bridge. 8.4 FActorS AFFecting PerFormAnce oF jointLeSS bridgeS Factors affecting the performance of jointless bridges include curvature, skew, bear- ing, and connection of superstructure and substructure. Other factors that should be considered include site conditions, deterioration of piles, and abutment walls. 8.4.1 Curvature Horizontal curvature changes the internal forces of integral abutment bridges. These changes are more important for bridges with either of the following conditions when the length and radius are measured at the centerline of the bridge: length-over-radius ratio greater than 0.5 or radius of curvature less than 1,000 ft. Figure 8.3. Lifetime cost analysis of jointed versus jointless bridge over time. Figure 8.4. Lifetime cost differential analysis of jointed versus jointless bridge.

350 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE If a curved bridge does not have either of these conditions, the response of the curved bridge can be estimated by the response of a straight bridge of the same length (Doust 2011). This estimation is not valid for the internal forces during construction. 8.4.2 Skew In skewed integral abutment bridges, the soil passive pressure developed in response to thermal elongation can prevent the transverse movement of the bridge. Appendix B provides additional detail on this subject. However, if the friction created by contact between soil and abutment wall is insufficient, and depending on the transverse stiffness of the abutment, either significant transverse forces or significant transverse movements could be generated (Oesterle et al. 2005). 8.4.3 Bearings In multispan integral bridges with rigid piers, the superstructure is commonly seated on piers through the use of bearing devices. In curved bridges or wide, straight bridges, fixed bearings are not recommended except at the points of zero movement. In curved integral bridges, there may be no point of zero movement throughout the bridge. Guided bearings are not recommended for curved or wide, straight integral bridges because the displacements do not happen in just one particular direction. In such cases, guided bearings behave like fixed bearings, creating large internal forces at the piers. Multidirectional elastomeric or sliding bearings are the proper types of pier bear- ings for integral bridges. If such bearings are used, the superstructure movement is mainly controlled by the integral abutments. 8.4.4 Connection Between Superstructure and Substructure The choice of how the superstructure is connected to the substructure has a significant impact on how the bridge will behave. Choosing the abutment type sets the major design considerations for the bridge with respect to jointless behavior. Methodology for designing the abutments for the various types of jointless bridges is presented in Section 8.6. The consideration for the connection to the piers is equally important. The super- structure can be made integral with a pier or designed to transfer loads to the pier with more traditional assumptions. It is important to note in the planning stages how the pier will react as the bridge expands and contracts. Piers must be sufficiently designed, whether they are intended to flex with the structure (slender piers) or are designed to resist the movement (stout piers). The latter case is generally not preferable as it often leads to overdesigned substructures, because the movement from the continuous deck superstructure can generally be accommodated by simply using an expansion bearing to accommodate the movement. The design of different pier types is discussed in more detail in Section 8.6.2.9.

351 Chapter 8. JOiNTLESS BRiDGES 8.4.5 other Considerations Other factors that can affect the performance of jointless bridges are primarily associ- ated with foundation conditions. 8.4.5.1 Site Condition Integral abutments for jointless bridges are usually supported on a single row of piles to provide flexibility. Also, piles are typically used to minimize settlement of the abutment and differential settlement within the superstructure. However, when rock is close to the substructure bearing surface, a different type of foundation may be required. One solution is to use semi-integral abutments, described in Section 8.3.2, in which the abutment foundations are supported directly by and keyed into the rock. The end dia- phragm serving as the abutment backwall and encasing the superstructure ends rests on bearings supported by the abutment foundations. The deck and approach slabs are cast monolithically with the backwall. The abutment foundations are rigid, and the longitudinal movement of the superstructure is accommodated through the bearings. As an alternative to the semi-integral abutments, spread footings may potentially be considered for integral abutments when rock is close to the surface, particularly for single-span bridges, and when the foundation is assumed to slide. However, sig- nificant friction forces would have to be overcome, and this concept has typically not been considered. Differential settlement would be another concern for the use of spread footings on soils to support abutments for multispan continuous bridges, but Moulton et al. (1985) and Hearn (1995) indicate that the magnitude of settlement for abutments supported by spread footings is similar to that for abutments supported by piles. However, there is very little experience with the actual use of spread footings for integral abutments either on rock or on competent soil near the surface. Hence, it is recommended that experience be gained by starting with relatively short simple-span bridges. Use can then progress to longer structures and multispan structures as success- ful experience is gained. The following recommendations pertain to abutments supported by shallow spread footings, in which end movement may be accommodated by sliding: • For footings founded on rock, a layer of granular fill should be used (on top of a leveling layer of fill concrete, as needed) between the footing and rock to facilitate sliding. The footing should not be keyed into rock. • The abutment wall should be designed for shear and moments resulting from both expansion and contraction movements. The resistance to contraction should in- clude friction on the bottom of the footing and passive soil pressure from the berm soil on the front face of the abutment. • Sufficient drainage, distance from the face of the slope, and slope protection are essential to keep soil from washing out below the footing. For footings supported on a layer of granular soil for sliding on rock, use of geotextile material may be considered to contain the granular soil. For footings supported on soil, mechanical stabilization of the soil below the footing may be appropriate.

352 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Another possible solution for use in conditions in which rock is close to surface is to drill large-diameter holes in the rock and use piles, which would consequently allow the use of typical integral abutment construction. It must be noted that the concepts of inte- gral abutments supported on spread footings or supported on piles placed in holes drilled into rock are not common practice. These two concepts are suggested for consideration when site conditions would otherwise inhibit the use of typical jointless construction. 8.4.5.2 Deterioration of Piling Accelerated pile deterioration is generally not considered except in specialized cor- rosive locations. Designers should consult with either a geotechnical engineer or ge- ologist to mitigate possible impacts for this condition. More commonly, corrosion is thought of as a minimal concern for piles, but it has been recorded (Beavers and Durr 1998) and more recently evaluated (Decker et al. 2008). Additionally, the state of Iowa has been investigating deterioration of piles just below the pile cap of integral abutments (Iowa Department of Transportation 2010)). Initial results note that Iowa investigators have discovered corrosion immediately below abutment footings of what would be considered normal conditions. Piling deterioration is of increased importance for integral abutments because of the additional strains placed on the substructure from the longitudinal expansion of the superstructure. The potential for section loss based on soil conditions should be accounted for as presented by Article 10.7.5 of the LRFD Bridge Design Specifica- tions (LRFD specifications), which states minimum considerations for the effects of corrosion and deterioration of piling (AASHTO 2012). Adhering to these guidelines should provide sufficient protection against advanced corrosion and thus failure of the integral abutment system. 8.4.5.3 Jointless Bridge Abutments with Mechanically Stabilized Earth Walls If setting the abutment on top of an embankment slope or reducing the total bridge length is impractical, full-height abutments with a mechanically stabilized earth (MSE) retaining wall may be considered in the design of jointless bridges. When MSE walls are used, steps must be taken to prevent excess pressure on the retaining wall intro- duced by the movement of the backwall and pile. For integral abutments, per FHWA Demonstration Project 82 (Elias et al. 1997) the horizontal force and its distribution with depth may be developed using pile load- deflection methods (p-y curves) and added as a supplementary horizontal force to be resisted by the MSE wall reinforcements. This force will vary depending on the level of horizontal load, pile diameter, pile spacing, and distance from the pile to the back of the panels. Per Demonstration Project 82, the following additional design details have been used successfully: first, providing a clear horizontal distance of about 1.5 ft (0.5 m) between the back of the panels and the front edge of the pile; and second, providing a casing around the piles, through the reinforced fill, where significant negative skin friction is anticipated.

353 Chapter 8. JOiNTLESS BRiDGES Where pile locations interfere with the reinforcement, specific methods for field installation must be developed. Simple cutting of the reinforcement is not permissible. For integral abutments and for seamless bridges, MSE walls can still be used, but they must be sufficiently isolated from the soil movement caused by the movement of the piles. Alternatives suggested by Nicholson et al. (1997) for the use of MSE walls with jointless bridge abutments are shown in Figure 8.5. Figure 8.5a illustrates the use of a semi-integral abutment or stub integral abut- ment on spread footings. In this approach, the MSE reinforcement should be designed for the sliding forces in the bearings of the semi-integral abutment or the frictional sliding forces of the spread footings. Figure 8.5. Alternatives to integral full-height wall abutments using reinforced-soil retaining structure. Source: Nicholson et al. 1997.

354 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Figure 8.5b illustrates use of pile encased in a pressure-relieving sleeve that isolates the pile movements from the surrounding soil. Hassiotis (2007) has reported tests with an integral abutment supported on piles encased in corrugated steel sleeves backfilled with sand. Lui et al. (2005) indicate that the Iowa DOT criteria for use of MSE walls with integral abutments requires each pile to be encased in a corrugated metal sleeve. The reinforced soil should include sand up to the bottom of the sleeve, and the remain- der of the sleeve should be filled with bentonite to the top. Figure 8.5c illustrates the use of a semi-integral abutment supported on a pier in front of the MSE wall. No additional considerations are necessary for semi-integral abutments because the lateral movement is dissipated through the bearings. 8.5 StrAtegy SeLection ProceSS Each type of jointless construction has a range of parameters that is appropriate for particular bridges or provides various advantages over another type of system. The fol- lowing tables provide guidance in selecting bridge type based on limiting parameters. Table 8.1 assists in the selection of the primary system and provides three options with regard to foundation types: integral, semi-integral, and seamless. The maximum length of each system is not set; rather, it is based on design calculations. Typical details that could be employed with each system are provided, and corresponding sections are noted. Table 8.1 also briefly describes the major advantages and disadvantages of each system. Relatively, the semi-integral abutment type provides larger longitudinal movement capabilities than the integral and seamless systems. The trade-off is the need to add sliding bearing, which will result in reduction in service life. As indicated in Table 8.1, the relative maintenance of integral and seamless abutment types is lower than maintenance costs for the semi-integral abutment type. The main reason for the difference in cost is the need for incorporating bearing at the abutment. All three abutment types are applicable to existing bridges when it is desirable to eliminate the expansion devices at the abutment. In some situations, cost may prevent use of integral or semi-integral abutment types. Table 8.2 provides further guidance on the substructure type that is appropriate for use with each type of jointless bridge. As indicated in Table 8.2, for the integral abutment type, H-piles or prestressed piles or concrete-filled tube (CFT) piles could be used. In the case of prestressed piles, the relative lateral displacement movement is lower than H-piles or CFT piles. In the case of prestressed piles, cracking and corro- sion is of concern. In the case of H-piles and CFT piles, corrosion is of concern. These concerns are reflected qualitatively in assessing the potential for each abutment type to achieve 100 years of service life. Table 8.3 provides guidance on the types of connections and bearings used at the piers when used in a jointless bridge. Considering the discussions provided in previ- ous sections, the four right-hand columns of Table 8.3 provide qualitative rankings of each option with respect to maximum longitudinal movement capabilities, relative maintenance cost, applicability to existing bridges, and potential to achieve 100+ year service life.

