Performance of partially and fully anchored wood frame shear walls under monotonic, cyclic and earthquake loads

Approved: Rakesh Gupta Thomas H. Miller The objectives of this study were to evaluate the performance of wood frame shear walls under monotonic, cyclic and earthquake loads by: (1) estimating the variability of shear wall performance, (2) comparing the performance of walls under each loading protocol, (3) evaluating the effects of anchorage on wall performance, and (4) evaluating the performance of walls qualitatively and quantitatively with respect to code defined performance measures. Tests were conducted on 2440 x 2440 mm (8 x 8 ft) wall specimens with 38 x 89 mm (2x4) Douglas-fir studs at 610 mm (24 in) on center. Two 1220x2440x11.1 mm (48x96x7/16 in.) structural oriented strand board (OSB) panels were installed vertically and fastened with 8d nails (0.113x2.375 in.; 2.87x60.33 mm) at 152 mm (6 in.) on center around the edges of the panels and at 305 mm (12 in.) on center in the panel fields. Two 12.7 mm (0.5 in.) gypsum wallboard (GWB) panels were installed vertically on the face of the wall opposite the structural panel sheathing. Partially anchored walls were connected using two 12.7 mm (0.5 in) A307 anchor bolts installed at 305 mm (12 in.) from each end of the specimen. Fully anchored walls were constructed identical to partially anchored walls except that SIMPSON Strong-Tie PHD-2A hold-downs were installed at the ends of the wall and connected with 15.9 mm (0.625 in.) Grade 5 anchor bolts. Sets of tests consisting of eight partially anchored walls and two fully anchored walls were conducted using the ASTM E564 monotonic protocol and CUREE cyclic test protocol for ordinary ground motions for a total of twenty walls. Eight walls were tested using two historical subduction zone ground motions scaled to a 10% in 50 year probability of exceedence for the Seattle area, with a 4545 kg (10,020 lb) seismic mass. Two partially anchored and two fully anchored walls were tested using each earthquake time history. Cyclic tests generally exhibited a coefficient of variation (COV) that was lower than monotonic tests. The COV was 14.9% and 9.6% for peak load (Ppeak), 15.6% and 17.2% for displacement at peak load (∆peak), 25.1% and 13.8% for energy dissipation (E), and 32.7% and 13.4% for initial stiffness (Ge) for monotonic and cyclic tests, respectively. When comparing average values between monotonic and cyclic tests, one tail T-tests showed that backbone parameters Ppeak, ∆peak, E, and Pyield are significantly different at an alpha level of 0.1. Performance parameters for fully anchored walls exhibited increases over partially anchored walls by a factor of about 2.5 for Ppeak and ∆peak and a factor of almost 9 for E. Partially anchored walls with dead load applied exhibited an increase in Ppeak proportional to the resisting moment applied by dead load. Partially anchored walls had a consistent failure mode (edge breakout along sill plate) regardless of test protocol used. Fully anchored walls demonstrated different failure modes between the monotonic and cyclic testing. Monotonic tests caused primarily nail pull-through type failures in the sheathing connections and crushing of the gypsum in the screwed GWB connections. The fully reversed cycling of the CUREE tests caused some nails to withdraw and GWB screws to fracture. A comparison of the test results with FEMA 356 m-factors also shows that the ductility of partially anchored walls is below the acceptance criteria for shear walls with structural panel sheathing. Earthquake tests had a lower Ppeak and Ge than similar walls tested monotonically. Peak load values were approximately equal to the walls tested using the CUREE cyclic protocol. The initial stiffness of the cyclically tested walls was between the values from the monotonic and earthquake tests. The monotonic tests exhibited ∆peak values that were more reflective than the cyclic tests of the earthquake results for fully anchored walls. The cyclic tests were more reflective of the earthquake results for the partially anchored walls. Fully anchored walls exhibited a “yield limit” beyond which variability in performance increased dramatically. This limit appeared to be associated with severe pinching in the hysteresis curves, also corresponding to a lengthening in the period of the drift response. The displacement at peak load from the cyclic tests was close to this “yield limit.” Different ground motions produced similarly shaped loaddeflection envelope curves, but had different maximum drifts. The positive and negative envelope curves were asymmetrical. The maximum transient drift (∆max), peak to peak drift (∆p-p), and cumulative drift (∆cumulative) showed similar dynamic response between fully and partially anchored walls, but uplift was much lower for fully anchored walls. The fully anchored walls tested under the subduction zone intraplate earthquake (FA-SE19) met the 3.0% drift limit required for collapse prevention performance as defined by FEMA 356. The maximum transient drift for all other earthquake tests exceeded the collapse prevention limit with values ranging from 4.0 to 5.8%. The earthquake tests caused more nail withdrawal than the cyclic tests and nearly all of the GWB screws to fracture along the panel edges. The average fundamental period of the walls tested was 0.342 sec and 0.485 sec for fully anchored and partially anchored walls, respectively. The period of the walls lengthened as the walls approached failure to average values of 0.684 sec and 0.658 sec for fully and partially anchored walls, respectively. CUREE tests generally appear to give a more conservative estimate of shear wall performance under actual earthquake loads than monotonic tests. Performance of Partially and Fully Anchored Wood Frame Shear Walls Under Monotonic, Cyclic and Earthquake Loads