Delegates to the 86th annual meeting of the Transportation Research Board (TRB) held in Washington, D.C., in January were exposed to dozens of papers
Delegates to the 86th annual meeting of the Transportation Research Board (TRB) held in Washington, D.C., in January were exposed to dozens of papers highlighting the structural, aesthetic, and economic benefits of precast concrete. Those presentations were offered among 3,000 scheduled discussions addressing topics of interest to public and private sectors responsible for transportation infrastructure. Following are selected summaries of precast-related papers reviewed at the TRB-sponsored event. A companion report on ready mixed research will follow in March. More information about TRB can be obtained at www.trb.gov.
FAST PRECAST BRIDGE REPLACEMENT
When the Ohio Department of Transportation needed to replace a 700-ft. concrete arch bridge of memorable design, it turned to HNTB Corp. and a precast scheme for one-year completion of the project. Precast was especially useful in meeting aesthetic requirements to rebuild the Perry Street Bridge over the Maumee River at Napoleon, as the new structure had to resemble the existing span and fit its footprint, according to Barker, Walters and O’Leary of HNTB Corp. in their technical paper, How to Build a 700-ft., $20 Million Concrete Bridge in One Year.
Designed in 1928 and built in 1930, Perry Street Bridge was a sand-filled, concrete barrel arch structure providing the only local crossing. For its replacement, the authors report, Ohio DOT set out four design criteria or parameters, pursuant to stakeholder input:
- The new bridge had to look like the old bridge, with the same shape and number of spans.
- Construction could not disturb the river bottom during the two-month period of April 15 to June 15 due to spawning fish.
- The new bridge had to be constructed essentially within the same footprint of the existing bridge.
- The old bridge had to be demolished and replaced with the new bridge in a single year.
The utilization of precast/prestressed concrete enabled the replacement in one year, the authors note. Our initial concept was to incorporate as much precast as possible, allowing the casting of components in the winter prior to the year of construction.
On-site construction, thereby, would be limited to building foundations and erecting large pieces of concrete. Since we had to replicate the arches, which determined the size of the concrete beam sections, we decided to attempt to design the new bridge for zero tension in the concrete, similar to the design criteria used for segmental box girders, say Barker, Walters and O’Leary. The replication of the arches provided plenty of concrete to carry the stresses, which enabled us to achieve the zero-tension state in both longitudinal and transverse directions, except for a few longitudinal sections under live load applications.
Exactly like the old bridge, the new structure had to comprise seven 103-ft.-long spans. Due to variable depth sections required to replicate the arches, an entire span length would be substantially too heavy to haul to the erection site. Accordingly, the HNTB team incorporated cast-in-place splices in the spans to obtain a module length and weight that a truck trailer could accommodate.
We also outlined a construction schedule to determine how much time we had to build the seven spans, the engineers recall. We had to make a decision regarding the casting of the bridge deck. Should we cast it in place conventionally, or should we design and detail something in precast? The preliminary construction schedule quickly showed us that we didn’t have sufficient time to cast the 700-ft. by 73-ft. bridge deck in place. They add, The final concept incorporates a combination of precast/prestressed construction schemes, all of which have been used individually, but to our knowledge, have never been used in this exact combination.
Additionally, the team was concerned with possible extra costs related to contractor-perceived risks. To help allay these concerns, we met with representatives of the precast industry in Ohio and held two contractor seminars where we explained the concept and asked for both precaster and contractor input, Barker, Walters and O’Leary affirm. Both groups provided valuable information and seemed to appreciate the opportunity to participate. The bid results indicated some success of alleviating risk concerns.
During concept development, the engineers also devised a scheme that would enable construction of drilled shafts throughout the spring, while remaining in compliance with the two-month river work restriction. We decided to drill shafts through the existing piers, which were founded with spread footings on shale, the authors report.