355 Chapter 8. JOiNTLESS BRiDGES tA B LE 8 .1 . St rA te gy t Ab Le F or A bu tm en t ty Pe S eL ec ti on in j oi n tL eS S br id ge S: S tr Ai gh t br id ge S St ra te g y M ax im u m B ri d g e Le n g th U se d Ty p ic al D et ai ls A d va n ta g es D is ad va n ta g es Q u al it at iv e Lo n g it u d in al M o ve m en t D em an d Q u al it at iv e M ai n te n an ce R an k in g A p p li ca b il it y to E xi st in g B ri d g es In te gr al Es ta bl is he d ba se d on d es ig n pr ov is io ns s ta te d in t he G ui de Fi gu re 8 .3 1, Fi gu re 8 .3 3, Fi gu re 8 .3 4, Fi gu re 8 .3 5 El im in at es n ee d fo r be ar in gs . D iffi cu lt to in sp ec t da m ag e to p ile s du e to pi le m ov em en t. Lo w Lo w Ye s Se m i- in te gr al Es ta bl is he d ba se d on d es ig n pr ov is io ns s ta te d in t he G ui de Fi gu re 8 .3 6, Fi gu re 8 .3 7, Fi gu re 8 .3 8 Pr ov id es r ed uc ed lo ng itu di na l f or ce t ra ns fe r to p ile s. N ee ds b ea rin gs . M ed iu m M ed iu m Ye s Se am le ss Es ta bl is he d ba se d on d es ig n pr ov is io ns s ta te d in t he G ui de Fi gu re 8 .2 5, Fi gu re 8 .2 6 El im in at es n ee d fo r ex pa ns io n jo in ts . El im in at es c on ce rn s w he n th er e is s ke w o r cu rv at ur e. El im in at es n ee d fo r be ar in gs a nd s le ep er s la b. Po ss ib le , i ni tia l h ig he r co st . D iffi cu lt to in sp ec t da m ag e to p ile s du e to pi le m ov em en t; lo ng tr an si tio n zo ne o ff br id ge . Lo w Lo w Ye s

356 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE tA B LE 8 .2 . St rA te gy t Ab Le F or F ou n dA ti on A t Ab ut m en tS in j oi n tL eS S br id ge S: S tr Ai gh t br id ge S St ra te g y Ty p ic al D et ai ls A d va n ta g es D is ad va n ta g es Q u al it at iv e Lo n g it u d in al M o ve m en t D em an d P o te n ti al f o r A ch ie vi n g 1 0 0 + Y ea rs o f Se rv ic e Li fe In te gr al a bu tm en t H -p ile Fi gu re 8 .3 3 Ec on om ic al fo r sm al l m ov em en ts . Ea sy t o co ns tr uc t. Re la tiv el y lo w st re ng th , d uc til ity a nd bu ck lin g ca pa ci ty . M ed iu m M ed iu m Pr es tr es se d pi le Fi gu re 8 .3 2 Ve ry s tif f a nd h ig h ax ia l l oa d ca pa ci ty . Pr on e to c on cr et e de te rio ra tio n an d co rr os io n of s tr an ds . Lo w M ed iu m C FT p ile Fi gu re 8 .3 2 C FT h as h ig h st re ng th a nd d uc til ity a nd hi gh er b uc kl in g ca pa ci ty , w hi ch w ill ac co m m od at e m uc h la rg er b rid ge le ng th . H ig he r in iti al c os t. H ig h M ed iu m Se m i-i nt eg ra l a bu tm en t A ny fo un da tio n ty pe c ou ld b e us ed w ith t he s em i-i nt eg ra l a bu tm en t. Se am le ss Th e st ra te gy fo r se le ct io n of s ea m le ss fo un da tio ns is t he s am e as fo r in te gr al a bu tm en ts . T he d et ai lin g fo r se am le ss b rid ge s is pr im ar ily in t he t ra ns iti on z on es a t th e en ds o f t he b rid ge a s de ta ile d in S ec tio n 8. 6.

357 Chapter 8. JOiNTLESS BRiDGES tA B LE 8 .3 . St rA te gy t Ab Le F or c on n ec ti on b et w ee n P ie rS A n d Su Pe rS tr uc tu re in j oi n tL eS S br id ge S: S tr Ai gh t br id ge S St ra te g y D et ai l Fi g u re A d va n ta g es D is ad va n ta g es Q u al it at iv e Lo n g it u d in al M o ve m en t D em an d Q u al it at iv e M ai n te n an ce R an k in g D eg re e o f D if fi cu lt y to A p p ly to E xi st in g B ri d g es P o te n ti al f o r A ch ie vi n g 1 0 0 + Y ea rs o f Se rv ic e Li fe G ird er s co nt in uo us ov er p ie r In te gr al - fr am e ac tio n Fi gu re 8 .2 7a , Fi gu re 8 .2 9 El im in at es n ee d fo r be ar in gs o ve r pi er . M ay c au se tr an sv er se cr ac ki ng in t he pi er . Lo w Lo w M ed iu m H ig h Fi xe d be ar in g (r ot at io na l m ov em en t al lo w ed ) Fi gu re 8 .2 7b N o lo ng itu di na l m ov em en t re qu ire m en t fo r be ar in g ov er p ie r. M ay c au se tr an sv er se cr ac ki ng in t he pi er . Lo w M ed iu m M ed iu m M ed iu m Ex pa ns io n be ar in g Fi gu re 8 .2 7c Re du ce d be nd in g of p ie r co lu m n. Be ar in g de si gn ed fo r bo th r ot at io n an d lo ng itu di na l m ov em en t. H ig h M ed iu m Lo w Lo w G ird er s no t co nt in uo us ov er p ie r Si m pl e fo r de ad a nd co nt in uo us fo r liv e lo ad Fi gu re 8 .2 7b , Fi gu re 8 .2 7c , Fi gu re 8 .2 8, Fi gu re 8 .3 0 El im in at es jo in ts an d pr ot ec ts gi rd er e nd s; vi ab le o pt io n fo r se is m ic r et ro fit . Re st ra in t m om en ts a nd cr ac ki ng in di ap hr ag m s. Va rie s w ith be ar in g ty pe Lo w Lo w H ig h Li nk s la b Fi gu re 8 .2 7d , Fi gu re 8 .3 1 Lo w c os t. M ay c ra ck a nd ca us e le ak ag e ov er jo in t. Va rie s w ith be ar in g ty pe H ig h Lo w M ed iu m

358 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 8.6 deSign ProviSionS For jointLeSS bridgeS Design procedures for integral abutment bridges can range from a simplified method of analysis to a more detailed approach. This section includes provisions for both. Sec- tion 8.6.1 provides requirements for using the simplified approach, and Section 8.6.2 describes detailed methods of analysis that should be used if Section 8.6.1 require- ments are not met. 8.6.1 Simplified method of Analysis The simplified analysis method is provided to eliminate many design steps for simple bridges that do not require detailed analysis. A bridge should meet the following re- quirements for use of the simplified analysis method (VTrans 2009): • The skew angle should be less than or equal to 20°; • The bridge can be straight or curved, but with parallel girders; • Abutments and piers should be parallel; • Abutment height should be limited to 13 ft.; • Heights of the abutment at the bridge ends should be approximately the same (maximum difference 20%); • The slope of the bridge in the longitudinal direction should be less than 5%; • The length of the wing wall, attached to the abutment, should be less than 10 ft; and • The length of the pile should be greater than or equal to 16 ft. The main characteristics of the simplified method of analysis are as follows: • The internal forces of the superstructure and substructure are obtained using a two-dimensional (2D) analysis. • Conservatively, the superstructure may be assumed to be simply supported at the two abutment ends. • The design of the pile can be accomplished by separating the pile from other bridge elements and treating it as an axial member. The moment capacity of the pile section is affected by the applied axial load. As the pile axial load increases, the moment that causes the formation of the plastic hinge in the pile will decrease. Once the plastic hinge is formed, the pile head can be assumed to act as a pin. In the simplified approach the pile is modeled as an axial element with one end (the end close to the abutment) subjected to axial load and constant moment equal to the moment capacity of the pile cross section. The interaction equation in LRFD specifications Article 6.9.2.2 can be employed to determine the capacity of the pile section.

359 Chapter 8. JOiNTLESS BRiDGES 8.6.2 Detailed method of Analysis If the requirements for the simplified method of analysis are not met, the bridge must be analyzed using the detailed analysis approach. In this approach, there is no limita- tion on total skew angle and so forth. For years, jointless integral abutment bridges were designed with an imposed maximum limitation on total bridge length, with max- imum length of steel structures less than that of concrete. Design provisions for the detailed method of analysis, as outlined in this chapter, do not include bridge length limitations; rather, one must meet the specified design provisions. In the detailed method of analysis, the superstructure and substructure are modeled as an integral system using three-dimensional (3D) finite element analysis, with girder webs modeled using shell elements. Use of grid-type analysis should be carried out with caution and is not recommended, primarily because the torsional stiffness of line elements in some of the available commercial programs does not include the contribu- tion of warping torsional stiffness. 8.6.2.1 Loads 8.6.2.1.1 Dead Loads Dead loads include the weight of all components including superstructure and sub- structure elements and include all permanent loads in accordance with LRFD speci- fications Article 3.5. The dead loads are distributed to the foundation through tradi- tional assumptions or in accordance with the owner’s bridge design provisions. 8.6.2.1.2 Live Loads Live loads and their associated impact are applied in accordance with LRFD speci- fications Article 3.6 or in accordance with the owner’s bridge design provisions. For integral abutments and piers, application of live loads will cause rotation and induce moments that will need to be considered in the design. Horizontal live load (braking force and centrifugal force) are subject to distribu- tion with respect to the stiffness of the integral and semi-integral abutments. In tradi- tional design, longitudinal forces are distributed to the substructure based on bearing fixity (expansion versus fixed against horizontal movement) and relative substructure flexibility. For jointless bridges, the backfill is in full contact with the end diaphragm (backwall) and provides a significant amount of stiffness relative to the other substruc- ture components. For integral abutments, in which the bearing condition is fixed, it is acceptable to assume for bridges with one to three spans that the longitudinal forces are absorbed by the passive pressure and stiffness provided by the backfill soil. This should be verified by a geotechnical engineer. As bridges get longer and additional substructure units are introduced, a relative stiffness analysis should be performed. However, even with multiple piers with some having fixed-expansion bearings, inte- gral abutments can be expected to absorb as much as 80% of the longitudinal force.

360 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 8.6.2.1.3 Soil Load on Abutment The magnitude of soil pressure behind the abutment wall and the nonlinear distribu- tion of this pressure depend on wall displacement, soil type, depth, pile stiffness, and also the direction of the displacement (Faraji et al. 2001). As a wall moves toward the backfill, passive pressure is engaged, and when it moves away, active pressure and surcharge pressure may be generated. Full passive pressure builds up for relatively long bridge lengths. For shorter bridge lengths, only part of the passive pressure is developed for expansion because thermal expansion is limited. For all bridges, the maximum passive pressure force is calculated as shown in Equation 8.1: P K H12p p 2γ= (8.1) where Pp = passive pressure force, Kp = passive pressure coefficient, g = unit weight of soil, and H = height of soil face. Kp is not necessarily the maximum Kp associated with full passive pressure. The value of Kp should be calculated using Figure 8.6 and Figure 8.7 (Clough and Duncan 1991). The extreme values for expansion and contraction are proportional to the height of the wall. The movement required to reach the maximum passive pressure is on the order of 10 times the movement required to reach the active soil pressure. The movement required to reach the extreme pressures are larger for loose soils than that for dense soils (Figure 8.6 and Figure 8.7, respectively). Table 8.4 highlights the move- ments required to achieve maximum pressures. The force-deflection relationship should be based on the design curves (Barker et al. 1991) shown in Figure 8.6 and Figure 8.7 (Clough and Duncan 1991). The stiffness of the springs behind the abutment wall is nonlinear and depends on the type of soil. 8.6.2.1.4 Soil Load on Piles The design of piles should consider the soil–structure interaction by using p-y curves such as in the procedure recommended by the American Petroleum Institute for off- shore platform design (API 1993). Soil–structure interaction analysis of piles can be performed using available soft- ware: LPILE, COM624P, and FB-MultiPier are several that use this approach. Further information on this topic is provided in Article 10.7 of the LRFD specifications.

361 Chapter 8. JOiNTLESS BRiDGES Figure 8.6. Relationship between wall movement and earth pressure. Source: Clough and Duncan 1991. 8.6.2.1.5 Thermal Loads In order to account for the effect of temperature changes in the design of jointless bridges, two effects should be considered: the effect of uniform temperature change and the effect of temperature gradient within the structure. These two effects are ex- plained in the following subsections. 8.6.2.1.5a Uniform Temperature Change The calculation of uniform temperature changes should be in accordance with LRFD specifications Article 3.12.2, in which two procedures are recommended, Procedure A and Procedure B, as described in this subsection. Either procedure may be used for concrete deck bridges that have concrete or steel girders. For all other types of bridges, Procedure A should be employed. Table 8.5 presents the temperature ranges used in Procedure A to calculate the design thermal movements. The difference between these values and the base construc- tion temperature should be used to calculate thermal movements.

362 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Source: Clough and Duncan (1991). Figure 8.7. Relationship between wall movement and earth pressure for a wall with compacted backfill. Figure 8.7. Relationship between wall movement and earth pressure for a wall with compacted backfill. S rce: Clough and Duncan 1991. tABLE 8.4. APProximAte mAgnitudeS oF movementS reQuired to reAch extreme SoiL PreSSure condition Type of Backfill Values of Δ/H Active Passive Dense sand 0.001 0.01 Medium-dense sand 0.002 0.02 Loose sand 0.004 0.04 Compacted silt 0.002 0.02 Compacted lean claya 0.01 0.05 Compacted fat claya 0.01 0.05 Source: Clough and Duncan 1991. Note: D = movement of top of the wall required to reach extreme soil pressure, by tilting or lateral translation; H = height of the wall. a Under stress conditions close to the minimum active or maximum passive pressures, cohesive soils creep continually. The movement shown would produce temporary passive pressures. If pressures remain constant with time, the movements shown will increase. If movement remains constant, active pressures will increase, while passive pressures will decrease.

