JAMES D. PFLUGER PEDESTRIAN AND BICYCLE BRIDGE INTRODUCTION Now the 18th largest city in the nation, Austin, Texas, is growing at a phenomenal rate, with its population soaring nearly 40% over the past ten years. Austin is also unique among large cities in that many of its citizens are avid walkers, runners, and bicyclists, not only for recreation, but also as a practical alternative for the daily commute. The Town Lake area—formed along the southern edge of downtown Austin by the damming of the Colorado River—is one of the most popular locations for recreational activity and one of the most heavily-traveled commuter corridors. The existing Lamar Boulevard Bridge, an historic six span reinforced concrete deck arch bridge spanning Town Lake, is widely used by both commuters and recreational pedestrians and bicyclists, although the bridge was not originally designed to accommodate such use and is now considered functionally obsolete. The City of Austin spent several years with much public involvement studying viable options in search of a remedy. The City opted to design and build a new, stand-alone crossing of Town Lake adjacent to the existing Lamar Boulevard Bridge instead of widening the historic bridge. An unusual “Double Curve Alignment” concept was proposed, using two curved alignments that meet and overlap over Town Lake, resulting in an hourglass-shaped deck. Final design work on the new bridge proceeded on an accelerated schedule to meet a deadline for obtaining partially matching federal funds for construction. The design engineer and architect worked in parallel to refine aesthetic concepts, while simultaneously developing construction plans. Several alternatives were considered for implementing the “Double Curve” concept; in the end, the most attractive and practicable scheme involved a weathering steel plate girder superstructure with a unique framing plan utilizing opposing curvature exterior girders to achieve a constant deck overhang. Given the complex geometry of the bridge and the constraints associated with construction over water, steel proved to be the most practical material for achieving both the structural and architectural goals of this project. The design engineer used many standard steel bridge design elements in innovative ways to simplify design and construction of the bridge while still conforming to, and indeed celebrating, the architectural concepts associated with the “Double Curve” scheme and the adjacent historic Lamar Boulevard Bridge. The result is a pedestrian and bicycle bridge that will satisfy the special needs of its users and serve as an attractive signature structure well integrated with its surroundings. This paper will highlight how this bridge used conventional materials and standard highway bridge elements in unique and innovative ways to achieve the dual goals of a consistent aesthetic theme and an economical, readily-constructed design. PROJECT HISTORY Six major bridges—carrying a variety of traffic, including automobiles, trucks, trains and, of course, numerous pedestrians and bicyclists—cross Town Lake. One of the major crossings is the existing Lamar Boulevard Bridge, a six-span concrete deck arch bridge built in 1940. The bridge features restrained-but- elegant Art Deco detailing and is a historically significant landmark, added to the National Register of Historic Places in 1994. However, the Lamar Boulevard Bridge suffers from a degree of functional obsolescence, with 10-foot traffic lanes and narrow sidewalks. Significant population growth in Austin over the past decade has resulted in substantial daily vehicular traffic on the bridge. This growth, combined with the bridge’s key location near the center of the north and south shore hike and bike trails, has also resulted in heavy pedestrian and bicycle traffic. The net effect of combining heavy traffic with narrow lanes and sidewalks was a facility that was less than ideal for many of its users. The City of Austin desperately needed an improved facility for crossing Town Lake. Recognizing this and other needs, the citizens of Austin approved one of the largest bond packages in the city’s history, including $8 million specifically earmarked to widen the Lamar Boulevard Bridge. In the early 1990s, the City of Austin secured approximately $950,000 of matching federal funding for the project, as part of the Intermodal Surface Transportation Efficiency Act (ISTEA). In 1995, the City signed a contract retaining a consultant team—led by HDR Engineering, Inc.