The final cross-sections developed during the conceptual phase included a precast deck bulb-T, which has proved efficient in carrying longitudinal loads, and a variable depth slab of minimum 8.5-in. thickness cast on the bulb-T section. To simulate the arch of the old bridge, the deck bulb-T sections were set to a 12-ft. depth over the piers and 4-ft. depth at mid-span. The beams were to be spliced at the spans’ quarter points, resulting in modules of approximately 50 ft. in length and similar weights. End-span modules at 75 ft. in length and heavier in weight were the single exception.
Precasters indicated they could build the modules to a 0.5-in. straightness tolerance for the 50-ft. lengths, because the components are extremely stiff. Thus, we detailed the joints separating adjacent slabs to be ±0.5 inch and required the joints to be filled with a stiff grout, which had to be trowled into the joint, the engineers explain.
Key to rapid construction of the piers was reuse of existing pier bases. The erection of precast modules, however, resulted in construction loads exceeding the capacity of the spread footings. Moreover, the condition of concrete comprising the existing piers was questionable, though cores indicated 8,000-psi concrete strengths. Consequently, the resultant design specified piers supported on shafts drilled through the existing pier bases, which were to be left in place from the waterline down.
As substructure work following installation of the drilled shafts occurred above the waterline, it was not subject to fish spawning restrictions. To further expedite substructure construction, the piers were designed as cap and column-type structures with precast pier caps. Dividing the precast caps into three sections to be post-tensioned in the field served to reduce their weight. Abutment beam seats also were designed as precast sections.
Since aesthetic requirements dictated that the piers should be wall-type piers, project plans included a cast-in-place fascia wall surrounding the pier columns. The fascia wall could be constructed, if necessary, after superstructure placement in order to meet the schedule.
The superstructure consists of three precast elements: pier modules, mid-span modules, and end-span modules. It was analyzed using HNTB’s in-house T187 computer program that allows the bridge to be constructed piece by piece within the model, while forces generated during erection as well as time-dependent effects are tracked. Typical for Ohio DOT projects, the design strength for the precast modules’ concrete was 5,500 psi at release and 7,000 psi at 28 days.
A combination of pretensioned and post-tensioned tendons was incorporated in the module design. Pretensioned strands were used to control stresses during shipping and erection, and the number of pretensioned strands varied for each module type. Continuity post-tensioning was provided by 12 0.6-in. strands in each of two 3-in.-diameter ducts. Shortening the duct through which strands would be fed, tendon couplers were used on the post-tensioned strands, allowing the contractor to reuse some of the strongbacks and begin the transverse post-tensioning earlier. Transverse post-tensioning ducts were designed to be installed parallel to the top of the flange, and flange depth was varied to achieve the required eccentricity.
Since FruCon Construction was the prime contractor for a large precast segmental cable-stayed bridge in Toledo, some 30 miles east of Napoleon, it chose to fabricate the modules in the Toledo casting yard and haul them to Napoleon. FruCon constructed three casting beds for the pier modules, mid-span modules, and the end-span modules. It purchased two sets of forms each for the pier modules and the mid-span modules. The total project includes 54 pier, 45 mid-span and 18 end-span modules. Commencing in early March 2005, casting was completed in July 2005, so that all modules were ready by the time on-site erection started.
HALF-BOX CULVERT SHEAR TESTS CLARIFY CANADIAN, AASHTO SPECS
Destructive test results involving half-box-culvert specimens indicate that the Canadian Highway Bridge Design Code (CHBDC) sectional method of shear design works better than AASHTO specs, particularly in terms of variability of predictions. In Investigating the Shear Strength of Concrete Box Culverts, authors Yee, Bentz and Collins of the University of Toronto summarize the results of a preliminary series of reinforced concrete shear experiments on precast box culverts. Specimens were half-boxes tested under uniform load with a passive tie-bar to allow full determination of internal moment throughout the testing process. Of a total 12 specimens tested, six failed in shear.
With the support of Ontario Concrete Pipe Association, research on shear strength of reinforced precast box culverts was conducted at the University of Toronto. Prior to the year 2000, concrete box culverts in Ontario generally followed the guidelines of the province’s specification, but in that year the Canadian Highway Bridge Design Code (CHBDC) became the governing document for culvert design.
