Preliminary Testing of Wood Structural Panel Shear ... - Bay Area … Cyclic Shear Wall Testing.pdf · Preliminary Testing of Wood Structural Panel Shear Walls Under Cyclic (Reversed)

50

0

80

70

20

30

40

10

60

90

5

023

87

6

9

4

1

R E S E A R C H R E P O R T

Report 158

Preliminary Testing of Wood Structural Panel Shear Walls Under

Cyclic (Reversed) Loadingby John D. Rose • March, 1998

Technical Services Division

THE ENGINEEREDWOOD ASSOCIATION

APA

RR,W260,158.0 11/9/01 10:53 AM Page 1

Wood is the right choice for a host of construction applications. It is theearth’s natural, energy efficient and renewable building material.

Engineered wood is a better use of wood. It uses less wood to make morewood products. That’s why using APA trademarked I-joists, glued laminated

timbers, laminated veneer lumber, plywood and oriented strand board is constructive ... for the environment, for innovative design, and for strong, durable buildings.

A few facts about wood.■ We’re not running out of trees. One-third of the United States land base– 731 million acres – is covered by forests. About two-thirds of that 731million acres is suitable for repeated planting and harvesting of timber. Butonly about half of the land suitable for growing timber is open to logging.Most of that harvestable acreage also is open to other uses, such ascamping, hiking, and hunting. Forests fully cover one-half of Canada’s land mass. Of thisforestland, nearly half is considered productive, or capable of producing timber on asustained yield basis. Canada has the highest per capita accumulation of protected naturalareas in the world – areas including national and provincial parks.

■ We’re growing more wood every day. American landowners plant more than two billion trees every year. In addition, millions of trees seednaturally. The forest products industry, which comprises about 15 percentof forestland ownership, is responsible for 41 percent of replanted forestacreage. That works out to more than one billion trees a year, or about

three million trees planted every day. This high rate of replanting accounts for the fact thateach year, 27 percent more timber is grown than is harvested. Canada’s replanting recordshows a fourfold increase in the number of trees planted between 1975 and 1990.

■ Manufacturing wood is energy efficient.Wood products made up 47 percent of allindustrial raw materials manufactured in theUnited States, yet consumed only 4 percentof the energy needed to manufacture allindustrial raw materials, according toa 1987 study.

■ Constructive news for a healthy planet. For every ton of wood grown, a young forest produces 1.07 tons of oxygen and absorbs 1.47 tons ofcarbon dioxide.

Wood. It’s the constructive choice for the environment.

Percent of Percent ofMaterial Production Energy Use

Wood 47 4

Steel 23 48

Aluminum 2 8

NOTICE:The recommendations inthis report apply only topanels that bear the APAtrademark. Only panelsbearing the APA trademarkare subject to theAssociation’s qualityauditing program.

RATED SHEATHING

EXPOSURE 1SIZED FOR SPACING32/16 15/32 INCH

000PS 1-95 C-D PRP-108

THE ENGINEERED

WOOD ASSOCIATIONAPA

©20

01 A

PA –

TH

E EN

GIN

EERE

D W

OO

D A

SSO

CIA

TIO

N •

ALL

RIG

HTS

RES

ERVE

D. •

AN

Y C

OPY

ING

, MO

DIF

ICAT

ION

, DIS

TRIB

UTI

ON

OR

OTH

ER U

SE O

F TH

IS P

UBL

ICAT

ION

OTH

ER T

HA

N A

S EX

PRES

SLY

AUTH

ORI

ZED

BY

APA

IS P

ROH

IBIT

ED B

Y TH

E U

.S. C

OPY

RIG

HT

LAW

S.

Be ConstructiveWOOD

RR157&158InsideFront 11/9/01 12:21 PM Page 1

3

ABSTRACT

Damage to wood-framed residentialbuildings in the January, 1994 earthquakeat Northridge, California raised questionsabout performance of shear wall assemblieswhen subjected to cyclic (reversed) loading.Factors such as the effect of monotonic(one-directional) loading versus cyclicloading on shear wall stiffness and strength;contribution of dissimilar materials tostiffness and strength of the wall; shear walldisplacement limits, to minimize cosmeticand structural building damage during anearthquake; ductility; fatigue resistance offasteners; and slip and deformation of shearwall hold-down connectors are considera-tions when walls are subjected to cyclicloading from seismic conditions.

Evaluation of shear wall performancehistorically has been based on monotonictesting, which is used to develop provi-sions for prescriptive wall bracing andengineered shear walls for incorporation

in model building codes. There arepresently no recognized U.S. standardsfor conducting cyclic load testing ofassemblies. However, a modification of asequential phased displacement testprocedure – developed cooperatively in1987 by U.S. and Japanese engineers andscientists participating on the JointTechnical Coordinating Committee onMasonry Research (TCCMAR) – has beenused as a basis for developing a cyclicload test method for evaluating perfor-mance of structural assemblies.