—to study project alternatives. Under the original contractual agreement, the consultant team evaluated six options, all variations on the theme of widening the existing Lamar Boulevard Bridge. The study phase included inspection of the existing Lamar Boulevard Bridge, evaluation of existing site conditions, traffic modeling and extensive public involvement activities. The consultant team held regular meetings with the City of Austin, the Texas Department of Transportation (TxDOT), the Texas Historical Commission and other key stakeholders. From early in this process, the Texas Historical Commission clearly indicated its preference to avoid any alterations to the existing Lamar Boulevard Bridge (currently owned by the State of Texas), citing its mission to preserve the historical integrity of this landmark structure. In addition, attendees at a 1996 public meeting indicated they did not want more traffic lanes added to the existing bridge. In its transportation work session in March 1998, the city council directed the project team to seek public input in the design of a separate pedestrian/bicycle bridge. Although constructing a separate bridge would not correct all the deficiencies of the existing Lamar Boulevard Bridge, it would create a safer facility for pedestrian and bicycle traffic. The idea of a separate bridge was passionately debated by the city council, but the council ultimately decided to proceed with a design process. The City considered many options for how to develop concepts for the new bridge, ranging from hosting a full-blown design competition among competing architecture and engineering firms to sponsoring a design idea competition open to the general public. Eventually, the City decided to hold a public workshop for key stakeholders. This workshop took place in May 1998, and generated fifteen proposed concepts, including: • Cable-stayed bridge options • Arch bridge options • Relocation of an existing, historical truss bridge • Several variations on beam bridge Figure 1: Original sketch from the May 1998 public options workshop, showing what would eventually become the concept for the final design of the James D. Pfluger One of the most innovative and intriguing Pedestrian and Bicycle Bridge. concepts proposed was the “Double Curve” shown in Figure 1. This unique concept was developed by a group of five workshop participants (Chas Tonetti, Tere O’Connell, Jamie Wise, Rush McNair and Chris Hutson) and focused on the function of the facility suggesting the form of the structure. The concept grew around the “paths of travel”: the function of connecting the trail system along the south shore of Town Lake to activity centers at 5th and 6th Streets and a future public activity center on the north shore. The concept featured two curved alignments, crossing over each other at Town Lake and creating a wide area that would serve as a “gathering [place] at the river” for people to stop and enjoy the view or watch lake activities. This “curved” theme would then be echoed in the design of the structure itself. The City selected four of the 15 proposed concepts for further development by the consultant team’s architect (Kinney and Associates + Carter Design Associates), including the Double Curve concept. The Figure 2: The developed concept prepared by the consultant architect set a criterion of “no straight team architect for the “Double Curve Alignment” option. lines” and conceived a structure with an hourglass-shaped deck plan resulting from the crossing’s curved horizontal alignments and featuring helical ramps and curved connector spans at each end. The architect’s original Double Curve concept is shown in Figure 2. The Double Curve concept remained true to the basic “curved” concept from the public workshop – it was simply developed from a freehand sketch to a higher level of refinement. The four developed concepts were presented to the Austin City Council in September 1998 for feedback, leading to the choice of one selected alternate that served as the base concept for the final design of the new bridge. The Double Curve concept emerged as the selected alternate. With only six months remaining before the matching ISTEA funding would expire, development of aesthetic concepts had to occur in parallel with the preparation of the final Plans, Specifications and Estimates (PS&E). DESIGN Basic Themes and Concepts As the first step in the final design process, the consultant team’s engineers further refined the architect’s developed concept for the Double Curve structure. The helical ramp at the south end of the bridge was eliminated following a geometric evaluation of the vertical profile and existing topography, hike and Figure 3: The “Double Curve Alignment” option, as refined bike trails, and several surface streets. by the engineering team. Structural design, constructability and cost considerations resulted in the decision to change the north end “triangle connector” span from curved to tangent. The engineering team established a span arrangement sensitive to both the adjacent Lamar Boulevard Bridge and users of Town Lake for crossing the lake, trails and streets at both ends of the bridge. As the plan developed, it stayed remarkably true to the original freehand concept sketch from the public workshop. Following these refinements, the consultant team constructed a scaled physical model of the proposed bridge. This model vividly illustrated how the bridge itself would look and, more importantly, how it would fit into its surrounding environment. Figures 4 and 5 show the physical model. Several structural systems—including cast- in-place, post-tensioned concrete box girders; precast segmental concrete box girders; and solid or voided cast-in-place, post-tensioned slab structures—were initially considered for the superstructure. However, the need for quick, simple