The CHBDC is similar in many respects to AASHTO LRFD Bridge Design Specifications, the authors state. The earlier shear provisions had been based on empirical rules, while the CHBDC shear design provisions were derived from a theoretical model called the Modified Compression Field Theory (MCFT). The new provisions suggested that shear reinforcement would be necessary for some box culverts that previously were exempt from the requirement.
As shear reinforcement is costly, determining where and when the steel is needed was a priority. Specimens dry cast according to standard industry practice by Hanson Pipe & Products Canada in Cambridge, Ont., were fabricated about two feet long in the direction of water flow in the culvert. After curing, the specimens were cut in half so that a nominally identical pair of specimens was delivered to the testing labs. Thus, prototype specimens were tested first to confirm the stiffness of tie rods; and, in most cases, two test results were obtained for each specimen type and geometry.
The authors conclude that the CHBDC sectional shear design method is conservative when applied to culverts and appears to capture effect of differences between the specimens. AASHTO Box Culvert Equations work less well than the CHBDC sectional method of shear design, they add, particularly in terms of variability of predictions.
AASHTO LRFD special box culvert shear design rules are conservative for thin slabs, but unconservative for thicker slabs 16-plus inches deep. In addition, the authors assert, they provide a less uniform level of safety than the CHBDC provisions.
While some culvert designs are shear critical and require reinforcement when the culverts are buried at sufficient depth, Yee, Bentz and Collins contend, available predictive tools can capture the load-deformation behavior of the specimens. The proposed method of load determination does an excellent job at predicting the failure load of these specimens, the affirm. Work is ongoing to determine whether the suggested method is safe for all practical cases.
STEAM-CURED PRECAST WITH SILICA FUME SHOWS NO SHRINKING
Steam-cured, high-performance precast containing silica fume meets all Florida DOT requirements for 12-, 18- and 24-hour curing periods, demonstrating excellent durability, note Yazdani, Filsaime and Islam, University of Texas-Arlington, in their paper, Accelerated Curing of Silica Fume Concrete.
While silica fume is a common admixture for high performance concrete mixes, its use leads to increased water demand. Consequently, Florida DOT allows only a 72-hour continuous moist-cure process for concrete containing silica fume. Because 72 hours at a precast plant may constitute an eternity, the industry seeks shorter cure times to increase form turnover.
Small laboratory specimens exposed to various steam-curing durations were tested for concrete compressive strength, surface resistivity, and shrinkage. All steam-cured samples demonstrated excellent durability, besides meeting Florida DOT requirements for the 12-, 18-, and 24-hour curing periods. No shrinkage cracking was observed in any samples up to one year age, the authors report, recommending that Florida DOT allow the 12-hour steam curing for concrete with silica fume.
Silica fume, also known as microsilica, has been used as a concrete property-enhancing material and as a partial replacement for portland cement for over 25 years, the researchers explain. In fresh concrete, silica fume affects the water demand and slump. The concrete water demand increases with the increased amounts of silica fume, due primarily to the high surface area of the silica fume. Silica fume is also known to affect the time of setting and bleeding of fresh concrete. Mechanical properties of silica fume concrete, such as creep and drying shrinkage, have been known to be lower than that of concrete without silica fume. At 28 days, the compressive strength of silica fume concrete is significantly higher than concrete without silica fume. Silica fume is also linked to chemical attack resistance, decrease of permeability, and enhancement of the chloride ion penetration resistance of concrete.
However, they add, the surface of silica fume concrete tends to dry quickly, subsequently causing shrinkage and cracking prior to final setting. Therefore, early-age moist curing of silica fume concrete is important.
Accelerated curing is favored at precast/prestressed concrete plants, the authors observe, emphasizing that this type of curing is advantageous where early strength gain in concrete is important, or where additional heat is required to accomplish hydration, as in cold weather. Accelerated curing reduces costs and curing time in the production of precast members resulting in economic benefits. But, a primary concern with accelerated curing is the potential for increased moisture loss during the curing process, as mentioned in ACI 517.2R.
