J.S.
Pei, R.D. Martin, C.J. Sandburg, and T.H.-K. Kang
School of Civil Engineering and Environmental Science, University of Oklahoma, 202 W. Boyd St., Room 334, Norman, Oklahoma 73019-1024
Shear capacity of real-world prestressed concrete girders designed in the 1960’s and 1970’s is a concern because AASHTO Standard Specifications (AASHTO-STD) employed the quarterpoint rule for shear design, which is less conservative for shear demands than today's AASHTO LRFD.
Shear tests were conducted on two full sized AASHTO Type II girders, one of which had been in service for nearly forty years before being replaced due to irreparable damage. As a means to improve analysis, additional experimental data are used to determine the effective prestressing force of these specimens. Comparisons are then made between three design codes and experimental results to assess the condition and safety of similar girders currently in use. The comparison of nominal shear capacities according to the 11th Edition AASHTO-STD (1973), AASHTO LRFD (2004), and ACI 318-08 including provisions for strut and tie models is carried out. Composite sections are analyzed with varying properties (concrete compressive strength, transverse reinforcement spacing, etc.) for AASHTO Type II, III and IV prestressed concrete girders. By examining the ratios of nominal shear capacity to demands for each code, considering all load and resistance factors, these code-to-code comparisons are better able to identify girders that may be deficient according to today's standards than a direct comparison of nominal capacities alone. Experimental results for shear capacity of real-world girders are compared with the codes' nominal capacities to check if the girders' are structurally sufficient. Girders are also rated according to AASHTO LRFR (2005) to check if AASHTO inventory and legal loads are permissible.
Preliminary results are presented on the estimation of effective prestressing force using static test data; an inverse problem is formulated where input and output are measured to determine system properties. Nominal shear capacity is particularly sensitive to effective prestressing force under current design codes, so it's important to have accurate values when making calculations. Although there are methods for determining long-term prestress losses, they apply to a wide variety of structural members and do not necessarily reflect the condition of girders and their uncertain histories studied here. In attempt to get more accurate results for effective prestressing force, span varying flexural stiffness is assumed. This assumption reflects that girders with long histories may be damaged, obvious or otherwise. The load balancing method is used in conjunction with the principle of virtual work to express camber and Δ/P in terms of effective prestressing force and/or piecewise-constant flexural stiffness. Least squares techniques are used to solve these overdetermined problems. The key challenges include the problem formulation considering time-dependent properties, the selection of appropriate initial values, and the interpretation the results.
For a given girder, the ratio of nominal shear capacity to demands has generally decreased with newer codes. Girders having this ratio near one for the 11th Edition AASHTO-STD may be structurally deficient. Experimental results from this study, however, indicate that the girders’ actual capacity exceeds nominal capacity of current codes. Additional shear capacity tests should be performed on more real-world girders to get a more definitive conclusion.
School of Civil Engineering and Environmental Science, University of Oklahoma, 202 W. Boyd St., Room 334, Norman, Oklahoma 73019-1024
Shear capacity of real-world prestressed concrete girders designed in the 1960’s and 1970’s is a concern because AASHTO Standard Specifications (AASHTO-STD) employed the quarterpoint rule for shear design, which is less conservative for shear demands than today's AASHTO LRFD.
Shear tests were conducted on two full sized AASHTO Type II girders, one of which had been in service for nearly forty years before being replaced due to irreparable damage. As a means to improve analysis, additional experimental data are used to determine the effective prestressing force of these specimens. Comparisons are then made between three design codes and experimental results to assess the condition and safety of similar girders currently in use. The comparison of nominal shear capacities according to the 11th Edition AASHTO-STD (1973), AASHTO LRFD (2004), and ACI 318-08 including provisions for strut and tie models is carried out. Composite sections are analyzed with varying properties (concrete compressive strength, transverse reinforcement spacing, etc.) for AASHTO Type II, III and IV prestressed concrete girders. By examining the ratios of nominal shear capacity to demands for each code, considering all load and resistance factors, these code-to-code comparisons are better able to identify girders that may be deficient according to today's standards than a direct comparison of nominal capacities alone. Experimental results for shear capacity of real-world girders are compared with the codes' nominal capacities to check if the girders' are structurally sufficient. Girders are also rated according to AASHTO LRFR (2005) to check if AASHTO inventory and legal loads are permissible.
Preliminary results are presented on the estimation of effective prestressing force using static test data; an inverse problem is formulated where input and output are measured to determine system properties. Nominal shear capacity is particularly sensitive to effective prestressing force under current design codes, so it's important to have accurate values when making calculations. Although there are methods for determining long-term prestress losses, they apply to a wide variety of structural members and do not necessarily reflect the condition of girders and their uncertain histories studied here. In attempt to get more accurate results for effective prestressing force, span varying flexural stiffness is assumed. This assumption reflects that girders with long histories may be damaged, obvious or otherwise. The load balancing method is used in conjunction with the principle of virtual work to express camber and Δ/P in terms of effective prestressing force and/or piecewise-constant flexural stiffness. Least squares techniques are used to solve these overdetermined problems. The key challenges include the problem formulation considering time-dependent properties, the selection of appropriate initial values, and the interpretation the results.
For a given girder, the ratio of nominal shear capacity to demands has generally decreased with newer codes. Girders having this ratio near one for the 11th Edition AASHTO-STD may be structurally deficient. Experimental results from this study, however, indicate that the girders’ actual capacity exceeds nominal capacity of current codes. Additional shear capacity tests should be performed on more real-world girders to get a more definitive conclusion.
No comments:
Post a Comment