design—combined with the overriding criteria of ease of constructability over the lake, ability to easily conform to geometric complexities and low construction cost— Figure 4: Physical model of the James D. Pfluger Pedestrian quickly led the engineering team to select and Bicycle Bridge. steel plate girders as the best possible choice for the superstructure system. The architect requested use of weathering steel to provide an “organic” or “natural” appearance to fit with the wooded shorelines of Town Lake. Additionally, this meshed well with cost implications associated with initial and long-term maintenance of coatings systems. The relatively tight construction budget led to the basic theme for the engineering design of the bridge: conventional materials and techniques used in unconventional manners. Basically, the engineering team set out to produce a set of plans closely resembling those for a major steel plate girder highway bridge, with the exception of the basic geometry and aesthetic treatments. The scale of this structure and the selection of steel plate girders as the main structural system made it clear that only heavy highway bridge contractors would be bidding on this project. In order to obtain low, competitive bids, the engineering team had to produce plans that would look familiar to these contractors to give them confidence in their understanding of the project and allow them to bid the project without fear of surprises during Figure 5: Physical model of the James D. Pfluger Pedestrian construction. and Bicycle Bridge. Design Criteria The scale of the James D. Pfluger Bridge suggested that although it is a “pedestrian bridge” covered by the provisions of AASHTO’s Guide Specifications for Design of Pedestrian Bridges (1), it is more the same scale and construction as a typical highway bridge, covered by the provisions of AASHTO’s Standard Specifications for Highway Bridges (2) and AASHTO’s Guide Specification for Horizontally Curved Highway Bridges (3). The design criteria selected for the bridge reflect this assessment of the bridge’s character. The basic bridge design criteria were AASHTO’s Standard Specifications for Highway Bridges (2) and AASHTO’s Guide Specification for Horizontally Curved Highway Bridges (3). However, the live load definitions were modified to reflect the nature of the structure as a pedestrian/bicycle facility, and a 100 psf live load was used as the primary live load. This 100 psf load was applied both over the entire structure and in various “checkerboard” patterns to determine maximum loading effects. 100 psf is a higher criteria than that provided in AASHTO’s Guide Specifications for Design of Pedestrian Bridges (1) but reflects the criteria used in several other similar structures where large groups of people are expected to congregate during major recreational and civic events (such as regattas on Town Lake or Independence Day fireworks shows). In addition, the bridge was checked for H10 truck loads (to reflect the occasional maintenance truck on the bridge), but it proved to control only in a few local loading checks. AASHTO’s Guide Specifications for Design of Pedestrian Bridges (1) also provides criteria for vibration analysis of pedestrian bridges. The consultant team performed vibration analyses of the various units of the James D. Pfluger Pedestrian Bridge and, as expected, found no cause for concern. These provisions of the guide specification are more intended for “lightweight” pedestrian bridge structures (most often constructed). The James D. Pfluger Pedestrian Bridge is of a scale, size and weight more akin to a highway bridge and has commensurate dynamic characteristics. A simple analysis approach quickly demonstrated the bridge’s fundamental frequencies were well above the threshold of concern. Framing Plan The Double Curve alignment resulted in an unusual plan for the bridge (see Figures 6 and 7). At the south end, two ramp structures curve toward each other. The southwest ramp is a two-span continuous unit (Unit A; 86’-120’), while the southeast ramp consists of two single-span units (Units B and C; 48’ and 111’ respectively). Units A and C each utilize three concentric, horizontally-curved composite plate girders. Unit B utilizes three concentric, horizontally-curved composite rolled beams. Unit B has a relatively short span, allowing the use of rolled beams to achieve a shallow superstructure depth, required to maintain Figure 6: Drawing and photograph showing the plan for the adequate vertical clearance over the hike main river crossing of the James D. Pfluger Pedestrian and and bike trail below. Units A, B and C are Bicycle Bridge (photo courtesy of White Photographic relatively narrow (for this structure), with Services, Austin, Texas). a total out-to-out width of 23’-0”. Figure 7: Unit A = Spans No. 1W and 2W (Abut. 1W to Bent 3); Unit B = Span No. 1E (Abut. 1E to Bent 2E); Unit C = Span No. 2E (Bent 2E to Bent 3); Unit D = Spans No. 3-5 (Bent 3 to Bent 6); Unit E = Spans No. 6W 6E and Cross Bridge (Bent 6 to Bents 7W and 7E); Unit F = Spans No. 7-10 (Bent 7W to Abut. 11); and Unit G = Elevated approach section north of Abut. 11 Units A and