Durable, prestressed/precast concrete panels offer an option for pavement construction, according to one study among several on precast performance presented at the 81st annual meeting of the Transportation Research Board (TRB) in Washington, D.C. As TRB research on ready mixed concrete was covered in March (see “TRB Reports Spotlight Advances in Ready Mixed,” page 76), summaries of representative presentations on precast concrete follow. More information about the TRB meeting can be obtained by contacting the Transportation Research Board, 2001 Wisconsin Avenue, N.W., Green Building, Washington, D.C. 20007, or visit TRB's web site at http://trb.org.

Precast panels offer option for concrete pavements

Prestressed/precast concrete panels are feasible for pavement construction, according to casting and field experience in Texas, contend David Merritt and B. Frank McCullough, Center for Transportation Research (CTR), and Ned Burns, University of Texas-Austin, in “Development of a Precast/Prestressed Concrete Pavement near Georgetown, Texas.” The authors report, “Precast concrete panels can be cast and cured in a controlled environment, stockpiled, and set in place in a short amount of time, allowing for construction to take place overnight or on weekend operations.”

In March 2000, University of Texas at Austin's CTR completed a Federal Highway Administration (FHWA)-sponsored feasibility study to develop a method for using precast panels to expedite concrete pavement construction. The final recommendation was a concept involving the use of prestressed/precast panels. A staged implementation strategy for testing and refining the proposed concept was presented.

Accordingly, a pilot project on the northbound frontage road of Interstate 35 near Georgetown, Texas, was initiated to further explore and develop the concept. It entailed the construction of 2,300 ft. of precast pavement on either side of a new bridge.

“The project will incorporate prestressing as a means of reducing the required thickness of the pavement and tying the precast panels together,” the authors note. “In the end, this project will demonstrate the viability of all of the different aspects of the proposed concept and the use of precast pavements in general.”

Roughness and pavement noise, sometimes associated with slipformed concrete pavements, can be more pronounced with precast panel pavements. The authors address this issue: “It is believed that a smooth enough riding surface can be attained with full-depth panels and occasional diamond grinding, as needed.”

Previous experience with prestressed pavements has shown that prestress in both the longitudinal and transverse directions is essential. “Transverse prestress will be incorporated through pretensioning all of the panels during fabrication, and longitudinal prestress will be incorporated through post-tensioning all of the panels together after placement,” Merritt, McCullough and Burns report.

The authors say base preparation is important, so panels are resting on a flat surface and fully supported. They emphasize that the method for ensuring vertical alignment of adjacent panels is critical, so ridges are not created at the panels' joints, thereby degrading ride quality. They also note that new research supports the possibility of placing a thin (25-50-mm) asphalt leveling course smooth and flat enough to evenly support the panels, which also permits traffic onto the leveling course prior to panel placement. “To ensure that vertical alignment is achieved between adjacent panels, the proposed concept features continuous shear keys cast into the edges of the panels, which will interlock the panels as they are set in place,” the researchers explain.

Three types of panels would be required, Merritt, McCullough and Burns explain: base, joint and central stressing. “All of the panels are pretensioned in the longitudinal direction (transverse pavement direction),” they say. “Post-tensioning ducts cast into each panel in the transverse direction (longitudinal pavement direction) will allow the panels to be post-tensioned together after they are set in place.”

Base panels are the “filler” panels between the joint panels and central stressing panels. The number of base panels required will depend on the post-tensioned slab length (between expansion joints). The joint panels feature an armored expansion joint, similar to bridge deck expansion joints, which will absorb the significant expansion and contraction movements of the post-tensioned slabs.

The precast pavement for the Georgetown project was designed for an equivalent fatigue life — in terms of expected 80 kN (18-kip) ESAL (equivalent single axle load) applications — to that of a 355-mm (14-in.) continuously reinforced concrete pavement. “Although a 355-mm pavement is a much thicker pavement than needed for a frontage road, the purpose of this project is to simulate what might be used on main Interstate pavement lanes,” the authors affirm.

Panels for the Georgetown precast pavement project were cast on a 400-ft. long-line bed. “The casting allowed for 10 full-width (11-m) panels, and up to 20 partial-width panels to be cast at one time,” they note. “The advantage of long-line casting is that the pretensioning strands can extend continuously the full length of the casting bed, passing through all of the panels.”

After release of prestress, the pretensioning strands between each of the panels are cut and the panels are removed from the forms. Long-line casting does require special attention for the side forms, however, to avoid imperfections or misalignment that might prevent the keyed panel edges from matching up.

The mix design used for the precast pavement panels is a mix similar to that used for precast/prestressed bridge beams. It is a seven-sack (Type III) mix with a water-cement ratio of 0.42 and super-plasticizer for increased workability. A mix of this nature — not typically used for pavements — was necessary to develop sufficient strength for release of prestress and removal from the forms the following day. The specifications for the precast panels required that the concrete reach a minimum compressive strength of 24.1 MPa (3,500 psi) at release of prestress and a 28-day compressive strength of 34.5 Mpa (5,000 psi).

