This report is one of a series of reports documenting the methods and findings of a multi-year, multi-disciplinary project coordinated by the Pacific Earthquake Engineering Research Center (PEER and funded by the California Earthquake Authority (CEA). The overall project is titled “Quantifying the Performance of Retrofit of Cripple Walls and Sill Anchorage in Single-Family Wood-Frame Buildings,” henceforth referred to as the “PEER–CEA Project.”
The overall objective of the PEER–CEA Project is to provide scientifically based information (e.g., testing, analysis, and resulting loss models) that measure and assess the effectiveness of seismic retrofit to reduce the risk of damage and associated losses (repair costs) of wood-frame houses with cripple wall and sill anchorage deficiencies as well as retrofitted conditions that address those deficiencies. Tasks that support and inform the loss-modeling effort are: (1) collecting and summarizing existing information and results of previous research on the performance of wood-frame houses; (2) identifying construction features to characterize alternative variants of wood-frame houses; (3) characterizing earthquake hazard and ground motions at representative sites in California; (4) developing cyclic loading protocols and conducting laboratory tests of cripple wall panels, wood-frame wall subassemblies, and sill anchorages to measure and document their response (strength and stiffness) under cyclic loading; and (5) the computer modeling, simulations, and the development of loss models as informed by a workshop with claims adjustors.
This report is a product of Working Group 4: Testing, whose central focus was to experimentally investigate the seismic performance of retrofitted and existing cripple walls. Two testing programs were conducted; the University of California, Berkeley (UC Berkeley) focused on large-component tests; and the University of California San Diego (UC San Diego) focused on small-component tests. The primary objectives of the tests were to develop descriptions of the load-deflection behavior of components and connections for use by Working Group 5 in developing numerical models and collect descriptions of damage at varying levels of drift for use by Working Group 6 in developing fragility functions. This report considers two large-component cripple wall tests performed at UC Berkeley and several small-component tests performed at UC San Diego that resembled the testing details of the large-component tests.
Experiments involved imposition of combined vertical loading and quasi-static reversed cyclic lateral load on cripple wall assemblies. The details of the tests are representative of era-specific construction, specifically the most vulnerable pre-1945 construction. All cripple walls tested were 2 ft high and finished with stucco over horizontal lumber sheathing. Specimens were tested in both the retrofitted and unretrofitted condition. The large-component tests were constructed as three-dimensional components (with a 20-ft 4-ft floor plan) and included the cripple wall and a single-story superstructure above. The small-component tests were constructed as 12-ft-long two-dimensional components and included only the cripple wall. The pairing of small- and large-component tests was considered to make a direct comparison to determine the following: (1) how closely small-component specimen response could emulate the response of the large-component specimens; and (2) what boundary conditions in the small-component specimens led to the best match the response of the large-component specimens.
The answers to these questions are intended to help identify best practices for the future design of cripple walls in residential housing, with particular interest in: (1) supporting the realistic design of small-component specimens that may capture the response large-component specimen response; and (2) to qualitatively determine where the small-component tests fall in the range of lower- to upper-bound estimation of strength and deformation capacity for the purposes of numerical modelling. Through these comparisons, the experiments will ultimately advance numerical modeling tools, which will in turn help generate seismic loss models capable of quantifying the reduction of loss achieved by applying state-of-practice retrofit methods as identified in FEMA P-1100Vulnerability-Base Seismic Assessment and Retrofit of One- and Two-Family Dwellings. To this end, details of the test specimens, measured as well as physical observations, and comparisons between the two test programs are summarized in this report.
Full List of PEER Reports: click here.