C meet at Interior Bent 3 and the bridge continues out over Town Lake on Unit D, a three-span continuous steel plate girder unit (114’-114’-114’). Unit D has variable width (minimum width of 31’-3”; maximum width of 42’-0”) and utilizes a very unusual “hourglass” framing plan (discussed in detail later in this paper). Unit D ends at Interior Bent 6, where two ramps split off to the northeast and the northwest. Unit E is a “triangular” unit consisting of Span 6W, a single span unit (104’ span; 21’-0” width) curving from Interior Bent 6 to Interior Bent 7W to the northwest; Span 6E (109’ span; 26’-0” width), a single span unit curving from Interior Bent 6 to Interior Bent 7E to the northeast; and Span 6X (49’ span; 18’-0” width), a single span unit spanning between support brackets on the exterior girders of Spans 6W and 6E. Spans 6W and 6E each utilize three concentric, horizontally-curved composite plate girders, while Span 6X utilizes three tangent composite rolled beams. The use of the relatively shallow rolled beams in Span 6X was possible due to the comparatively short span length and was desirable since it simplified the detailing of the beam ends at their supports. Two further units, Units F and G, were designed running to the northwest over the hike and bike trail, Cesar Chavez Boulevard and Sandra Muriada Way. Unit F is a four-span continuous steel plate girder unit utilizing two concentric, horizontally-curved composite steel girders. Unit F runs from Interior Bent 7W to Interior Bent 11W, where Unit G begins. Unit G is a nine span, conventionally-reinforced concrete slab and T-beam unit that continues to Abutment 20, where a retained fill section with a switchback ramp leads to the sidewalk of the existing Lamar Boulevard. Units F and G were fully designed and detailed for this project but were identified as deductive alternates during bidding and were not included in the final construction contract. Provisions were made at Interior Bent 7E for a future ramp running to the northeast toward the existing Seaholm power plant, which is planned to have a future public facility. Interior Bent 7E also serves to support a short-span, conventionally-reinforced concrete slab and beam span that links Span 6E to the Helix Ramp, a conventionally-reinforced helical ramp structure that rotates through 540 degrees to link up with the hike and bike trail below. A second link to the hike and bike trail is provided via a stairway from Span 6X to the trail below. Units A, B, C and E utilize framing plans and design and construction techniques that are indistinguishable from those used in typical highway bridge construction. Beyond their unique pedestrian live load criteria, these bridges are identical to their highway bridge brethren. Due to its highly unusual framing plan—not at all typical of highway bridge construction—Unit D merits further discussion. The unique framing plan of Unit D—with a tangent center girder and two exterior girders with opposing horizontal curvature—derives from its unusual hourglass plan. The hourglass shape of the Unit D deck plan is a direct result of the refinement of the original Double Curve alignment concept into a workable design plan. The final deck plan was essentially unchanged from its first incarnation as a dimensioned engineering drawing, except for the replacement of the re-entrant corners at the alignment overlap locations with radiused edges. The final framing plan for the girders, however, was the second of two concepts examined. Originally, the engineering team proposed using two tangent girders for Unit D. The girder spacing was set at approximately 20’ to provide a nominal 3’-3” overhang at the narrowest points of the deck plan. With this arrangement, adding a third, center girder would have resulted in an inefficient girder spacing. This arrangement appeared to have merit because it resulted in simpler, cheaper-to-fabricate tangent girders and uniform diaphragm dimensions. However, the engineering team was struggling with the deck design, since at the widest points of the deck plan, the deck overhangs were up to 12’. The team studied various options, including conventionally-reinforced, cast-in-place decks; precast, prestressed deck panels; and cast-in-place post-tensioned decks. The cast-in-place options suffered from constructability concerns related to relatively long span shoring and construction over the lake, while the precast option suffered from complicated detailing for the interfaces between panels and the interfaces with the shear studs provided to obtain composite action in the girders. The tangent girder option also met with disapproval from the project architect, who disliked the uneven shadow line created from the variable width deck overhang and requested a constant width deck overhang instead. The engineering team was initially reluctant to consider such a plan, since it resulted in girders with complex reverse curvature and variable girder spacing. However, after a few minutes of discussion, the engineering team realized this was actually a win-win suggestion that not only resolved many of the architect’s concerns, but many of the engineering team’s concerns