363 Chapter 8. JOiNTLESS BRiDGES tABLE 8.5. Procedure A: temPerAture chAngeS Climate Steel or Aluminum Concrete Wood Moderate (°F) 0 to 120 10 to 80 10 to 75 Cold (°F) –30 to 120 0 to 80 0 to 75 Source: LRFD specifications Table 3.12.2.1-1. Source: LRFD Specifications Figure 3.12.2.2-1. Figure 8.8. Maximum design temperature for concrete girder bridges. Figure 8.8. Maximum design temperature for concrete girder bridges. Source: LRFD specifications Figure 3.12.2.2-1. Procedure B considers the range of temperature change, which is the difference between maximum design temperature and minimum design temperature. The maxi- mum design temperature for concrete girder bridges with concrete deck is provided by Figure 8.8, and the minimum design temperature is given in Figure 8.9. The maximum and minimum design temperatures for steel girder bridges are given in Figure 8.10 and Figure 8.11, respectively. 8.6.2.1.5b Temperature Gradient The effect of temperature gradient may typically be ignored; however, if the designer decides to consider the effect of temperature gradient, the following provisions, taken from LRFD specifications Article 3.12.3, are recommended. The profile of the tem- perature in steel and concrete girder bridges may be taken as shown in Figure 8.12, in which t is the thickness of the concrete deck. Dimension A in Figure 8.12 should be taken as follows: • 12.0 in. for concrete superstructures deeper than 16 in.; • Depth of superstructure minus 4.0 in. for concrete superstructures shallower than 16 in.; and • 12.0 in. for steel superstructures.

364 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE The values for T1 and T2 are given in Table 8.6 and vary by solar radiation zone as determined from the map shown in Figure 8.13. The values in Table 8.6 are positive temperature values. The negative temperature values are obtained by multiplying the values from the same table by –0.3 for plain concrete decks and by –0.2 for decks with asphalt overlay. The value of T3 should be taken as 0°F, unless a specific field study is carried out to determine this value, in which case T3 should not exceed 5°F. Source: LRFD Specifications Figure 3.12.2.2-3. Figure 8.10. Maximum design temperature for steel girder bridges. Figure 8.10. Maximum design temperature for steel girder bridges. Source: LRFD specifications Figure 3.12.2.2-3. Source: LRFD Specifications Figure 3.12.2.2-2. Figure 8.9. Minimum design temperature for concrete girder bridges. Figure 8.9. Minimum design temperature for concrete girder bridges. Source: LRFD specifications Figure 3.12.2.2-2.

365 Chapter 8. JOiNTLESS BRiDGES Source: LRFD Specifications Figure 3.12.2.2-4. Figure 8.11. Minimum design temperature for steel girder bridges. Figure 8.11. Minimum design temperature for steel girder bridges. Source: LRFD specifications Figure 3.12.2.2-4. Figure 8.12. Positive vertical temperature gradient in concrete and steel superstructures. Source: LRFD specifications Figure 3.12.3-2. tABLE 8.6. bASiS For temPerAture grAdientS Zone T1 (°F) T2 (°F) 1 54 14 2 46 12 3 41 11 4 38 9 Source: LRFD specifications Table 3.12.3-1.

366 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE When considering the temperature gradient in the section profile, the analysis should consider axial extension, flexural deformation, and internal stresses (LRFD specifications Article 4.6.6). The response of the structure to temperature gradient can be divided into three parts: axial expansion, flexural deformation, and additional stresses. These components are discussed in this subsection. Axial expansion is due to the uniform portion of the temperature gradient and can be calculated as shown by Equation 8.2 (LRFD specifications Equation C4.6.6-1): T A T dw dz 1 UG c G∫∫= ⋅ (8.2) where TG = temperature gradient (D°F), TUG = temperature averaged across the cross section (°F), Ac = cross-section area transformed for steel beams (in. 2), w = width of element in cross section (in.), and z = vertical distance from center of gravity of cross section (in.). The corresponding uniform axial strain (εu) is then taken as Equation 8.3 (LRFD specifications Equation C4.6.6-2): T Tu UG Uε α ( )= + (8.3) where α is the coefficient of thermal expansion (in./in./°F), and TU is the uniform speci- fied temperature (°F). Figure 8.13. Solar radiation zones for the United States. Source: LRFD specifications Figure 3.12.3-1.

367 Chapter 8. JOiNTLESS BRiDGES The consequence of a temperature gradient is flexural deformation, the develop- ment of curvature (f) over the cross section, which can be calculated using Equa- tion 8.4: I T zdw dz R . 1 c G∫∫ α φ = = (8.4) where Ic is the inertia of the cross section transformed for steel beams (in. 4), and R is the radius of curvature (in.). Any additional stresses because of curvature, created by thermal gradient, shall be calculated as E T T zE G UGσ α α[ ]= − − φ (8.5) where E is the modulus of elasticity (ksi). 8.6.2.1.6 Creep Concrete creep strains should be calculated using LRFD specifications Article 5.4.2.3.2. Time dependence and changes in concrete strength should be taken into account in determining the effect of concrete creep. The creep coefficient (ψ) can be determined using Equations 8.6 through 8.10 (LRFD specifications Equation 5.4.2.3.2-1): t t k k k k t, 1.9i s hc f td i 0.118( )Ψ = − (8.6) in which k V S 1.45 0.13 1.0s = − ≥ (8.7) k H1.56 0.008hc = − (8.8) k f 5 1f ci = + ′ (8.9) k t f t61 4td ci = − ′ +     (8.10) where H = relative humidity (%). In the absence of better information, H may be taken from Figure 8.14; ks = factor for the effect of the volume-to-surface ratio of the component; kf = factor for the effect of concrete strength; khc = humidity factor for creep; ktd = time development factor;

368 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE t = maturity of concrete (day), defined as age of concrete between time of loading for creep calculations, or end of curing for shrinkage calculations, and time being considered for analysis of creep or shrinkage effects; ti = age of concrete at time of load application (day); V/S = volume-to-surface ratio (in.); and fci′ = specified compressive strength of concrete at time of prestressing for pre- tensioned members and at time of initial loading for nonprestressed mem- bers. If concrete age at time of initial loading is unknown at design time, fci′ may be taken as 0.80 fc′ (ksi). 8.6.2.1.7 Shrinkage Concrete shrinkage should be calculated in accordance with the provisions of LRFD specifications Article 5.4.2.3.3, where appropriate. For concrete elements, shrinkage strain (εsh) can be calculated using Equations 8.11 and 8.12 (LRFD specifications Equation 5.4.2.3.3-1): k k k k 0.48 10sh s hs f td 3ε = × − (8.11) in which k H2.00 0.014hs ( )= − (8.12) where khs is the humidity factor for shrinkage. Figure 8.14. Percentage annual average ambient relative humidity. Source: LRFD specifications Figure 5.4.2.3.3-1.

369 Chapter 8. JOiNTLESS BRiDGES This article states that if the concrete is exposed to drying before 5 days of curing have elapsed, the shrinkage as determined in Equation 8.11 should be increased by 20%. 8.6.2.1.8 Settlement Settlement is not a deterrent to the use of jointless bridges if sufficiently accounted for in the design of the affected components. AASHTO provides guidance on estimating settlement for structures in LRFD specifications Article 10.7.2.3. Bridges with simple spans and simple abutment bearings are able to accommodate shifting and the associated rotation of the end spans with flexibility of the bearings. With continuous jointless superstructures and integral abutments, vertical or longitu- dinal movement of the foundation will introduce additional stresses in the superstruc- ture, deck, or both. In addition, with semi-integral abutments, vertical movement of the foundation will introduce additional stresses in the superstructure, deck, or both. Figure 8.15 demonstrates this concept with an exaggerated illustration showing settle- ment (D). In instances in which traditional bearings are used, the superstructure is free to rotate to accommodate the movement. In contrast, when the superstructure is inte- gral with the substructure, the superstructure is not permitted to rotate or shift, and thus forces are introduced from the fixed-end displacement. The best approach to address settlement, in general, is to increase pile length so that settlement is not a design consideration. If increasing the pile length is not an option and design for settlement must be considered, then one of two strategies can be used to reduce or eliminate the effect of settlement: (1) evaluate the anticipated settle- ment and account for the resulting forces in the design, or (2) determine the maximum permissible displacement allowable by design and take measures to ensure that that settlement limit is not exceeded. Figure 8.15. Comparison of settlement effects on the superstructure (not to scale). Δ Δ M M V V

370 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 8.6.2.1.9 Wind Wind load needs to be considered in accordance with LRFD specifications Article 3.8. As with braking and centrifugal forces, longitudinal and transverse forces resulting from wind loads should also take into consideration the considerable stiffness of the integral abutments (see Section 8.6.2.1.2). 8.6.2.1.10 Other Loads All other loads prescribed by the LRFD specifications, such as collision forces and water and ice loads, need to be applied to jointless structures in the same manner as other structures. As with all designs, it is the engineer’s responsibility to determine and apply the necessary load conditions appropriate for the unique situation of each jointless bridge. 8.6.2.2 Load Combinations and Limit States This section summarizes available information in the LRFD specifications and de- veloped by SHRP 2 Project R19A related to load combinations and limit states to be considered for jointless bridges. 8.6.2.2.1 Load Combinations The following loads should be considered for jointless bridges: DC = dead load of structural components and nonstructural attachments, DW = dead load of wearing surfaces and utilities, EH = horizontal earth pressure load, LL = vehicular live load, WS = wind load on structure, WL = wind on live load, TU = uniform temperature, CR = creep, SH = shrinkage, TG = temperature gradient, and SE = settlement. 8.6.2.2.2 Load Factors and Combinations Table 8.7 lists load combinations required in the design of jointless bridges, and Table 8.8 shows load factors for permanent loads. The LRFD specifications state that the load factor for the temperature gradient (γTG) should be considered on a project-specific basis or may be taken as follows: • 0 at the strength limit states; • 1.0 at the service limit states when live load is not considered; and • 0.50 at the service limit state when live load is considered.

371 Chapter 8. JOiNTLESS BRiDGES Because the effects of γTG are typically self-limiting and do not significantly affect strength or ductility at strength limit states for the types of bridge girders typically used in jointless bridges, γTG can commonly be taken as zero for the design of foundations in integral and semi-integral abutments. Similarly, the load factor for settlement (γSE) should be considered on project- specific information or may be taken as 1.0. Load combinations that include settle- ment should also be applied without settlement. 8.6.2.3 Bridge Movement Three methods are provided for calculating bridge maximum end displacements. The first approach is applicable to straight bridges, the second approach addresses trans- verse movement of skewed bridges, and the third approach is a general method to calculate the movement of curved girder bridges. tABLE 8.7. LoAd combinAtionS And LoAd FActorS Load Combination Limit State DC DW EH LL WS WL TU CR SH TG SE Strength I γp 1.75 — — 0.50/1.20 γTG γSE Strength II γp 1.35 — — 0.50/1.20 γTG γSE Strength III γp — 1.40 — 0.50/1.20 γTG γSE Strength IV γp — — — 0.50/1.20 — — Strength V γp 1.35 0.40 1.00 0.50/1.20 γTG γSE Service I 1.00 1.00 0.30 1.00 1.00/1.20 γTG γSE Service II 1.00 1.30 — — 1.00/1.20 — — Service III 1.00 0.80 — — 1.00/1.20 γTG γSE Service IV 1.00 — 0.70 — 1.00/1.20 — 1.00 Note: — = not applicable. Source: LRFD specifications Table 3.4.1-1. tABLE 8.8. LoAd FActorS For PermAnent LoAdS (γp) Type of Load and Foundation Type Load Factor Maximum Minimum DC: Component and attachments DC: Strength IV only 1.25 1.50 0.90 0.90 DW: Wearing surface and utilities 1.50 0.65 EH: Horizontal earth pressure Active At rest 1.50 1.35 0.90 0.90 Source: LRFD specifications Table 3.4.1-2.