While Florida DOT Standard Specifications 346 allows the use of silica fume in concrete as 7 to 9 percent replacement of cementitious material, together with the usage of high range water reducing admixtures, ASTM C 1240 requirements must be met. Florida DOT spec modifications, effective January 2004, specify extended moist curing requirements for silica fume concrete.
Immediately after finishing, curing blankets must be applied to all exposed surfaces and saturated with water. The moist curing must continue for a minimum of three days; and, immediately afterwards, two coats of curing compound must be applied and the surfaces kept undisturbed thereafter for a minimum of seven days.
The authors assert:
- Steam curing of silica fume precast elements can be conveniently accomplished with current technology and facilities available in large precast yards. Meeting and even exceeding Florida DOT specifications regarding temperature regimens during steam curing is easily achieved.
- Steam-cured silica fume concrete can achieve target specified minimum compressive strengths. In this study, most steam-cured laboratory specimens and all field pile specimens reached a 28-day target strength of 41.37 MPa (6,000 psi) for Class V mixes. All steam-cured samples continued to gain in strength over time. At 365 days, samples displayed significantly higher compressive strength than that at 28 days.
- Steam-cured samples displayed lower compressive strengths at all ages than their moist cured counterparts. Such results are consistent with previous research on silica fume concrete.
- Steam curing times of 12, 18 and 24 hours do not seem to play a major role in controlling concrete compressive strengths. No consistent pattern of maximum strength was displayed by samples from a single source of steam cured duration.
- Surface resistivity of all samples increased significantly over time. Moist-cured specimens gained surface resistivity at a much higher rate than the steam-cured specimens. At an age of 365 days, moist-cured samples displayed the greatest resistivity.
- Steam-cured specimens displayed low-very low permeability and very low permeability at 28- and 364-day measurement, respectively. Greater surface resistivity indicates lower permeability and increased long-term durability of concrete.
- All steam- and moist-cured specimens showed a general increase in shrinkage over time. Steam-curing duration did not appear to be a major factor in shrinkage rates. At later stages, e.g., 364-day age, longer steam-curing periods such as 18 and 24 hours accelerated shrinkage growth.
The authors suggest that precast yards consider a 12- to 24-hour period of steam curing for silica fume concrete elements, while maintaining the practice of curing silica fume concrete products 72 hours total. The balance of the 72 hour requirement may be included Û through curing blankets or curing compound application Û after the end of steam curing.
THREE-POINT BEARINGS PROVIDE UNIFORM SUPPORT
Three-point bearings are among innovative design features of the Interstate system bridges incorporating high-strength, prestressed concrete U-beams in New Mexico, say Bowser, Jones, Camp and Meyers of New Mexico DOT in Design of I-40 Bridges in Guadalupe County, N.M. The I-40 Overpass Bridges comprise a set of three medium-span structures accommodating two lanes of traffic each. The project team addressed several design challenges with innovative solutions, besides incorporating aesthetic elements from the early stages.
Designed continuous for live load, the two-span bridges feature high-strength prestressed concrete U-beams and cast-in-place decks, the authors write. Three-point bearing provides uniform support of the girders, they assert. To alleviate problems with harping and end debonding, strands were added in the top flange, debonded in the center of the beam, then cut after release. Spread footings on top of MSE [mechanically stabilized earth] wall fill support the abutments to reduce differential settlement between the approach and abutment.
Differential settlement between pier and abutments was included in the design to account for footing support variances. The semi-integral abutment stub wall was shortened on each end to allow the integral wing walls to slide freely, solving a common movement restraint problem, Bowser, Jones, Camp and Meyers explain. 3-D modeling was used to assist visualization of the completed form. Artwork was created to form uniform recessions in the concrete.
Average bridge costs for these structures, including aesthetics, were below state average, the authors report. Savings can be attributed to several factors, including design considerations of simplicity and ease of construction; consultation during design with suppliers and contractors; use of repetitive parts and materials; and, a creative bid process.