as well. The resulting framing plan provided a short, 3’-3” deck overhang and more reasonable interior deck spans – the girder spacing now varied from 12’- 4” minimum to 17’-9” maximum. This allowed the use of a conventionally–reinforced, 12” thick concrete deck that could easily be constructed using typical highway bridge deck construction materials, equipment and techniques. Detailed Design of the Unit D Girders The design of the Unit D girders themselves was surprisingly simple. At first glance, the unusual framing plan with a tangent center girder and two exterior girders with both opposing and reverse curvature would appear to be quite complex to design. However, it was actually simpler to design than a typical horizontally- curved steel plate girder highway bridge. In a typical horizontally-curved steel girder highway bridge, there are two main curvature effects that must be accounted for: global overturning and local lateral flange bending. The global overturning effect results from the center of gravity of any horizontally-curved bridge span being offset from a chorded centerline drawn between the two supports. This global overturning moment manifests itself as an increase in the vertical loading of the girder on the outside of the curve and a decrease in the vertical loading of the girder on the inside of the curve. The “shifting” loads are carried in the diaphragms, which in a typical horizontally-curved highway bridge become primary load-carrying members. The lateral flange bending effect results from the axial forces in the flanges being applied to structural elements that are curved. Thus flanges in compression experience lateral flange bending moments that tend to bow the flanges more, while flanges in tension experience lateral flange bending moments that tend to pull the flanges straight. In Unit D, the global overturning moment effect did not exist. Since the girders were essentially symmetrical in plan, the center of gravity of each span essentially fell on the centerline of the supports. Thus, the only effect of curvature requiring consideration was the lateral flange bending effect, which could easily be included in the design calculations using the simple lateral flange bending moment formula presented in the V-Load curved girder Figure 8: Drawing showing the rolled beam diaphragms. design methodology (4): Md 2 M Lat = 12 Rh Where: MLat = Lateral Flange Bending Moment at Section Under Investigation M = Primary Vertical Bending Moment at Section Under Investigation d = Diaphragm Spacing R = Radius of Horizontal Curvature h = Depth of Girder, from Center of Top Flange to Center of Bottom Flange In fact, the exterior girders in Unit D could be designed as tangent girders with the lateral flange bending moment effects added in manually. This is exactly what the engineering team did for the preliminary design of the Unit D exterior girders. First, a tangent girder design was performed using the commercial STLBRIDGE tangent plate girder design program (5). Then, the stresses resulting from this design were imported into a spreadsheet. Other key parameters, such as the radius of horizontal curvature, diaphragm spacing and girder depth, were input and the lateral flange bending stresses calculated and compared to stress limits calculated in accordance with AASHTO’s Guide Specification for Figure 9: Photo showing the rolled beam diaphragms. Horizontally Curved Highway Bridges (3). The diaphragm spacing and flange sizes were then adjusted to result in an optimized design. This approach proved quite successful, and these preliminary designs exhibited excellent correlation with the results of a later, detailed 3D finite element analysis performed using the proprietary BSDI 3D System computer modeling service (6). Diaphragm Detailing and Design Given the variable girder spacing in Unit D, the detailing of diaphragms warranted special attention during the design process. The engineering team initially examined several options, including plate diaphragms, rolled beam diaphragms, X- frames, W-frames and K-frames. Frame diaphragms are perhaps the most typically used in highway bridge design, Figure 10: Sketch of two column bent. but would have been cumbersome for this bridge due to the variable girder spacing— a 12’-4” minimum to a 17’-9” maximum—in Unit D. With such wide girder spacings, the diagonals of X-frames or K-frames would have inefficiently shallow angles from horizontal. Similarly, W-frames would either have shallow diagonals or require multiple W’s to span between the girders. Combined with these inefficiencies, there would have been virtually no identical diaphragms anywhere in Unit D (because of the 15 degree skew, half of the commonality due to symmetry was also lost). Finally, all the different bracing details would have been aesthetically distracting for recreational boaters under the bridge on Town Lake. As a result, the engineering team selected rolled beam diaphragms (see Figures 8 and 9). In addition to eliminating problems with inefficient diagonal angles in the frame options, rolled beam diaphragms also offered substantial advantages in fabrication. Instead of developing jigs for many