372 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 8.6.2.3.1 Displacement of Straight (Nonskew) Bridges Bridges expand and contract because of temperature changes and time-dependent vol- ume changes associated with concrete creep and shrinkage. In jointless bridges, it is important to estimate the maximum expansion and contraction at each end of a bridge to determine the longitudinal displacement expected for the abutment piles. It is also important to predict the movement at each pier and the joint width needed between the approach slab and the pavement. Another important movement is the maximum total thermal movement at each end resulting from the total effective temperature range. The starting point to determine the maximum passive pressure should conser- vatively be at the maximum contraction (Oesterle et al. 2005). The maximum passive pressure is related to the end movement, with reexpansion for the full effective tem- perature range. Calculation of the length change for a prestressed concrete bridge can be accom- plished through use of typical design values for the coefficient of thermal expansion combined with creep and shrinkage strains. However, the overall variability of these factors adds uncertainty to the calculated end movements. Although a coefficient of thermal expansion for concrete is typically assumed to be 5.5 × 10–6 to 6.0 × 10–6/°F, it is known that this value can range from approximately 3.0 × 10–6 to 7.0 × 10–6/°F ( Kosmatka and Panarese 1988). Also, the variability of creep, shrinkage, and modu- lus of elasticity of concrete is known to be significant (Bazant and Panula 1980). In addition, resistance to length change from abutments and piers, combined with the variability of the restraint (primarily caused by the variability of the soil), leads to unequal movement at each end of a bridge (even in theoretically symmetrical bridges) and uncertainty as to the magnitude of the movement at each end. Finally, the effective setting temperature of the bridge and the age of concrete girders at completion of the superstructure are typically unknown, making the relative magnitude of expansion and contraction and the starting point for temperature, creep, and shrinkage calcula- tions uncertain. To investigate the effects of the variability of these parameters and to provide guid- ance in formulating recommendations for design calculations, Monte Carlo studies were carried out to calculate bridge movements in order to generate a large number of computer analyses using the statistical variation of material parameters affecting the movement (Oesterle 2005; Oesterle and Volz 2005). Within each analysis, val- ues for the coefficient of thermal expansion, temperature at construction, creep and shrinkage parameters of concrete, modulus of elasticity of concrete, and soil stiffness were selected based on statistical distributions of the values of these parameters. The variations in calculated bridge end abutment movements were then used to deter- mine a 98% confidence interval for the maximum calculated movements. These maxi- mum values were used to determine magnification factors, referred to as Γ factors, for modification of calculated values to account for uncertainty in the various parameters affecting results.

373 Chapter 8. JOiNTLESS BRiDGES The procedures presented in the following sections outline how to determine the maximum end movements of jointless bridges, including use of these Γ factors. In these calculations, it is assumed that the bridge has unknown construction timing and that no specific data on material properties are available. For prestressed concrete bridges, the following steps should be used to estimate the longitudinal movement: • Determine the average construction temperature using the procedure described in Section 8.6.2.1.5a. • Determine the maximum and minimum effective bridge temperatures based on the recommendations of Procedure B in Section 8.6.2.1.5a. • Assume the parameters for concrete presented in Table 8.9. tABLE 8.9. concrete PArAmeterS Value Type Coefficient of Expansion Modulus of Elasticity Value (English) 6.0 × 10–6/°F 57,000 fc′ (psi) Value (metric) 10.8 × 10–6/°C 4.700 fc′ (MPa) Source: Oesterle et al. 2005. • Determine the point of zero movement of the fixity point of the bridge based on the stiffness of the piers and the abutments. Use Section 8.6.2.1.3 provisions (Clough and Duncan 1991) to estimate the backfill passive pressure and the p-y method to evaluate the nonlinear behavior of the soil surrounding the piles. For symmetric bridges, the middle of the bridge will be the point of fixity. • Use Equations 8.13 through 8.16 to calculate the strain values in the bridge: Tthε α= ∆ (8.13) EA EA 1 sh sh sh sh ,girder ,deck ,girder girder deck ε ε ε ε ( ) ( ) = + − + (8.14) EA EA 1 1 cr cr,girder girder deck ε ε ( ) ( ) = +               (8.15) totalλ ε λ∆ = Γ (8.16)

374 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE where Dl = maximum end movement; eth = thermal strain; esh = shrinkage strain; ecr = creep strain; a = coefficient of thermal expansion; E = modulus of elasticity; A = cross-section area; l = length from the point of fixity to the end of the bridge. Note that for an unsymmetrical bridge two values of l are involved; and Γ = magnification factor to account for uncertainty listed in Table 8.10, where etotal = eth – esh – ecr for expansion, and etotal = –eth – esh – ecr for contraction. • For maximum expansion, which occurs shortly after construction, use the tem- perature difference between the maximum effective bridge temperature and the mean construction temperature for the bridge location based on the Federal Con- struction Council’s Technical Report No. 65 (FCC 1979). For creep and shrinkage calculations, assume the girders are 90 days old. Based on Monte Carlo simula- tion, Γ should be 1.6 to account for uncertainties with 98% confidence that the movement will be less than the calculated value. • For maximum contraction, which occurs after several years of service, use the temperature difference between the minimum effective bridge temperature and the mean construction temperature. For creep and shrinkage, assume ultimate values with the girder to be 10 days old at the time of casting the deck. Based on Monte Carlo simulation, Γ should be 1.35 to account for uncertainties with 98% confi- dence that the movement will be less than the calculated value. • For maximum thermal reexpansion from a starting point of full contraction, use the full effective bridge temperature range without any creep and shrinkage movements. Based on Monte Carlo simulation, Γ should be 1.2 to account for uncertainties. • The Γ values in the first two columns of Table 8.10 for maximum expansion and maximum contraction are relatively large and possibly overconservative because they are affected by the relatively large uncertainty of the construction or setting temperature. Further studies to include a more deterministic method to incorpo- rate the construction temperature for a given bridge may reduce these magnifica- tion factors for a more efficient design approach. For reinforced concrete bridges, the same procedure as that used for prestressed concrete bridges should be used to calculate bridge end movements. In the case of shortening, movement caused by creep is not a factor. Magnification factors for differ- ent cases are listed in Table 8.10.

375 Chapter 8. JOiNTLESS BRiDGES For steel girder bridges, the same procedure as that used for prestressed concrete bridges should be used to calculate bridge end movements, except that the extreme effective bridge temperatures should be calculated using the recommendations of Sec- tion 8.6.2.1.5. Other steel material parameters are provided in Table 8.11. The effective coefficient of thermal expansion for steel composite bridges can be estimated as shown by Equation 8.17 (Emanuel and Hulsey 1977): EA EA EA EAe girder deck girder deck α α α( ) ( ) ( ) ( ) = + + (8.17) Magnification factors for different cases are listed in Table 8.10. 8.6.2.3.2 Displacement of Skewed Bridges For information on displacement of skewed bridges, refer to Appendix B. 8.6.2.3.3 Displacement of Curved Bridges A procedure has been developed (Doust 2011) to determine the magnitude and direc- tion of bridge end displacement in the case of curved integral abutment bridges. The related material is provided in Appendix F, which explains the assumptions and limita- tions of the approach. tABLE 8.10. SummAry oF recommended mAgniFicAtion FActorS Bridge Type Maximum Expansion Maximum Contraction Maximum Thermal Reexpansion Prestressed concrete bridges Γ = 1.6 creep + shrinkage + thermal Γ = 1.35 creep + shrinkage + thermal Γ = 1.2 thermal Reinforced concrete bridges Γ = 1.6 shrinkage + thermal Γ = 1.4 shrinkage + thermal Γ = 1.2 thermal Composite steel bridges Γ = 1.7 shrinkage + thermal Γ = 1.5 shrinkage + thermal Γ = 1.2 thermal Source: Oesterle 2005. tABLE 8.11. recommended SteeL PArAmeterS Value Type Coefficient of Expansion Modulus of Elasticity Value (English) 6.5 × 10–6/°F 2.9 × 107 (psi) Value (metric) 11.7 × 10–6/°C 2.0 × 105 (MPa) Source: Oesterle 2005.

376 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 8.6.2.4 Design of Pile Foundation The main steps in design of piles are as follows: • Based on subsurface explorations, develop a soil profile for the site. Details of strength profiles, compressibility characteristics, stress history, and geology of the subsurface materials should be included. Further, identify favorable and unfavor- able strata in the affected subsurface zones. • Estimate the loads for the strength and serviceability limit states. • Determine the water profiles for the site and the expected depth of scour during 100-year and 500-year flood events. • Select technically feasible pile types and pile lengths based on constructability, and consider the strength, serviceability and extreme event limit states. Eliminate the unsatisfactory alternatives. • Make a general comparison between the technically feasible piles, and then design with the most cost-effective alternative based on the following steps: – Estimate the axial and lateral pile nominal resistance considering soil and struc- tural capacity. – Determine the required number of piles and their spacing. – Estimate the resistance of the pile group on the basis of pile group interaction. If the group resistance is not sufficient, modify the number of piles and/or the pile spacing. – Check the possibility of punching of the pile into any weak stratum that may be present beneath the bearing stratum. – Determine the tolerable deformations of the structure and estimate its verti- cal and lateral deformations. If the deformations are greater than the tolerable magnitudes, increase the length of the piles or number of the pile spacing. – If the pile group is subject to uplift, check its uplift lateral. – Determine the loads on top of pile under design lateral displacements to deter- mine design forces for interaction with the pile cap. – Determine whether pile load tests are needed to verify the design and apply the appropriate resistance factors. These requirements are summarized in Table 8.12 and categorized as to whether the requirement applies to the strength or serviceability limit state. In certain cases the extreme-event limit state governs the design of piles.

377 Chapter 8. JOiNTLESS BRiDGES 8.6.2.4.1 Pile Orientation: Straight and Curved Bridges Abutment piles of straight bridges should be oriented so that the strong axis of the piles is perpendicular to the longitudinal direction of the bridge (Doust 2011). This orientation results in strong-axis bending of the piles in response to longitudinal move- ment of straight nonskew bridges. A procedure has been developed to determine the optimum abutment pile orienta- tion in the case of curved girder integral bridges (Doust 2011). Appendix F provides the suggested approach and current limitations. 8.6.2.4.2 Pile Design Design of piles should consider strength; ductility; fatigue; stability; pile group interac- tion; and minimum penetration length required to satisfy the requirements for uplift, scour, downdrag, liquefaction, lateral loads, seismic forces, and other extreme-event loadings. Figures 8.16 and 8.17 provide the design aids for design of piles for integral abut- ment systems. These design aids are based on research conducted within SHRP 2 Project R19A (Azizinamini et al., submitted for publication; Sherafati 2011). A sum- mary of the steps in developing these design aids, which include consider strength, fatigue, and local and global stability, is provided in Appendix C. Figures 8.16 and 8.17 provide four charts that allow the determination of maximum lateral movement capacity of a single pile versus the applied axial load to the pile. The design aids are for tABLE 8.12. SummAry oF Strength, ServiceAbiLity, And extreme-event Limit StAteS thAt muSt be conSidered in deSign oF PiLe FoundAtionS Design Consideration Strength Limit State Serviceability Limit State Extreme-Event Limit State Structural capacity of single pile X — X Bearing capacity of single pile X — X Bearing capacity of pile groups X — X Punching into lower weak stratum X — X Settlement of pile groups — X X Tensile capacity of piles during uplift X — X Uplift capacity of single piles X — X Structural capacity of piles under lateral loading X — X Lateral movement of pile groups when subjected to lateral loads — X X Note: X = applicable; — = not applicable. Source: Adapted from Barker et al. 1991.

378 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE two HP piles (HP 10x57 and HP 12x84) and four soil conditions. The yield strength of the HP piles is assumed to be 50 ksi. The design aids provided in Figures 8.16 and 8.17 make design of piles for integral abutment system an easy process. Knowing the applied axial load to the pile allows determination of maximum lateral movement that the pile can accommodate. Figure 8.16. Maximum displacement of compact HP sections in soft clay (cu = 2.9 psi) for (a) HP 10x57 and (b) HP 12x84. Figure 8.17. Maximum displacement of compact HP sections in medium clay (cu = 5.8 psi) for (a) HP 10x57 and (b) HP 12x84. (a) (a) (b) (b)

379 Chapter 8. JOiNTLESS BRiDGES 8.6.2.4.2a Geotechnical Axial Resistance The axial nominal resistance of a pile is the sum of its tip and friction resistance minus the weight of the pile, as shown by Equation 8.18: Q Q Q Ws tnom = + − (8.18) where Qnom = nominal bearing capacity of a pile, Qs = pile shaft resistance (Asqs), Qt = pile tip resistance (Atqt), W = weight of the pile, As = surface area of the pile shaft, qs = unit skin resistance of the pile, At = area of the pile tip, and qt = unit tip resistance of the pile. In most situations (except for large concrete piles in pile bent piers), the weight of the pile is small compared with the other terms and is usually disregarded. 8.6.2.4.2b Global Stability Global stability is referred to as buckling of the pile between end supports as opposed to local flange or web buckling. In general, global stability is not a governing design provision unless a significant length of pile is above ground level and is unsupported against lateral buckling (Sherafati et al. 2012). 8.6.2.4.2c Lateral Deformation of Pile Groups Provisions of LRFD specifications Article 10.7.2.4 should be used when the p-y method of analysis is used to evaluate pile group horizontal movement. 8.6.2.4.2d Minimum Penetration Length LRFD specifications Article 10.7.1.5 specifies the provisions for the minimum penetra- tion length necessary to satisfy the requirements for uplift, scour, settlement, down- drag, liquefaction, lateral loads, seismic response, and other extreme-event loading conditions. This guidance is also appropriate for the design of jointless bridges and should be followed by the designer. Tensile loading of a foundation (uplift) may be caused by swelling soils, frost heave, buoyancy, lateral loads, and tensile loading during construction activities. Piles subjected to uplift should be designed to withstand tensile stresses and pullout from the subsurface materials.

