Seismic Performance of Nonstructural Partition Walls

Graduate Students: Rodrigo Retamales

Principal Investigator: Andre Filiatrault

Co-Principal Investigator: Gilberto Mosqueda

Project Completion Date: 11-30-2009

The NEES Nonstructural Grand Challenge Project: “Simulation of the Seismic Performance of Nonstructural Systems” aimed to improve the seismic performance of buildings through analyzing, testing, and modeling of common nonstructural systems (ceilings, piping systems, and partition walls). The first phase of this project, emphasizing on nonstructural partition walls, was conducted at the University at Buffalo, using the Nonstructural Components Simulator (UB-NCS). Walls were subjected to quasi-static and dynamic protocols to assess their seismic fragility.

Fifty full scale light gauge steel studded gypsum partition wall specimens were tested. Thirty-six of them were tested in-plane, while the other fourteen were tested out-of-plane. Sixteen different wall configurations for in-plane testing were developed by the experimental team at UB in coordination with the Practice Committee and Advisory Board of the NEES Nonstructural Grand Challenge Project. The partition walls were categorized into the six groups, based on similar detailing and observed mechanical response. The group categories were: Commercial Slip Track; Commercial Full Connection; Institutional Slip Track; Institutional Full Connection; Partial Height; and Improved Details.

Slip track and full connection configurations vary in the connection of studs and gypsum boards to the top and bottom tracks. The primary differences between the commercial and institutional configurations is the thickness of steel framing material, respectively 18 and 30 mils, and the typical details of wall intersections. Partial height walls are 8 ft tall with diagonal braces stabilizing the walls. Improved details corresponded to special designs developed and tested to offset damage states to higher drift ratios or remove them completely.

Damage observed in the partition walls was mainly concentrated at wall intersections. Initial damage observations included rocking of screws attaching the gypsum to the top and bottom tracks, openings at wall intersections, and crushing of gypsum at wall corners. Higher levels of damage to the partition walls included bending and cracking of gypsum at wall intersections and bending in boundary studs. The most severe damage observed included tearing in steel tracks around concrete fasteners, fasteners pulling through tracks, bending in top track flanges of transverse walls, and hinges forming in studs.

This project is supported by the George E. Brown, Jr. Network for Earthquake Engineering Simulation program of the National Science Foundation under Grant No. CMMI-0721399 (NEES Research).