different frame diaphragms, the fabricator simply cut rolled sections to the lengths needed for the variable girder spacing in Unit D. In order to reduce weight and keep the diaphragm stiffnesses reasonable, shallow rolled beams were selected for the diaphragms. The difference between the diaphragm depth and the depth of the girders was made up using a simple curved gusset detail at the ends of the diaphragms. Figure 11: Photo of two column bent. Beyond the greater efficiency and simpler fabrication of the rolled beam diaphragms, they also provided a much simpler and cleaner appearance. The curved gusset detail also worked extremely well with the overall curved theme of the bridge. Thus, the choice of rolled beam diaphragms proved to be another example of a win-win solution that best satisfied aesthetic, engineering and construction criteria. Since the diaphragms perform as primary load-carrying members, particularly in the curved ramp structures of Units A, B, C and E, they were carefully designed using diaphragm forces obtained from the results of BSDI 3D finite element analyses. Substructure Design The substructure system for this bridge is Figure 12: Photo of single column bent. quite conventional in its materials but quite unconventional in its form. All substructures are conventionally-reinforced, cast-in-place concrete bents, with pile caps straddling conventionally–reinforced, concrete drilled shaft foundations. Depending on the superstructure width at any given bent location, either two column frame bents or single column hammerhead bents were used. In following the “curved” theme for the structure, the two column bents ended up with a rather unusual appearance. Following an exhaustive aesthetic design process, the final selected bent form included the use of curved columns and a haunched bent cap (see Figure 10 and 11) to allow views of scenic Town Lake through the bents. In and of itself, this two column bent, although not typical in appearance, was not overly unconventional. However, the narrower superstructure sections utilized a single column hammerhead bent. To stay consistent to the established theme, these bents also used curved columns; hence the asymmetrical curved column hammerhead bents (see Figure 12). Although unconventional in appearance, these bents posed no unusual challenges from a design standpoint. The engineering team strove to find ways to simplify the construction whenever possible to help keep construction costs low. For example, the architect had originally drawn essentially freehand curves for the bent cap soffits and column alignments. The engineering team approximated these curves with circular curves set with given PC and PT locations that allowed the contractor to build and reuse one set of Figure 13: Erection of girders at the north end of the main forms for multiple bents on the bridge. river crossing. BIDDING AND CONSTRUCTION – LESSONS LEARNED The construction contract was bid twice. The first round of bidding involved four bidders, all heavy bridge construction contractors. The bids were relatively widely separated and all bids exceeded the City of Austin’s allowable budget for construction of the project. However, detailed evaluation of the bid tabulations revealed some lessons. While the high and low bids were separated by 24 percent, comparison of only the 20 bridge-related bid items (which essentially included all items except the helix ramp, retaining walls at the trails, architectural bridge railing, lighting details, colored concrete wearing course, and landscaping and associated Figure 14: Formwork for the deck at the south end of the main site work) revealed the three low bids river crossing. were separated by only about five percent. Since all bidders on this project were heavy bridge construction firms, they all felt comfortable bidding the project very tightly on the 20 bridge- related bid items because they looked like the typical highway bridge plans they routinely bid. For the second round of bidding, the construction contract was restructured using deductive alternates that allowed the City to restructure the project after bidding in an effort to fit the project within the construction budget. This second round of bidding included two bidders and resulted in a winning bid that, with Deductive Alternate 1 (which eliminated Units F and G [see Figure 7]), fit the City’s construction budget. The construction contract was awarded to Jay-Reese Contractors, Inc. of Austin, Texas, in April 2000, after an up- to-the-last minute debate on project options and costs in the Austin City Council. Groundbreaking occurred on May 15, 2000, and the contractor was allocated a one-year schedule with incentive bonuses for early completion. After allowances for foul weather days and change orders, the contractor completed the project ahead of schedule on June 16, 2001. Construction usually progressed very smoothly, a Figure 15: Concrete placement for deck at the sound end of testament to the philosophy followed in the main river crossing preparing the bridge plans. Generally, construction of the main lake crossing (Units A through E) followed the typical construction sequences