380 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Tensile loading for a foundation design is well covered in the LRFD specifica- tions, which provide guidance to design against uplift for both single piles and pile groups. The design of single piles, drilled shafts, and micropiles in groups is addressed in Articles 10.7.3, 10.8.3, and 10.9, respectively. Scour around the foundation is an important issue that should be considered in the design. In geotechnical analysis, it should be assumed that the subsurface materials above the scour line do not exist to provide bearing or lateral support. Three scour types should be considered in design (Barker et al. 1991): • Aggradation and degradation are long-term effects. Aggradation is defined as the deposit of stream bed material eroded from other portions of a stream. Degrada- tion is the removal of stream bed material and thus lowering of the bed elevation. • General scour and contraction scour are distinguished by removal of bed material across the entire width of the stream as a result of increasing flow velocities. • Local scour occurs when bed material is removed from a small portion of the width of the stream. Bridge piers and abutments induce acceleration of the flow because of obstruction of the flow and cause vortices that wash away the bed material. Scour is usually evaluated for a design flood with a return period of 100 years, with a check flood not to exceed the 500-year event, or from an overtopping flood of lesser recurrence (AASHTO 2012). To increase the safety against pile failure caused by scour, a few longer piles should be used rather than many short piles. Settlement is not a deterrent to the use of jointless bridges if it is accounted for in the design of the affected components (see Section 8.6.2.1.8). Minimum penetration lengths with respect to settlement calculations for the foundation are not an additional concern for jointless bridges. 8.6.2.4.3 Analysis Tools This section provides a general discussion of different analysis approaches and avail- able tools designers can use to analyze jointless systems. 8.6.2.4.3a Simplified Analysis (p-y Method) The ability to estimate the response of laterally loaded piles is of great importance in the design of jointless bridges. This design consideration is similar to a beam-on-elastic foundation model. If the piles are deep enough, modeling the soil with Winkler springs is a useful method. In this method the soil is considered as a series of independent layers providing resistance (p) to the pile deflection (y). This resistance (p) may be a highly nonlinear function of the deflection (y). The proper form of a p-y relation is influenced by many factors, including • Variation of soil properties with depth; • Shape of the pile deflection;

381 Chapter 8. JOiNTLESS BRiDGES • The state of stress and strain throughout the affected soil zone; and • The rate sequence and history of load cycles. 8.6.2.4.3b Finite Element Analysis Finite element modeling can be used to analyze a jointless bridge. There can be several levels of finite element analysis for such a structure ranging from a simplified analysis to a refined analysis. In a simplified finite element analysis, different elements including composite girders, abutment walls, piers, and piles are modeled using frame elements. The model- ing can be 2D or 3D; however, a 3D analysis is preferred. The soil–structure interac- tion should be modeled by means of springs. Each spring’s load-deflection curve can be assumed to be linear for a simplified model. In the case of 2D models, the girder distribution factors should be calculated using appropriate equations from the LRFD specifications. In contrast, a refined finite element analysis is a 3D modeling of a jointless bridge. In this approach, shell elements can be used to model the bridge elements. The soil– structure interaction can be modeled using nonlinear springs, which can model the abutment–soil interaction of the gap created between the abutment wall and soil as a result of contraction. Based on the importance and complexity of the bridge, the level of detail included in the finite element model can vary. Engineering judgment should be exersized in developing the 3D finite element model. 8.6.2.5 Design of Other Foundation Types It is recognized that other foundation types may be appropriate for jointless applica- tions depending on the requirements for the individual bridge. Additional consider- ations for other foundations are discussed here. 8.6.2.5.1 Drilled Shafts Drilled shafts should be designed considering the same design requirements as piles. Note, however, that traditional drilled shaft diameters of 30 in. and larger may prove to be too stiff for longer bridge lengths. Semi-integral abutments may be designed using drilled shafts with no additional consideration. Refer to LRFD specifications Article 10.8 for more information on the design of drilled shafts. 8.6.2.5.2 Spread Footings Use of spread footings directly over rock with integral abutments is not common prac- tice and is not recommended.

382 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 8.6.2.5.3 Micropiles Micropiles may be a viable option for jointless bridges. Note that micropiles, similar to regular piling, should only be used in a single row for integral abutments. In addition, the micropile design must include consideration of the cyclic nature of the bending load resulting from the integral abutment configuration. Multiple rows of micropiles should only be used for semi-integral abutments. 8.6.2.6 Design of Pile Cap Depending on the selected jointless system, the pile caps of jointless bridges may re- quire special consideration. The pile cap of an integral abutment no longer serves solely as a transfer for gravity loads. The pile cap must transfer longitudinal move- ments and other forces introduced by making the abutment integral. 8.6.2.6.1 Integral Pile Cap Design Pile head elevation design for integral abutments can take one of two forms. The first option is to fix the pile head against rotation. Alternatively, as recently demonstrated by SHRP 2 Project R19A (Azizinamini et al., submitted for publication; Sherafati and Azizinamini 2014), the pile head can be fitted with an elastomer-based collar that al- lows for limited end rotation and displacements to occur, which alleviates some of the stresses induced by bending of the piles. 8.6.2.6.1a Encased Piles (Fixed-Head Condition) The pile cap for integral abutments must take into account and be able to develop the moment resulting from the restraint of the embedded pile head. Figure 8.18 illustrates how the shear restraint develops as a moment over the pile length due to fixity (assum- ing no soil support) as the force couple develops. In turn, this force is resisted by the pile cap, as shown in Figure 8.19. Note that the shear (V) and bending resultant (Cm) will be additive. Wasserman and Walker (1996) indicate that the depth of the resulting stress block is as shown by Equation 8.19: a l 0.85 2p pe=     (8.19) where ap is the depth of the stress block, and lpe is the pile embedment length within the cap. It can be seen intuitively that increasing the embedment length will directly decrease the bending resultant stresses on the cap, both by increasing the moment arm and the length of ap. Taking Mp as the plastic moment of the pile, the force couple balance is repre- sented by Equation 8.20: M C Dp m= ′ (8.20)

383 Chapter 8. JOiNTLESS BRiDGES or M a bf l ap p cb pe p2 ( )= − (8.21) where b is either the pile section depth (weak axis bending) or flange width (strong axis bending), respective to pile orientation, or pipe diameter; and fcb2 is the bending resultant stress. Cm and D′ are given in Figure 8.19. Figure 8.18. Moment transfer from pile to cap. Figure 8.19. Moment transfer from pile to cap: internal force balance. (Not to scale.)

384 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE It follows then that the maximum stress on the concrete cap is the combined resul- tant of the bending and shear stresses (fcb1), as shown in Equation 8.22: = +f f V a bcb cb p1 2 (8.22) 8.6.2.6.1b Pin Head Piles (Flexible Condition) By providing rotational capacity at the pile head, the stiffness of the piling system is reduced, and the moments developed in the pile as a result of lateral movement are decreased, because the pile will deform in a single curvature rather than double cur- vature shape. Because the major criterion limiting the application of jointless bridges is the capacity of the piles to accommodate lateral movement, the proposed detail can allow the application of jointless construction to longer bridge lengths (Azizinamini submitted for publication). The proposed pile cap detail consists of an elastomeric casing at the pile head. To alle viate the stress concentration at the top of the pile caused by rotation, steel plates that slide by each other are key to the design detail. One of these plates is welded to the end of the pile, and the other is embedded in the concrete with shear studs. Figure 8.20 shows the pin head detail for the case of a concrete-filled tube (CFT) pile. However, the suggested detail can be adopted for H-piles, as well. This system offers several advantages; it • Can provide longer service life (by allowing integral construction for longer bridges); • Can effectively be used for jointless skewed or curved bridges; • Allows construction of longer jointless bridges; • Develops smaller forces in the abutment and superstructure, since the lateral stiff- ness of the pile is reduced; and • Reduces construction cost and time. Design considerations include the material around the pile head, which is intended to have a very low elastic modulus to provide rotational capacity. Since the material for the detail experiences large strains, it must be able to accommodate these strains when subjected to the applied cyclic rotations. Elastomeric material, regularly used as bearings for girder bridges, is recommended for the detail (Sherafati and Azizinamini 2014). Sufficient thickness needs to be provided to ensure the efficiency of the detail. Preliminary results indicate that the minimum thickness of elastomer should be 4 in. 8.6.2.6.2 Semi-Integral Pile Cap Design The pile cap in semi-integral bridges is mainly subjected to axial load and possibly the moment created by axial load eccentricity applied to the pile cap. Figure 8.21 shows a prefabricated pile cap.

385 Chapter 8. JOiNTLESS BRiDGES Figure 8.20. Proposed pin head detail for a concrete-filled tube (CFT). )b( )a( Figure 8.21. Prefabricated pile cap: (a) lowering cap into place; (b) final position.

386 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 8.6.2.6.3 Seamless Details The design of the pile cap for a seamless bridge should follow the same procedure as for integral abutment bridges. 8.6.2.7 Design of End Diaphragm (Backwall) In addition to being integral with the superstructure, the end diaphragm acts as a backwall for integral and semi-integral jointless bridge systems. The end diaphragm is designed to resist forces resulting from soil loads; in the remainder of this discussion it is referred to as the backwall. The soil loads include the passive pressure force that is created by superstructure thermal expansion. The calculation of this passive pressure (Pp) is shown in Sections 8.6.2.1.3 and 8.6.2.1.4. Modeling, as shown in Figure 8.22, can be used to design the backwall. The following subsections discuss additional backwall design considerations that vary for each jointless bridge type. 8.6.2.7.1 Integral The backwall for an integral bridge abutment must be designed to adequately transfer forces across the construction joint and into the foundation cap for each direction in which the pile bends. This transfer of forces is illustrated through an example of a strut-and-tie model in Figure 8.23. In this figure, Section AA shows the local section recommended for a local region (dp + b) over which the forces can be transferred and a suggested reinforcing pattern. The length dp is the distance from the forward face of the pile cap to the face of the pile. Figure 8.22. Lateral pressure restraint by superstructure. Source: Oesterle et al. 2005. S Pp Beam (typ.)

387 Chapter 8. JOiNTLESS BRiDGES Figure 8.23. Lateral pressure restraint by superstructure. Source: Oesterle et al. 2005. 8.6.2.7.2 Semi-Integral Semi-integral backwalls do not require additional considerations above those outlined in Section 8.6.2.7.1. The one item of note, however, is that if removable forms are not used to form the bottom of the backwall over the foundation cap, the joint-fill material used should be sufficiently stiff to support the concrete weight, yet flexible enough not to interfere with the movement permitted by the bearings. This has been successfully accomplished with expanded polystyrene filler. 8.6.2.7.3 Seamless The design of backwalls for seamless bridges should be the same as for integral bridges. 8.6.2.8 Design of Approach Slab Jointless bridges require approach slabs for two main reasons: (1) the slab needs to be positively attached to the deck or substructure, or both, to eliminate the joint over the abutment; and (2) the slab must span the area behind the abutment where the potential

388 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE for backfill settlement exists. Backfill settlement will occur and introduce voids regard- less of the degree of compaction and must be considered in design (Schaefer and Koch 1992). 8.6.2.8.1 Integral and Semi-Integral For both integral and semi-integral abutments, the length of the approach slab is de- termined by the extent of the backfill. Gangarao and Thippeswamy (1996) determined that the rate of backfill settlement decreased significantly beyond 20 ft from the back face of the backwall. This is a typical standard approach slab dimension shown in several state standards. The study by Schaefer and Koch (1992) demonstrated that backfill movements occur within a 1.5 horizontal–to–1.0 vertical line from the bot- tom of the abutment for integral abutments. A general recommendation for the design length of the approach slab is to conservatively set at a 2.0 horizontal–to–1.0 vertical slope from the bottom of the abutment. A 20-ft minimum should be considered for both integral and semi-integral abutments, as shown in Figure 8.24. In addition, experience from several states has found that the approach slab should be positively attached to the backwall by at least No. 8 reinforcing bars anchored with a hook, as shown in Figure 8.24. The condition shown in the figure allows for a sepa- rate pour of the approach slab designed as a simple span. Creating a moment connec- tion between the approach slab and the deck slab is not recommended. The connection should be detailed to act as a pin with tension steel transferred across the approach span into the backwall for integral and semi-integral abutments. If a moment connec- tion is desired, it is recommended to use a seamless deck transition for the design (see Section 8.6.2.8.2). A final consideration for the approach slab is the development of compression forces. Sufficient allowance for expansion of the superstructure must be accommo- dated in the sleeper slab. (See Section 8.7.3 for sleeper slab details.) Otherwise, com- pression can be introduced into the slab resulting from closing the expansion gap and then activating the passive pressure behind the sleeper slab or contact with the adja- cent roadway pavement, which is a major issue for spalling and buckling of adjacent pavement. 8.6.2.8.2 Seamless Deck Transition Zone Details of seamless systems developed by SHRP 2 Project R19A are provided in Ap- pendix E. The system is shown in Figure 8.25 and Figure 8.26. Figure 8.25 shows that beyond the abutment a “Transition Zone” is required that replaces the approach slab. The proposed transition introduces simplicity and ease of construction (Jung et al. 2007). The concept slowly transitions from a heavily rein- forced region to a plain jointed condition over an extended transition length. Within the heavily reinforced region, crack spacing is quite small. As the level of reinforce- ment is reduced, the crack spacing increases. These cracks may be allowed to occur naturally or may be forced by shallow saw cuts in the pavement.

