The seismic response of an instrumented tilt-up wall building is initially investigated under the action of four earthquake ground motions that were recorded at the site . The strongest base acceleration was 0.18g recorded during the Big Bear, California, earthquake. Building response was obtained from nine sensors located in the building, and these data were used to evaluate the behavior of the building and the accuracy of the numerical calculations obtained from a three- dimensional building model. A site visit indicated that the response of the walls to the recorded ground motions had been primarily linear elastic; however, minor cracking had occurred in some walls and pilasters due to the strongest ground motions. Since the building was occupied, it was difficult to inspect the roof diaphragm and connections to the walls. Fourier spectra analysis of the recorded response indicated an increase in the fundamental period of the building during each of the four ground motions.
Initial elastic dynamic analyses compared the calculated accelerations and displacements with those obtained from corresponding sensor lo cations. In general, these comparisons were quite good. Base shear demands from recorded ground motions were compared with the design base shear used for the original design as well as with the current requirements of the 1997 UBC. Force demands on the roof to wall connections we re also determined and plotted. Force contours including in-plane shear, in-plane moment, and out-of-plane moment were plotted and used to estimate maximum stresses in the reinforced concrete walls and timber diaphragm. These results indicate that in-plane shear in the diaphragm is critical followed in importance by the glulam beam to pilaster connection and the out-of-plane moment in the wall. The in-plane shear capacity of the walls is well above the minimum required by the codes and also well above that demanded by the four recorded ground motions.
Following these analyses, the effects on the building of three stronger ground motions having pulse-type displacement characteristics were evaluated. The base shear demands of these ground motions exceed the minimum code design re quirements in both principal directions. A comparison of the forces in the connections and force contours in the walls and diaphragm with their capacities indicates that of these, only the in-plane shear capacity of the walls has sufficient capacity to remain elastic. As in the case of th e ground motions recorded at the site, the critical demand is the in-plane shear in the roof diaphragm.
With these results, a nonlinear three-dimensional model of the building was developed within the constraints of the computer program (SAP2000). Nonlinear elements were incorporated for the connections of the roof di aphragm to the glulam beams and purlins, and the nonlinear behavior of the diaphragm was modeled using a Hrennikoff model of the continuum. Inelastic characteristics for these components were obtained from the results of a limited number of component tests conducted at the University of California at Irvine. This modeling permitted consideration of both old and new connections to the pilasters, along with the effects of dense and sparse nailing in the diaphragm. Static pushover analyses and dynamic time history analyses were conducted. The results of these analyses indicate that nonlinear behavior can have a significant effect on the force and displacement demands of the different components. The use of the new connections, along with the dense nailing in the diaphragm, produces the best results; however, the displacement (ducti lity) demands may require a dditional strengthening of the critical roof diaphragm.
Inclusion of the vertical ground accelerations in the analyses of the elastic model with nonlinear connections had a significant effect on the vertical shear force component of the glulam to pilaster connections , with increases of more than 100% being obtained under pulse- type ground motions. However, this connecti on force component, including the increase, is approximately 33% of the more dominant axial force component. Moreover, this increase, and the fact that two of the force co mponents are perpendicular to the grain, indicate the need to test these critical connections under triaxial loading.
Full List of PEER Reports: click here.