The mix was also required to be fluid enough during casting that a carpet drag finish could be applied at sheen loss. The use of an intermediate curing compound, similar to monomolecular film, was also needed to minimize water loss during casting, the authors add.

When casting the panels, handheld concrete vibrators and a vibratory screed were used to ensure a flat surface and proper consolidation of the concrete around the keyways, post-tensioning ducts, and pockets. After any necessary hand finishing, a carpet drag finish was applied to each of the panels along their length — transverse to the flow of traffic. The finish followed almost immediately after the vibratory screed as very little bleed water migrated to the surface. Immediately following the carpet drag, a double-coat of curing compound was applied to the panels to minimize further water loss from the concrete, a particularly important measure as many of the panels were cast during July and August.

Panels were cured in forms overnight before releasing the prestress, then removed from the forms and stacked. Once the panels were stacked, wet mats were applied to the stack and a tarp was used to cover the panels. The panels were allowed to cure in this condition until 48 hours from the time of casting. In total, 123 full-width (11-m) and 216 partial-width (5-m and 6-m) panels were cast.

Tom Kuennen is Principal of Expressways Publishing, Wheeling, Ill., specializing in public works construction.

Code methods overpredict HPC beam prestress loss

Conventional code methods for calculating prestress loss overpredict for steam-cured high performance concrete (HPC) beams, based on field measurements, note Rola Idriss Ph.D., P.E., New Mexico State University, and Amor Solano, M.S., CSR Prestress, Albuquerque, in “Prestress Losses In High Performance Concrete Beams: Actual Versus Predicted.” Emphasizing the significance of reliable estimates, the authors state, “Accurate prediction of the prestress losses is a very important step in the design of a highly stressed high performance concrete (HPC) girder and can affect the service behavior of the girder, such as deflections, camber and cracking.”

Current methods for calculating prestress losses according to the American Association for Highway and Transportation Officials (AASHTO) and Prestressed Concrete Institute (PCI) were developed for conventional concrete, they contend. The higher the prestressing force in the girder, the larger the concrete compressive strength needed at release of the prestressing strands. “To help achieve the higher release strength,” the authors explain, “the precast plants have been using longer curing times. Many precast plants are using steam curing to increase the curing rate. A better understanding of the effects of this heating on HPC is needed.”

Therefore, an optical fiber monitoring system was designed and built into a three-span high performance concrete highway bridge. Located 15 miles west of Albuquerque on Route 66, the Rio Puerco Bridge is a three-span, prestressed concrete structure with simple support for dead load and continuous for live load.

The data collected was analyzed to calculate the prestress losses in the girders, compare the results to the predicted losses using available code methods, and establish a better understanding of the properties and behavior of high performance concrete.

HPC was used for the cast-in-place concrete deck and the prestressed concrete beams. Normal strength concrete was used for the substructure. The primary members of the bridge consist of four I-beam type BT-1600. The beams are prestressed by 42 Grade 270 steel tendons, 26 straight and 16 draped.

The beams were fabricated at CSR Prestress in Albuquerque. The casting and installation of the sensors in the four beams took place in July 2000. Measurements were collected during casting of the beams, steam curing, strand release, storage of the beams, through transport, casting of the slab and the full year following construction.

Following release the beams were stored in the casting yard, and data was collected four times a day. The beams were then transported to the site, where construction was under way. Monitoring started again the week prior to deck casting and was maintained continuously during the procedure. Following deck casting, measurements were collected daily.

The concrete was designed for a release strength of 48.3 MPa (7,000 psi) and a 28-day strength of 68.9 MPa (10,000 psi). Cylinders were taken during the pouring of the girders, steam cured alongside the girders and tested in the laboratory. “Monitoring the beams during the fabrication process gave an insight into the effect of the steam-curing temperature,” the authors note. Material properties and prestress losses were measured using the built-in sensor system. Idriss and Solano state that the research conducted during the course of the project supports the following conclusions:

• Current code methods overpredict the prestress losses for steam-cured HPC. “There is a need to modify these methods or develop new ones to better predict prestress losses for HPC,” the authors affirm.

• Estimates by four methods — the complex time-consuming PCI general, the ACI-ASCE, the AASHTO LRFD refined, and the simple AASHTO LRFD Lump Sum — fell within a close range. “In sum, all four methods over-predicted the prestress losses for HPC and were found to be very conservative,” they report.

• HPC exhibited much less creep in the month following transfer than factored by the PCI general method, the authors report. “The steam-curing temperature can have a significant effect on the early prestress losses in HPC,” they contend. “A lower curing temperature was associated with larger early prestress losses following transfer. This is most pronounced in the month following transfer and tapers off with time.”

• An imbedded monitoring system can be used to measure the in-situ material properties of the concrete.