389 Chapter 8. JOiNTLESS BRiDGES Figure 8.24. Determination of approach slab length. Figure 8.25. Seamless paving over bridge transitioning to jointed pavement. Immediately adjacent to the bridge is a thickened and reinforced approach zone. The approach zone behaves similarly to a reinforced concrete slab bridge and is intended to carry flexural forces that may arise as a result of settlement. The design of the approach zone is similar to the design of an approach slab for an integral or semi-integral bridge. The transition zone is not designed, per se, but the reinforcing spacing is reduced in specified stages. This transition zone reinforcement is shown in Figure 8.26.

390 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 8.6.2.9 Design of Superstructure–Pier Connection By definition, bridge decks in jointless bridges are continuous, including the region over the piers. The connection between the piers and the bridge deck could be integral, pinned, or expansion types, or they could be connected with a link slab. Figure 8.27 shows these different configurations conceptually. In integral-type connections (Figure 8.27a), the pier and superstructure are mono- lithic with frame action developed between the superstructure and substructure. The advantage of this type of connection is the elimination of bearings. Further, the system provides higher levels of redundancy, especially in highly seismic areas. The longitudi- nal movement of the bridge superstructure is not affected by making the piers integral with superstructure. However, the longitudinal expansion of the deck must be consid- ered in the design of the pier columns, pier foundations, and their connection to the superstructure. In pinned connections (Figure 8.27b), bearings are used to restrict longitudinal movement. Rotation at the bearing is allowed. Although designated as a pin-type connection, typical bridge terminology in which a bearing is not permitted to move Figure 8.26. Continuously reinforced transition zone to jointed pavement. Source: Jung et al. 2007. Plan View 12' 12' 15' Reinforcing Steel Profile View Optional Dowel Dowel CRCP Longitudinal Steel CRC PavementSaw Cuts or Induced Design Crack 100% Steel Zone 60% Steel Zone 30% Steel Zone Transition Transition JC Pavement

391 Chapter 8. JOiNTLESS BRiDGES Figure 8.27. (a) Integral, (b) pinned, and (c) and (d) expansion type bearings for jointless bridges. longitudinally is designated as a fixed bearing, commonly denoted as “F” (Fixed) in traditional design plans. For this connection, longitudinal movement between super- structure and pier is not permitted. Similar to integral-type connections, the longitudi- nal expansion of the deck must be considered in the design of the pier. In expansion connections (Figure 8.27c), bearings are necessary and are required to accommodate both rotation and longitudinal movements. This detail uses tradi- tional expansion bearings as determined by design requirements. For the first three connection types shown in Figure 8.27 (integral, pinned, and expansion), the superstructure is made continuous over the pier. This structural con- tinuity can be accomplished in one of two ways: (1) the superstructure splices can be positioned such that they are made at or near the dead load inflection points for the continuous bridge, or (2) a continuity splice can be used over the pier. This second option is commonly referred to as simple for dead load, continuous for live load. This construction method is shown conceptually in Figure 8.28 for an integral pier. The beams are placed as simply supported over the pier; the beams are either spliced mechanically or additional reinforcing is provided for the diaphragm; and finally, a closure pour is made. The last construction option is the expansion condition using a link slab (Fig- ure 8.27d). A linkage slab is used when the beams are not positively connected, as is the case with the other details shown in Figure 8.27. For this condition, the super- structure is designed and constructed with traditional bearing considerations. The design considerations for each of these pier cap connections are provided in the following sections. 8.6.2.9.1 Integral Pier Cap When making the superstructure truly integral with the pier cap, it must be recognized that both positive and negative moments will be introduced to the cap from live loads and other transient loads. Sufficient strength needs to be provided through the deck,

392 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Figure 8.28. (a) and (b) Simple for dead and continuous for live pier detail after the placing of girders and (c) after the closure pour. Source: Azizinamini et al. 2008. integral diaphragm, and pier cap. Although this type of connection eliminates the need for bearings at the piers and can increase clearance, it introduces more complex forces in the superstructure and the piers (see Figure 8.29). The longitudinal deflection movement of the foundation and the pier accommo- dates the total longitudinal movement expected at the top of the integral piers. (See Section 8.6.2.10 for additional information.) When designing the integral cap, the connection between the beams, continuity diaphragm, pier cap, and pier column must be sufficient to transfer the moments resulting from this deflection. Resolution of these forces should be computed by an analytical method or structural model with the abil- ity to properly capture the behavior of the whole bridge system. Figure 8.29. Integral cap as completed. (a) (b) (c)

393 Chapter 8. JOiNTLESS BRiDGES 8.6.2.9.2 Fixed (Pinned) and Expansion Pier Caps Similar to the integral pier cap, for both fixed (pinned) and expansion pier caps the superstructure is made continuous over the pier. The difference is that it is not made integral with the pier caps. The connection between the superstructure and the pier is treated with traditional bearings. Various details are used over the interior supports of multispan bridges to elimi- nate joints. One of the concepts implemented by various owner agencies for concrete bridges has been the use of simple spans for dead load made continuous for live load (Freyermuth 1969; Oesterle et al. 1989). The girders are simply supported for dead load, but continuity is achieved with deck steel as negative moment reinforcement over the piers. In addition, the girders are made integral with the interior pier diaphragms, and commonly positive moment reinforcement is included, as shown in Figure 8.30. Badie et al. (2001) discussed the alternative use of an interior steel pier diaphragm with prestressed girders to speed construction and achieve better overall design econ- omy. The concept of a simple span made continuous has also been applied to eliminate interior joints and improve the construction speed and design economy for short- and medium-span steel girder bridges (Azizinamini et al. 2008). As discussed in Section 8.6.2.4, attention must be paid to the effects of provid- ing positive moment restraint at the diaphragms. Some simple-made continuous pre- stressed concrete girder bridges have experienced severe cracking in the girders near the interior diaphragms. One extensively studied example is the Francis Case Memo- rial Bridge spanning the Washington Channel of the Potomac River in the District of Columbia (Telang and Mehrabi 2003). The prime cause of this distress was the restraint of upward camber of the prestressed girders under the influence of prestress- ing and temperature gradient. According to Telang and Mehrabi (2003), “By provid- ing a large amount of positive moment reinforcement at the diaphragms, designers inadvertently make the diaphragm area stronger than the adjacent girder sections, thereby forcing the cracking to occur in far more critical but weaker areas of the girder span.” Their article continues, “In closing, it is important to note that this seemingly simple transformation of simple-span prestressed girders to continuous spans should be attempted with caution, and significant attention must be paid during analysis and design to include loading conditions that can cause counterintuitive behavior such as Figure 8.30. Precast, prestressed girders connected with live load continuity. Deck Reinforcement Positive Moment Reinforcement

394 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE secondary positive moments at the piers. Most importantly, positive moment rein- forcement should be designed and detailed such that any cracking, if it occurs, should be limited to the relatively less critical diaphragm region of this type of structural sys- tem.” Further discussion of this problem and solutions to avoid it have been published by Oesterle et al. (2004a) and Arockiasamy and Sivakumar (2005) and are discussed in Section 8.6.2.4. 8.6.2.9.3 Link Slab Expansion Pier Cap A link slab is a type of detail used in conjunction with existing or new bridges hav- ing girders that act as simple beams for both dead and live loads. In this type of deck detail, the slab spans continuously over the longitudinal gap between the adjacent span girders, and the girders are kept as simple spans (see Figure 8.31). The length of the deck connecting the two adjacent simple-span girders is called a link slab (Caner and Zia 1998). Link slabs generally require less deck reinforcement, but they have more girder positive moment demands than simple-made-continuous designs. Limited analysis and laboratory experiments were carried out and design recommendations are provided in Caner and Zia (1998). The use of this detail has been very limited due to field-observed cracking. In fact, link slabs are not common in Snow Belt states. A crack is invariably formed due to deck slab rotation as the bridge is loaded with live loads. SHRP 2 Project R19A research studies on the link slab indicated that it offered negligible rotational end restraint to the bridge girders and that the link slab can be analyzed as a beam subjected to the same end rotations as the adjacent girders. The researchers found that under service-load conditions, the link slab would crack pri- marily due to bending. Prior methods developed by Gastal (1986) and El-Safty (1994) were capable of predicting the forces, stresses, and crack widths in the link slab due to thermal and shrinkage effects and creep. Caner (1996) modified the procedures devel- oped by Gastal and El-Safty to properly capture the link slab actions. All these solu- tions were based on beam theory. The reinforcing bar stresses compared reasonably well with the data measured from the experimental tests. The predicted crack widths Figure 8.31. Conceptual detail for link slab.

395 Chapter 8. JOiNTLESS BRiDGES were somewhat larger than the measured crack widths. The researchers concluded that bending and cracking under live load plus impact are the governing factors that must be considered in the design of the link slab. Caner and Zia (1998) suggested design of the link slab using only one layer of rebar placed near the top of the deck, but they suggested that two layers could be used to improve performance in bridges having horizontal restraints. 8.6.2.10 Design of Integral Piers As discussed in the section on design of integral pier caps, the total longitudinal move- ment expected at the top of integral piers is accommodated by two modes of defor- mation: longitudinal movement via rotation of the foundation system and flexural deflection of the pier. Pier deflection can be both elastic and inelastic in response. 8.6.2.10.1 Foundation Rotation For spread footings, Zederbaum (1969) provides an equation, shown here as Equa- tion 8.23, to estimate the rotational stiffness of the soil or rock responding to an applied moment: K E I b 3 s f=θ (8.23) where Kq = rotational stiffness of the foundation, b = one-third of the spread footing width, Es = compression modulus of the soil or rock, and If = the moment of inertia of the footing base. For pile-supported and drilled shaft–supported foundations, the rotational stiff- ness is estimated from the elastic stiffness of the pile or shaft group. Rotation of the foundation can be attributed to the elastic shortening and elongation of the piles or shafts for multiple rows. Note that the elongation and shortening add additional uplift and downward forces, respectively, that must be accounted for in the foundation design. In a single row of piles or drilled shafts, the rotational stiffness is based on the cantilever response of the single row. The length of the cantilever is based on the soil– structure interaction at the foundation and can be based on the assumed or calculated point of fixity for the pile or shaft. 8.6.2.10.2 Pier Displacement The differential between the pier displacement at the integral cap and the rotation of the foundation is the deflection of the pier column. The resulting design moment can be estimated by the following steps. First, the expected movement of the superstruc- ture at the pier cap should be calculated, as outlined in Section 8.6.2.3. Alternatively, this can be sufficiently approximated by determining the point of zero movement on the bridge and multiplying the end displacement by the ratio of the distance from the

396 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE fixed point to the pier to the distance from fixity to the end support. Second, assume that 30% of the expected lateral deflection is accommodated by the foundation rota- tion. Thirty percent is based on a parametric study that demonstrated foundation rotation can vary from 30% to 80% with an average close to 45%. Third, the antici- pated bending moment should be calculated using Equation 8.24: M EI b H 6 e 2= ∆ (8.24) where E = concrete modulus, Ie = effective section modulus, Db = lateral deflection at the pier cap, and H = height of the pier. Note that the value of the effective section modulus may depend on the applied loading and thus a simultaneous or iterative solution may be required. For fixed (pinned) continuous piers, divide the result of Equation 8.24 by 2 (fixed-end moment for a fixed-guided beam is one-half that of a fixed-fixed beam). The value of the effec- tive section modulus for reinforced concrete piers can be obtained from Equation 8.25: =     + −             I M M I M M I1e cr a g cr a cr 3 3 (8.25) where Mcr = cracking moment, Ma = applied moment, Ig = moment of inertia of gross section, and Icr = moment of inertia of cracked section. 8.6.2.11 Design of Wingwalls The design of wingwalls depends on their orientation relative to the abutment stem, their method of support, and the abutment skew. There are various possible configura- tions for wingwalls, but the traditional configurations include U-shaped, straight, or flared, the last being some degree of angle between the first two. Oesterle et al. (2005) indicate that the U-shape configuration is preferable for wingwalls in that this configuration inherently reduces the passive pressure introduced by the longitudinal movement of the abutment end diaphragms. Additionally, they note that the U-shape configuration conveniently contains the soil behind the abut- ment and decreases bulging of the embankment soil. Use of both straight and flared walls leads to the development of passive pressure on the wingwalls as the jointless abutment moves. Oesterle et al. (2005) note that this pres- sure can be expected to decrease as the distance from the abutment increases, but that the degradation cannot be effectively predicted. Thus, the wingwalls need to be designed for the same passive pressure as that of the abutment end diaphragm across the length.