and appearances of routine, large highway bridge construction projects (see Figure 13). The contractor was well-versed in this type of construction and the plans were relatively error-free. There were only two minor added-charge change orders during the main lake bridge construction. In fact, there were relatively few problems with the main lake bridge construction at all. One of the minor problems that did occur is highlighted below as a “lesson learned.” The single biggest construction problem was associated with the alignment of the single column bent caps. At the insistence of the architect, all bent caps for the main Figure 16: View of the new James D. Pfluger Pedestrian and lake crossing were detailed with widths Bicycle Bridge from under Bent 7E. identical to the thickness of their supporting columns. For the two column bents, this did not cause problems. However, the asymmetrically- curved single columns had a recurrent problem of shifting forms; during concrete placement, the column forms would shift and twist slightly. The rotational twist resulted in the corners of the tops of the column being out of position only ¼” on the worst column, but since the bent cap was the same width as the column, there was no tolerance for this out-of-alignment twist and the cap forms had to conform to it. As a result, a ¼” misalignment at the top corners of the column cased a misalignment of up to 2” at the end of the bent caps. Since steel forms were used for the bent caps, it was practically impossible to recover the correct alignment, as the forms could not be deformed to move the ends of the bent cap back into their proper positions. This misalignment complicated all other bent cap details, such as the positions of anchor bolts and bearings. In retrospect, the engineering team should have been more adamant about disallowing this questionable detailing. Many other unique or unconventional details worked well during construction, but for the most part, the smooth construction was due to detailing consistent with typical highway bridge detailing. This made it easy for an Figure 17: View of the new James D. Pfluger bridge looking experienced, qualified highway bridge south at deck level. Note planters (not complete in this photo) contractor to construct this bridge. that help to delineate “gathering” places from travel lanes. CONCLUSION This was a challenging, exciting and rewarding project for the design team, even though the number of stakeholders often made the project process arduous. The short schedule for final design and the need to prosecute detailed structural design simultaneously with conceptual aesthetic design added to the challenges. However, the design team had a truly inspired and well-conceived basic concept from which to work and the unusual nature of the bridge always kept the project interesting. The engineering team successfully used conventional materials and techniques in unconventional manners to produce a set of plans that resulted in very competitive bidding for construction. The unit cost for the main lake crossing worked out to be approximately $150/SF. This is high compared to typical highway bridge construction costs in Texas (which can range as low as $34-40/SF), but given the aesthetic and functional requirements of the project and in comparison to equivalent projects this unit cost relatively low. This approach also greatly facilitated the construction phase, with very few Requests for Information (RFIs), and no significant change orders during construction of the main lake bridge. Assembling the plans for this unique pedestrian/bicycle bridge to look Figure 18: On June 16, 2001, hundreds of like the plans for a typical highway bridge was a gambit runners sprinted over the new James D. that paid off well. Pedestrians and bicyclists will now Pfluger Pedestrian and Bicycle Bridge, have a new, safe, attractive route across Town Lake and marking the bridge’s grand opening. The the bridge’s owners and users are sure to be delighted bridge was named in honor of the Austin with their new facility. All in all, this project architect who conceived the trail system on demonstrates that aesthetics, functionality, either side of Town Lake. constructability and cost effectiveness need not be mutually exclusive goals for a bridge project. REFERENCES 1. American Association of State Highway Transporation Officials (AASHTO), Standard Specifications for Highway Bridges, 16th Edition, 1996, with Interim Updates through 1998. 2. American Association of State Highway Transporation Officials (AASHTO), Guide Specifications for Design of Pedestrian Bridges, August 1997. 3. American Association of State Highway Transporation Officials (AASHTO), Guide Specifications for Horizontally Curved Highway Bridges, 1993, with Interim Updates through 1995. 4. National Steel Bridge Alliance (NSBA), “V-LOAD Analysis,” Highway Structures Design Handbook, Volume 1, Chapter 12, pg. I/12/16, December 1996. 5. Bridgesoft, Inc., “STLBRIDGE – Design of Continuous Steel Bridge Girders,” Omaha, Nebraska, 1997. 6. Bridge Software Development International, Ltd. (BSDI), “Bridge-System (SM), 3D System,” Coopersburg, Pennsylvania, 1987.