The bridges carry U.S. 84 and local road traffic over I-40. The three spans replaced original bridges, on an offset from the original bridge, allowing free design of the new bridges within standard roadway parameters. The Interstate cross section the bridges were to span consisted of four 12-ft. driving lanes (two in each direction), 10-ft. outside shoulders and 4-ft. inside shoulders in each direction, with a 52-ft. median. Although the necessary span length including required clear zone was 168 ft., an extra lane in each direction was added to accommodate future expansion of the Interstate, creating an actual span length of 192 ft.
Tying the bridges together aesthetically was a prime consideration in choosing bridge structures, the authors note. That was most easily accomplished if all the bridges had identical structures, they add. Artwork could then reflect each locality, while the bridges maintained visual continuity. Therefore, each structure had to fulfill the requirements for all bridges. This task was simplified by the fact that the Interstate highway maintained its cross section throughout the entire project range. The decision was made to put one pier in the center of the median, creating a two span bridge.
Prestressed U-beams were selected for form as well as structural stability. Both I-beams and U-beams were considered, and analysis of I-beams resulted in a requirement of five 63-in.-deep beams. Instead, four 54-in.-deep U-beams were selected to obtain the required clearance with minimal fill requirements. The district also thought the U-beams had better visual appeal.
The concrete mix design for the U-beams had a strength of 9,500 psi at 28 days, with a 5,500-psi strength at release, a standard New Mexico mix design. An 8.5-in. continuous for live load deck was made composite to the beams. The cast-in-place deck was constructed of 4,000-psi concrete. Deck and integral diaphragms were placed in one continuous pour. Strands were 0.6-in. diameter. The larger strands resulted in less total strands required, reducing congestion, especially at the ends.
U-beams do not accommodate strand harping easily, Bowser, Jones, Camp and Meyers assert, adding that space for harping exists only in the sides of the beams. The sides are sloped at one and a half vertical to one horizontal, and widen as they rise.
Conversations with engineers from one of the local precast producers, Steven Ruiz and Amor Solano from Rinker Materials (now Coreslab Structures), indicated that harping for these beams presents difficulties, as the position of the strand relative to the edge of the beam is not constant throughout the length of the harped section. An analysis of the stress indicated that the condition could also cause unwanted torsional stresses in the webs.
Continuous beam analysis over the required span length resulted in excessive compressive forces on the bottom of the beam, unless they were debonded at the ends in excess of AASHTO code recommendations Û a solution unacceptable to the designers. Bursting forces around debonded strands at the end of the beams can cause cracking around the strands leading to premature deterioration of the concrete. The increased bursting force due to the larger-diameter strands required special consideration of debonding forces. The problem was addressed by adding strands in the top flange and debonding them in the center of the beam. The resulting need for debonding at the ends was minimal.
Debonded top strands were cut at the center between release and erection. Timing of the cut was left up to the supplier to assist with control of beam camber. The top strands also made placement of other reinforcing steel easier, and the supplier now recommends the use of top strands in U beams, whether required by design or not.
Adding diaphragms at the end of the beams resulted in 54-in.-deep concrete throughout the bearing area of the component. Both the abutment stem walls and pier caps were 3 ft. deep. The bearing area was 32 in. _ 10 in. for abutments and 20 in. _ 10 in. for piers. Analysis of shear requirements indicated that further disbursement of the load was unnecessary.
To fulfill the function of welcoming visitors to New Mexico and to Guadalupe County, aesthetics were considered from the beginning of the project. 3-D modeling with basic rendering was used to aid visualization of the completed form. Girders, piers, and MSE walls were designed to complement one another and provide an overall pleasing form to the bridge.
Artwork was incorporated into the MSE walls, wing walls, piers, and barrier curbs, designed using a technique of disposable foam or plastic of a given thickness to create recessions in the concrete. The recessed areas were then painted in colors contrasting with the background. Recessed areas give definitive paint boundaries for ease in initial and maintenance painting, the authors affirm. Painting the recessed areas provides contrast and definition, while adding color and interest to the bridge. The cost for forming and painting with this system was minimal with respect to other art system costs.