397 Chapter 8. JOiNTLESS BRiDGES For integral and seamless bridges, additional considerations for wingwalls include the loading effect they have on the bridge structure. When cantilevered from the abut- ment stem, the weight of the wingwalls will create additional torsion and/or bending along the length of the abutment. These forces are resisted by a counteracting negative moment at the end of the external beam or girder. If wingwalls for integral abutments are placed on supports, such as piles or a spread foundation, the support must also be able to accommodate the movements of the joint- less bridge. For this condition, Oesterle et al. (2005) note that the shear and moment developed in the wingwall foundation must be transferred through the wingwall struc- ture to the abutment and superstructure. They also note that U-shaped wingwalls on piles create significant resistance to abutment rotation, which creates partial fixity for beam end moments on the exterior beams or girders. These additional moments need to be included in the design of the connections of the exterior beams to the integral abutment. 8.7 detAiLS The introduction of different mechanisms for transferring force to the foundations requires that additional details be considered when designing jointless bridges. The following sections present specific details for each jointless bridge type. The term back- wall is used to describe the end diaphragm that resists soil loads. 8.7.1 general Abutment Details for Jointless Bridges In this section, details that have been used successfully in the past by some states are presented along with general concepts. The figures represent recent research efforts and the accumulated experience of several states that have used jointless bridge technology. 8.7.1.1 Integral Abutments Figure 8.32 shows the overall concept for an integral bridge abutment, including the typical layout with the beam, end diaphragm (backwall), and pile cap, all integral. Although it is not necessary in all cases, the beam shown in this figure is sitting on a temporary pedestal to achieve proper alignment before being cast integral with the rest of the abutment. Alternatively, the cap can be stepped to accommodate elevations before pouring the backwall. For proper alignment and to allow for rotations that occur when placing the beam, a small elastomeric pad should be placed at the girder bearing even though each beam will eventually be cast composite with the abutment. Note that the need to design the pads and for what capacity has not been thoroughly studied. The pads need not be designed to meet the criteria for rotational capacity, which is now addressed as a shear strain component. The maximum rotation of the pad is realized during placement of the beam. A reasonable assumption is to design the pad to accommodate only noncomposite bearing pressure. Drainage is also important to avoid ice expansion and removal of the backfill by washout. A drain pipe should be placed at the appropriate location to properly remove any water that might otherwise accumulate behind the backwall.

398 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Figure 8.32. General integral abutment concept. Additional end diaphragm details are presented in Figure 8.33. Note that an H-pile foundation is shown in the figure; however, each of the foundation types noted in the strategy table in Section 8.5 can be interchanged. The minimum embedment length of 2 ft, within the pile cap, should be maintained for H-piles, prestressed piles, and CFT piles, as shown in Figure 8.34. Also shown in the figure is an approximate cap height of 5 ft, typical of cold-weather regions, which allows for embedment below the frost depth and to provide 2 ft between the finished grade and the bottom of the beam. A depth of 3 to 3.5 ft is more common where frost depth need not be consid- ered. Another alternative to the holes through the beam shown in Figure 8.34 is to use threaded inserts, which are preferred by some precast concrete companies to ease securing them in the forms.

399 Chapter 8. JOiNTLESS BRiDGES Figure 8.33. Integral abutment details. Figure 8.34. Integral abutment rotation detail. Source: Ohio DOT.

400 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE Figure 8.34 is an adaptation from an Ohio DOT standard drawing showing a pre- stressed concrete beam. A standard detail for most prestressed girders includes provid- ing holes through the beam for reinforcing. This reinforcing provides continuity though the backwall for bending and limits the differential deflection between the superstruc- ture and backwall where tension forces develop in the top portion of the web. In contrast to the design recommendations in Section 8.6.2.7, the state of Ohio allows for rotation in the backwall across the construction joint instead of designing rebar to transfer the forces through the stem. The configuration is used to accommo- date the rotation of the superstructure, as shown in Figure 8.34. At the centerline of bearing, reinforcing is crossed at the bearing pivot location and expansion-joint mate- rial is placed to permit a limited amount of rotation. Note that Ohio limits the length of its bridges with integral abutments to 250 ft, so consideration of this limit should be made before adopting this detail for other bridges. Figure 8.35 presents another standard integral detail drawing from the New York State Thruway Authority that shows a steel beam connection. This detail is more typical of DOT design standards in that the reinforcing is continuous across the con- struction joint. When comparing this detail with Figure 8.33, although both details have had repeated success, there are two obvious differences: (1) Figure 8.33 shows Figure 8.35. Integral abutment details. Source: New York DOT.

401 Chapter 8. JOiNTLESS BRiDGES a bent hook bar connecting the approach slab, but Figure 8.35 shows that continuity is maintained by a straight bar connecting the approach slab to the deck; and (2) the Figure 8.33 detail uses a shear key, but the Figure 8.35 detail relies solely on the con- tinuity of the reinforcing across the construction joint. Each detail has demonstrated success in application, and the designer should consider which option may be more appropriate for each bridge’s unique situation. For more information on backwall detailing, see Section 8.7.2. 8.7.1.2 Semi-Integral Abutments Figure 8.36 shows the overall concept for a semi-integral bridge abutment. It includes the typical layout with the beam and end diaphragm (backwall) cast integral. Drainage and porous backfill are necessary for the same reasons as for integral abutments: formation of ice and integrity of the backfill. In semi-integral abutments, two bearing strategies have been used successfully. In the first method, the pile cap Figure 8.36. General semi-integral abutment concept: (a) section (with pedestal) and (b) elevation (with no pedestal). (a) (b) 1” PFJ Expanded Polystyrene Elastomeric Bearing

402 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE may be cast level and the superstructure superelevation can be accommodated through the use of bearing pedestals. The second method is to step the pile cap. In the latter case, the polystyrene filler must be used on both the top of the cap and on the sides of the step to allow for movement. Due to the nature of the superstructure movement, it is recommended that the first case with pedestals be used for locations of high skew (larger than 20°) and bridges on a curve. If it is desired to inspect the bearings during the life of the bridge, removable filler material should be placed in front of the bearings. Figure 8.37 shows the successful detailing strategies that have been used in vari- ous states. The foundation shown is for a drilled shaft, but other foundation types are equally applicable. Similar to integral abutments, dowel holes are placed through the beam or girder. Unlike integral abutments, bearings are used to accommodate move- ment between the superstructure and the foundation. Efforts must be made to seal the gap between the cap and backwall yet still accommodate movement. This seal has traditionally been a preformed filler surrounding the bearing area and a layer of waterproofing applied to the rear face of the seam before placing the backfill. For more information on backwall detailing see Section 8.7.2. Other than the backwall and treatment of the bearing area, detailing for the rest of a semi-integral abutment is the same as for a traditional design. Figure 8.38 shows an alternative detail used in instances in which the diaphragm is extended and a lip is dropped down over the pile cap. This detail replaces the neo- prene sheeting that provided the barrier between the porous backfill and the expanded Figure 8.37. Semi-integral details.

403 Chapter 8. JOiNTLESS BRiDGES polystyrene filler surrounding the bearings. Preformed elastomeric material is placed between the extended diaphragm and the abutment stem. 8.7.1.3 Seamless Abutments Detail recommendations for the transition zone are not well established, and no stan- dard details are available for reference. However, the recommendations for the abut- ment cap and backwall are the same as those presented in Section 8.7.1.1. 8.7.2 Pile Cap and Backwall Oesterle et al. (2005) recommend that vertical reinforcement for the moment from the soil load be distributed with 75% of the bars within 25% of the beam spacing on either side of the beam. Furthermore, for crack control they recommend the center- to-center spacing (s) of the flexural reinforcement not to exceed (in inches) the value calculated by Equation 8.26: ≤ − ≤     s f c s f 540 2.5 or 12 36 s c s (8.26) where cc is clear cover from the nearest surface in tension, and fs is calculated stress (ksi) at service load, or alternately, as 0.60Fy. This limitation is taken from ACI 318-05 Section 10.6.4 rules for the distribution of flexural reinforcing to control cracking in one-way slabs. Further commentary on this requirement can be found in that section. Figure 8.38. Semi-integral details with extended diaphragm.

404 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 8.7.3 Sleeper Slab A sleeper slab is appropriate for all integral or semi-integral bridges and is placed at the roadway end of the approach slab. The intent of this slab is to provide a relatively solid foundation for the far end of the approach slab and to provide a location for limited expansion and contraction (see Figure 8.39). Although no formal design is sug- gested, a typical suggested detail has been provided by Wasserman and Walker (1996). A potential problem with the design shown in Figure 8.39 is that it presents a potential for cracking in the approach pavement where it suddenly transitions to the thin piece above the sleeper slab. Although this type of transition might ease final grad- ing, it is preferable to have the stem of the inverted “T” of the sleeper slab extend to final grade and thus avoid any sharp transitions. The state of New York has adopted this sleeper slab detail and modified it to marry the adjoining pavement design based on the type of surfacing used. Figure 8.40 and Figure 8.41 show the sleeper slab for concrete and asphaltic pavements, respec- tively. In these figures, note how the state formed the joint such that both the pave- ment and approach slab are graded at full depth up to the sleeper slab that provides the transition. The location of the sleeper slab should be placed so that the entirety of the slab is outside the failure plane, as discussed in Section 8.6.2.8.1. Figure 8.39. Suggested helper (sleeper) slab details. Source: Wasserman and Walker 1996.

405 Chapter 8. JOiNTLESS BRiDGES Figure 8.40. Sleeper slab with concrete pavement approach. Source: New York DOT. Figure 8.41. Sleeper slab with asphalt pavement approach. Source: New York DOT.

406 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 8.7.4 Details for Skewed and Curved Bridges Transverse movements of integral abutments associated with large skews or horizontal curves should be accommodated by the details for barrier walls, drainage structures, and the ends of the approach slabs. In addition, the foundation and pier structure stiff- ness will likely be significant for movement parallel to the pier cap. It is recommended that the connection between the bottoms of the girders and diaphragms and the pier caps be flexible in this direction. This approach, however, may not be appropriate for seismic design, in which case the design of the diaphragms should consider the interior pier restraint of the rigid body rotations that result from passive abutment restraint of longitudinal thermal expansion. 8.8 conStruction 8.8.1 Construction Stability In response to concerns about the repetitive bending stresses on the pile, it is recom- mended that no seam (weld) be placed at the top 30 ft of the pile. This placement of the seam will ensure proper ductility and eliminate the possibility of having a poor fatigue detail near the region of higher bending response. In addition, it will better ensure proper alignment of the pile at the cap. The order of construction is also important, as described in Section 8.8.4. 8.8.2 Utilities Nonflexible utilities should not be permitted to pass through integral and semi-integral abutments. Multiple DOTs report experiencing problems in which the flexibility of the integral cap creates issues with rigid utilities. Only utilities that are able to sufficiently flex with the movement of the integral abutment should be permitted, but it is prefer- able to locate all utilities adjacent to the bridge structure. 8.8.3 Cracking Control Vertical cracks have often been found at the bottom of diaphragms between precast beams over the piers in the positive moment connection region near the external (fascia) girders. On the interior girders encased in the diaphragm, spalling of the diaphragm has been observed near the bottom flange. This spalling results from the bottom flange slipping outward (away from the diaphragm) due to the end rotation of the girder asso ciated with creep and thermal changes. These vertical cracks in the diaphragm and end rotation of the girders serve to relieve tensile stresses resulting from creep, shrink- age, and thermal movement and are not detrimental to the integrity of the structure. Attempts to control this cracking through overreinforcing may result in cracks in less desirable locations.

