1298378158

SEISMIC PERFORMANCE OF LOW-RISE LIGHT FRAMED WOOD BU1LDINGS

LASołtk*

Failures in sili piąte connections and homes shifting

•Supanrkory Research inęineer, Forest Products Laboretory, U.S. Department of Apricuiture. The La bort tory is meintalned et Madison, Rńacomin in cooperethn with the Unlrertlty of WHcomin.

Abstract Thit peper reelewt literatura on the per-formence of wood ttructuret In eerthguekat, axam-inm component end building retponte, and dfecuum current deeign phlloaophy. Wood ttructuret perform edegueta/y when they ero tymmetrk In pleń end ełeyetion end heve edeguete thoor woflt Bod performance occurt mhen there k e lock of or nontym-metric errengement of recking wel/t.

Iow-rise light-framed wood buildings have performed adequately in eerthquakes provided they acted as a unit. had adequate shear walls, and were reasonably symmetric in plan and eievation. The 1971 San Fernando earthquake. however. did extensive damage to wood structures and forced a rethlnking of design and construction concepts.

This paper reviews literaturę on the performance of wood structures in earthquakes, examines component (floors. walls. and roofs) and building response. and discusses current design phllosophy. This review i$ limited to low-rlse light-frame wood buildings with a diaphragm/shear wali lateral load-resisting system. This type of construction is used in a large number of resldential, commerclal. and industrial buildings. Post and beem, laminated frame and arch. and pole buildings as weil as other structures such as timber brldges. retaining walls, and poles are not included.

PAST EARTHQUAKE EXPERIENCE

Prlor to the 1971 San Fernando earthquake. Iow-rise timber structures. In generał, performed well when subjected to selsmic ground motion. Keenan (11 reviewed ancient structures in Asia and the Far East that had been subjected to numerous eerthquakes but showed little dlstress. Timber structures performed well In generał (2, 3) in the 1964 Alaska earthquake that measured 8.6 on the Richter scalę. Failures were observed when inadequate lateral bracing occurred due either to lack of wali sheathing under the slding or to large openings. particul8rly near corners.

The 1971 San Fernando earthquake (6.6 Richter). however. did extensive damage to timber structures. Of the 58 deaths and 5,000 injuries, one report had two (4J and another four (5J deaths oocurring in resldences. No estimate of injuries occurring In residences was given; but sińce the earthquake occurred at approximately 6 a.m. (local time), it could be assumed that many occurred in homes. McCIure [6] reported that about 20,500 single family houses were damaged, of which 730 were demolished or requlred major rehabllitation. He reported single family home dollar losses at between $58 and $114 million (1971 dollars); this dollar loss was larger than that of any other building category in the private sector. Housner (4) reported 6,000 residences damaged of which 450 were declared unsafe for occu-pancy. The discrepancy between these reports is related to geographic area induded and whether data were obtained by indlvidual survey or building permit records. Blume and Scholl (7) estimated that 30% to 60% of the Iow-rise (mostly wood) buildings in the exposed area were damaged; 3lthough resistant to collapse many of these buildings experi-enced condemnation-level damage. No specific numbers were given for commercial or industrial buildings.

The primary cause of overall residentlal damage (4, 6-10] In the San Fernando earthquake was inadequate lateral support. The lack of or nonsym-metrical arrangement of racking walls caused damage or collapse. most vulnerable were two-story and spllt-level homes with a garage on the first floor.