Click for next page ( 11

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 10
8 One of the units was instrumented and subjected to load tests. about the same time in Alabama have not exhibited the extent Characteristics of the girders and results from the unit tested and type of cracking observed on the I-565 bridge. are summarized below: In conclusion, several points can be made: The girders in the subject span on the main line struc- 1. The effects of the thermal gradient across the depth of ture were PCI BT-54 girders. There were nine girders in the superstructure may be significant in the design of the cross section, spaced at 8 ft. The pier-to-pier dis- bridges with precast concrete girders made continuous. tance was 100 ft, with a design span of 98.5 ft between 2. The Alabama bridges continued to perform as designed bearings. even with significant cracking at the ends of some gird- The girders were manufactured in December 1988 and ers in the cross section. January 1989. The deck was cast in late April and early 3. Very few bridges with precast concrete girders made May 1989. continuous have experienced significant cracking of the Cracks in the girders were monitored during load test- type observed on the I-565 and US 280 bridges in Ala- ing of the bridge. The maximum crack opening reported bama. In general, performance of this type of bridge from the effect of two load-testing vehicles was reported has been very satisfactory. to be about 0.0055 in. A maximum crack opening of 4. The thickness of the continuity diaphragm and the about 0.030 in. was reported during the course of a day embedment of the girders into the diaphragms may have as the sun heated the deck slab. been a significant factor in causing the observed crack- The camber at midspan was observed to be about 0.41 in. ing in the girders. due to the solar effect. ALDOT concluded that the cracking was caused by pos- INITIAL ANALYTICAL STUDIES itive moments that developed as a result of the thermal gradient across the section. Using the information gained in the literature search and Measured deflections from two load-testing vehicles from the survey, the next task was to propose specimen con- placed on the bridge were only about 75% of the deflec- figurations for the experimental work. To accomplish this task, tion computed using a finite element model; therefore, an analytical model was created. This model, a modernized ver- it was concluded that the presence of the cracks did not sion of BRIDGERM (11), was named RESTRAINT and works significantly affect the structural behavior of the bridge. within a standard spreadsheet program (see Appendix A). The program models a two-span continuous structure. The As a result of the experience of cracking with this and other support conditions assume that there is a support at each end bridges, ALDOT no longer allows the use of prestressed con- of the girder (see Figure 3) because this was the most common crete girders made continuous. support condition identified in the survey. This is also consis- Subsequent to the investigation of the I-565 bridge in tent with the support condition used in analysis program Huntsville, another bridge was found with cracking in the gird- BRIDGERM given in NCHRP Report 322 (11). RESTRAINT ers near the continuity diaphragms. This bridge was located used flexibility-based analysis by discretizing the span and the on US 280 in Lee County, Alabama. The bridge was con- diaphragm into several elements. Prior to using the restraint structed using AASHTO Type II girders with mild reinforce- program, moment curvature relationships are developed for ment extending into the continuity diaphragms to make a pos- the cross section. Any convenient method of determining the itive moment connection. The girders were also embedded moment-curvature relationship (e.g., hand calculations, com- several inches into the continuity diaphragm. The behavior of puter program, finite element analysis, or experimental data) this bridge was investigated by monitoring crack widths and can be used. For this study, the RESPONSE Program (22) was performing a load test. The investigation produced the same used to find the moment-curvature relationship. These data are results regarding the cause of cracking and the remaining then input into the spreadsheet. capacity of the bridge as for the previous bridge. The time the diaphragm and deck are cast as input into RESTRAINT, assuming that release of the pretensioning force is time = 0 (this can be a different time based on the age Conclusions Drawn of girder; however, the reference is made to the time after the from ALDOT Bridge Cracking release of post-tensioning). Because some states cast the dia- phragm first, the program allows the time the diaphragm and The cracking observed in the girders and continuity dia- deck are cast to be different. Basic material properties are phragms of the Alabama bridges was significant and defi- also input. nitely a matter of concern. However, similar situations with With the basic information available, the program calcu- widespread cracking of this type have not been reported in lates the internal moments that would result from creep of the the United States for bridges with prestressed concrete gird- prestressed girder and shrinkage of the girder and deck. Creep ers made continuous. In fact, similar bridges constructed at and shrinkage strains are found from the relationships given

OCR for page 10
9 Figure 3. RESTRAINT Model. in the American Concrete Institute Report 209 (23). The pro- performs a consistent deflection analysis. The center reac- gram also accounts for loss of prestressing force using the tions are removed to make the system statically determinate. method given in the Precast/Prestressed Concrete Institute Using the curvatures, the deflection at the center supports can Handbook (24). In the span, shrinkage of the deck and girder be found. The required reactions needed to restore the center is assumed to be uniform, while creep caused by dead load support deflection to 0 are found. The other reactions are plus prestressing force is assumed to be parabolic. At the dia- found from equilibrium and are used to calculate the conti- phragm there is no prestressing, so the creep is 0. Since the nuity moments. The continuity moments are then added to slab and diaphragm are usually cast together, the differential the other moments, and the entire analysis is repeated until shrinkage between them is assumed to be 0. the answer converges. Once the internal moments are known, the program adds To verify this program, the 1/2-scale I girders used in the the dead-load moments. If desired, a live load, consisting of original PCA tests were modeled (6). The results showed a a point load at midspan, can be included and the moment reasonable agreement with the experiment (see Figure 4), from this force is also added. The program then divides each although the effect of differential shrinkage (the first peak in span into 10 or more elements (defined by the user). A single the curve) was overestimated. element is used for the diaphragm. With the moments known, A parametric study was conducted on a two-span bridge the program can determine the curvature of each element consisting of AASHTO Type III girders. The spans were 65 ft, from the moment-curvature relationship. The program then and the girder spacing was 8 ft. A 2-ft-wide diaphragm was 2.5 PCA Tests Present Study 9.5 2 7.5 Reaction at Center Support (kips) Reaction at Center Support (kN) 1.5 5.5 1 3.5 0.5 1.5 0 Observed Cracking -0.5 of Diaphragm5 -0.5 Predicted Cracking of Diaphragm -2.5 -1 -4.5 0 150 300 450 600 750 Time from Removal of Deck Formwork (Days) Figure 4. Comparison of RESTRAINT (Present Study) with PCA data.