407 Chapter 8. JOiNTLESS BRiDGES Horizontal cracks and efflorescence have been found on the forward face of inte- gral abutments at the construction joint on top of the pile cap. These alterations can be alleviated by placing adequate sealing from water behind the stem across the con- struction joint. Settlement of the approach slab is common and can cause cracking and further damage to the barrier rail. Rails that are attached to both the deck and approach slab should be jointed to accommodate the differential settlement. 8.8.4 Construction Sequencing Guidelines for concrete bridge-deck materials and construction to control transverse cracking in concrete bridge decks are presented in NCHRP Report 380 (Krauss 1996). Among the issues that affect deck cracking are weather, time of placement, curing, vibration, finishing, loads, and placement sequencing. Certain current practices are presented here for jointless bridges. For jointless bridges, the construction sequence should generally be as follows: 1. Embankment should be completed before pile driving and should allow for con- solidation (if required). 2. Piling should be placed, and predrilled holes filled and forms constructed (if used). 3. Abutments and wingwalls should be constructed to the elevation of the bearing seat. 4. Semi-integral elastomeric bearings should be set; or integral beam pads should be set allowing for rotation from beam and deck dead load. 5. Beams should be set. 6. The deck slab and the integral backwall should be cast. The ends of the slab should be poured last to minimize locked-in stresses at the supports. 7. Drainage and backfill should be placed behind the abutments after the deck has achieved the appropriate strength. It is important that the backfill be placed simul- taneously behind each abutment so the bridge is not inadvertently shifted in the unsupported direction. 8. The approach slab should be cast, ideally with the bridge in the thermally con- tracted position (i.e., early morning). This avoids putting the slab into tension until the concrete has gained sufficient strength. It should be emphasized that simultaneous placement of the abutment backfill (Step 7) is particularly important for semi-integral abutments because in semi-integral bridges the superstructure sits on flexible bearings rather than being positively attached to the abutment, and it is more likely to move in response to the pressure from the compact- ing procedures.

408 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 8.8.5 fill Compaction Construction can follow normal compaction procedures as specified by the owner agency except as noted in the section on construction sequencing. Fill compaction has been modified and adjusted using several variables, including the use of specialized material. However, general experience has indicated that properly compacted normal fill material is sufficient for jointless bridge construction, and proper drainage behind the backwall is more important. 8.9 mAintenAnce And rePAir 8.9.1 Problems with Jointless Construction Although integral-type bridges will eliminate some of the more troublesome prob- lems associated with jointed bridges and yield significantly more durable structures, they will not eliminate endemic highway construction problems that are somewhat related to accelerated construction, all-weather construction, marginal construction super vision, and other construction issues identified in Chapter 3 on materials and Chapter 4 on bridge decks. Transverse and diagonal deck slab cracks, stage construction issues, lateral rota- tion of superstructure, erosion of embankments, marginal quality of structure move- ment systems, and other problems have appeared to trouble design, construction, and maintenance engineers. Except for early-age deck slab cracking, these problems are generally the result of failure to anticipate and apply typical design and construction provisions to achieve trouble-free construction and more durable structures. 8.9.1.1 Deck Cracking Diagonal deck slab cracks located at acute corners of integral-type bridges are occa- sionally reported. When constructing integral-type bridges, stationary abutments and moving superstructure must be joined together by cast-in-place continuity connections. Consequently, these connections could be stressed and cracked if a substantial tem- perature drop were to occur during initial concrete setting, or if concrete placement sequences were not suitably controlled. To address this problem, one or more of the fol- lowing procedures should be used: place continuity connections at sunrise, place deck slab and continuity connections at sunrise, place continuity connections after deck slab placement, or use crack sealers. 8.9.1.2 Lateral Rotation of Semi-Integral Bridges One of the primary aspects of semi-integral bridges that must be considered and ad- dressed is the design and proper orientation of guided bearings for the superstructure of skewed bridges. Unfortunately, many of the retention devices currently being used are not fully functional because of friction and binding and consequently, the long-term stability of some abutments, especially those not supported by rigid foundations, may not have been provided for effectively. However, it appears inevitable that this aspect

409 Chapter 8. JOiNTLESS BRiDGES of the semi-integral bridge concept will be improved when bearing manu facturers and bridge design engineers combine their expertise to design and manufacture more func- tional structure movement systems for these applications. 8.9.1.3 Approach Slabs Shortly after the state of Ohio adapted the integral concept to continuous steel beam bridges in the early 1960s, slab distress was experienced. Where the bridges in ques- tion were constructed adjacent to compressible asphalt concrete approach pave- ments, approach slab seats at the ends of bridge superstructures were found to be fractured, approach slabs had settled, and the vertical discontinuity in the roadway surface at the approach slab–superstructure interface was hindering movement of vehicular traffic. 8.9.1.4 Drainage Washout has been noted on several existing structures in which drainage was not properly designed or maintained, including some where the piles became exposed. It is imperative that proper drainage material including geotextiles and perforated piping be placed behind the abutments. The preferred alternative is to direct water away from the bridge approach, but it is acknowledged that this can be difficult to accomplish in many cases. In addition, improper drainage can lead to washout at one end of the bridge and not the other. For semi-integral abutments, this leads to an unbalanced soil pressure, which can lead to additional maintenance issues at the bearing locations. Drainage can also affect settlement of the sleeper slab and create settlement of the approach slab. It is recommended that runoff be intercepted or diverted so that it does not reach the end of the approach slab. In regions that experience freezing temperatures, proper drainage is also important to minimize the potential for frozen soil behind the abutment. The magnitude of the potential restraining force is unknown for frozen soil, but it will be minimized with proper drainage (Briaud et al. 1997). 8.9.1.5 Cycle-Control Joints Probably the most significant unresolved problem regarding integral and semi-integral bridges is the availability of cost-effective functional and durable cycle-control joints, which are the moveable transverse joints used between approach slabs of integral-type bridges and approach pavements. The usual pavement movement joints, composed of preformed fillers, are currently being used for the shortest bridges. For the longest bridges, finger-plate joints with easily maintainable curb inlets and drainage troughs have been successfully employed. However, for intermediate-length bridges, develop- ment of a suitable cycle-control joint is still in the evolutionary stages.

410 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE 8.9.2 Deck Replacement Figure 8.42 shows what can happen when the proper procedures and sequences for deck replacement and integral abutment backfilling are not followed. It should be anticipated that large compressive forces are acting on the whole structure as a result of soil pressure on the abutments and restrained expansion of the girders. To ensure the global stability of the structure, one of two procedures must be followed. The first procedure, which should always be used for whole deck replacement, is to use proper construction sequencing as follows: 1. Remove the approach slab. 2. Remove backfill to the bottom of the stem for integral abutments or to the bottom of the end diaphragm for semi-integral abutments. Excavation should be done simultaneously behind both backwalls. 3. Remove the deck. 4. Replace the deck according to the guidance provided in Section 8.8.4. The second option is to calculate the stress applied by the passive pressure of the abutment backwalls. This force can then be applied to the superstructure with por- tions of the deck removed to check the stability of the system and each structural item that might be affected by the removal of the deck. This procedure includes checking both local and global buckling stability. It is recommended that this be used only for partial-width deck repair. 8.9.3 Bearing Replacement for Semi-integral Jointless Bridges An additional factor when detailing semi-integral jointless bridges, should bearing re- pair or replacement be required, is to incorporate appropriate features at the initial design stage to facilitate superstructure jacking. In general, since superstructure and abutments in semi-integral bridges are separated by elastomeric bearings, it would be easy to place flat jacks between the abutment and the end diaphragm to raise the super structure and approach slab and allow replacement of the bearing. Figure 8.42. Buckled girder flanges resulting from improper deck sequencing.

411 Chapter 8. JOiNTLESS BRiDGES 8.10 retroFitS A large percentage of existing bridges are simple-span bridges that rely on expansion joints at piers and abutments to accommodate longitudinal movements. Most of the deficient bridges in the United States include these jointed structures, which require upgrade and repair. Retrofitting existing jointed bridges to jointless ones is highly recommended. The following considerations are required in integral conversion (Leathers 1990): 1. The existing structural elements should be able to properly function without the expansion joint. 2. Movement calculation should be based on the LRFD specifications. 3. Continuity can be achieved by making either the deck or the girders continuous. 4. All obsolete and deteriorated bearings should be replaced with elastomeric bearing devices. 5. If the abutment is unrestrained, a fixed integral condition can be developed for many of the shorter bridges. Abutments that are free to rotate, such as a stub abut- ment on one row of piles or an abutment hinged at the footing, are considered un- restrained. A semi-integral condition can be developed for restrained abutments. 8.10.1 Details over the Pier Two practical options that can be used with or without integral abutments are avail- able for retrofitting existing jointed bridges into jointless bridges: 1. Provide beam continuity for live load only. In this case, the negative moment con- tinuity is provided over the piers, with or without positive moment continuity at these locations; or 2. Provide deck slab continuity only. In this option, although the deck is continuous, beams are technically, simply supported. This method involves removing some length of slab at the ends of the adjacent beams, splicing the existing reinforcement and adding new bars, then recasting that part of the deck. When retrofit of an existing open joint is considered, the following approach may be used, as shown in Figure 8.43 (note for this detail only the deck is made continuous): 1. Remove concrete as necessary to eliminate existing armoring. 2. Add negative moment steel at the level of existing top-deck steel sufficient to resist transverse cracking. 3. Reconstruct with regular concrete to original grade. Because the deck slab will be exposed to longitudinal flexure as a result of the rota- tion of beam ends responding to the movement of vehicular traffic, cracks will occur over the link slab. However, for short- and medium-span bridges, the deck cracking

412 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE associated with such behavior is preferred over the long-term consequences associated with open movable deck joints or poorly executed joint seals. 8.10.2 Details over the Abutment For existing stub abutments with a single row of piles, the following procedure (shown in Figure 8.44) should be used (integral abutment retrofit): 1. Check the capacity of piles and the pile cap connection for the expected movement. 2. Remove the approach slab, and excavate backfill to the elevation of the existing ground on the front face. 3. Demolish the existing backwall to the top of bridge seat. Cast reinforced concrete around beam ends. 4. Provide drainage, backfill, and new approach slab behind the new abutment. For existing stub abutments with rigid foundation or existing full-height wall abut- ments, the following procedure (shown in Figure 8.45) should be used (semi-integral abutment retrofit): 1. Remove the approach slab and excavate the backfill to the elevation of the existing ground on the front face. 2. Remove the existing abutment backwall to the top of the bridge seat. 3. Provide a sliding surface between the pile cap and the abutment stem, which is cast integrally with the beam ends and approach slab. 4. Provide details for both horizontal and vertical sliding joints by using lateral guide bearings, sheet seals, and drainage and backfill. 8.10.3 Converting Jointed Bridges to Jointless Bridges General experience has shown that most common bridge types can be converted to jointless bridges to enhance their performance with the same goal as new construc- tion (i.e., joint elimination). Examples of candidates that have already been converted are pin-and-hanger bridges and multispan and simple-span bridges for both steel and concrete superstructures. Several states have had success converting old pin-and-hanger expansion joints to a bolted full-moment connection, thus eliminating the expansion joints. Figure 8.43. Integral conversion at piers. Source: Leathers 1990.

413 Chapter 8. JOiNTLESS BRiDGES Figure 8.44. Conversion of a bridge with moveable deck joints at the superstructure– abutment interface with integral abutment (a) before and (b) after conversion. Figure 8.45. Conversion of a very-short-span bridge with moveable deck joints at the superstructure–abutment interface with integral abutment (a) before and (b) after conversion. )b( )a( )b( )a(

414 DESiGN GUiDE FOR BRiDGES FOR SERviCE LiFE The state of New Mexico has presented several case studies (Maberry et al. 2005). In one of them, simple-span concrete girders were converted by incorporating a link slab. The project demonstrated that attention must be paid to the bearings. The greatly increased expansion that would transfer to the outer bearing locations was overlooked by the retrofit assessment. The resulting expansion loads were absorbed by the pile caps, which quickly deteriorated. The key to any conversion is the ability of the bridge to withstand the new conti- nuity loading and expansion demands introduced by the changed load path. Because of the complex nature of the converted structure, it is recommended that conversions be treated with the same level of analysis as required for a new design.