Eight 8-foot x 8-foot shear walls withwood structural panel sheathing weretested using the modified TCCMARprocedure. Sheathing materials andfastener schedules were varied to obtaininformation on the load-displacementcharacteristics and load capacity of typicalconstructions. Gypsum wallboard wasincluded as a sheathing material in onetest, to evaluate the contribution of dis-similar sheathing materials to wall stiffnessand strength. Shear wall performance, asdetermined by a simplified cyclic load test

procedure used for evaluating compo-nents for steel structures, also was inves-tigated for comparison with resultsobtained by the modified TCCMAR proce-dure. This information has been used toassist in developing a standardized testprocedure for cyclic (reversed) loading ofstructural assemblies, and in understand-ing the performance of wood-framed shearwalls under cyclic loading, and the factorswhich govern their performance.

Reported by:

John D. RoseSenior EngineerTechnical Services Division

Reviewed by:

William A. Baker, P.E.Manager, Market Support ServicesTechnical Services Division

RR,W260,158.0 11/9/01 10:53 AM Page 3

5

0

80

70

60

90

5

023

87

6

9

4

1

4

CONTENTS

ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . .3

CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . .4

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . .5

OBJECTIVE . . . . . . . . . . . . . . . . . . . . . . . . . . .6

TEST SPECIMENS AND MATERIALS . . . . . . . .6

Framing . . . . . . . . . . . . . . . . . . . . . . . . . .6

Wall Sheathing . . . . . . . . . . . . . . . . . . . . .8

Fasteners . . . . . . . . . . . . . . . . . . . . . . . . .8

Hold-Down Connectors . . . . . . . . . . . . . .10

TEST SET-UP AND PROCEDURE . . . . . . . . . .10

TEST RESULTS AND DISCUSSION . . . . . . . .15

Comparison of Calculated vs. Measured Shear Wall Displacement . . . . .17

Load-Displacement Relationships . . . . . . .18

Shear Wall Load Capacity – Cyclic vs. Monotonic Tests . . . . . . . . . . . .19

Plywood vs. OSB Wall Sheathing . . . . . . .20

Contribution of Gypsum Wallboard . . . . .21

Pneumatically Driven Nails . . . . . . . . . . .21

TCCMAR vs. “Ramped” Cyclic LoadingProcedure . . . . . . . . . . . . . . . . . . . . . . . .21

Shear Wall Hold-Down Slip . . . . . . . . . . .21

DESIGN AND CONSTRUCTIONCONSIDERATIONS . . . . . . . . . . . . . . . . . . .22

DESIGN ANALYSIS . . . . . . . . . . . . . . . . . . . .23

CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . .27

LITERATURE CITED . . . . . . . . . . . . . . . . . . . .29

ACKNOWLEDGMENT . . . . . . . . . . . . . . . . .29

APPENDIX A – LOAD-DISPLACEMENT CURVES (APA/UNIVERSITY OF CALIFORNIA - IRVINE TESTS) . . . . . . . . . . .30

APPENDIX B – TEST RESULTS AND LOAD-DISPLACEMENT CURVES(SCHMID/UNIVERSITY OF CALIFORNIA - IRVINE TESTS) . . . . . . . . . . .34

APPENDIX C – DESIGN EXAMPLE . . . . . . . . 37

Laboratory Accreditation

APA – The Engineered Wood Association is committed to providingits clients with high-quality service and information through doc-umented test procedures and thorough, accurate collection ofdata. As a part of that commitment, a Quality Program has beenestablished by APA based on the international document ISO/IECGuide 25: General Requirements for the Competence ofCalibration and Testing Laboratories. The APA Quality Programfollows the Accreditation Criteria and Requirements for TestingOrganizations (CAN-P-4), National Accreditation Program forTesting Organizations, Standards Council of Canada (SCC) andProcedures and General Requirements of the National VoluntaryLaboratory Accreditation Program (NVLAP) as outlined in theNational Institute of Standards and Technology (NIST) Handbook150. APA is accredited as a testing laboratory for the specificscopes of accreditation by the following agencies:

National Voluntary Laboratory Accreditation Program (NVLAP),as a Testing Organization (NVLAP Lab Code No. 100423-0)

Standards Council of Canada (SCC), as a Testing Organizationand a Certification Organization (Registration No. 89)

National Evaluation Service (NES), as a Compliance Assuranceand Inspection Agency and as a Testing Laboratory(NER-QA397 and NER-TL527)

Japanese Agricultural Services, Foreign Testing Division (JAS), asa Foreign Testing Organization (Nos. 5-471, 3-1565, 7-2341,3531 and 3532)

City of Los Angeles, as a Compliance Assurance and TestingAgency (No. 22192)

Metro - Dade County, as a Testing Laboratory (Certification No.97-1021.06)

This report contains data generated through testing of engineeredwood products according to various test methods. Many testmethods conducted by APA are accredited by the organizationslisted above. A list of accredited methods is available upon request.Any test data in this report that is derived from test methodswhich deviate from accepted accredited procedure are noted.Accreditation does not constitute endorsement of this report bythe accrediting agency or government.

Standards Council of CanadaAccredited Certification Organization

Conseil canadien des normesOrganisme de certification accrédité

TM

RR,W260,158.0 11/9/01 10:53 AM Page 4

5

INTRODUCTION

In the United States, model buildingcode recognition of allowable designshear loads for shear walls is based onresults of tests using “monotonic” load-ing; e.g., shear loads are applied in theplane of the wall, acting in one directiononly.1,2 Observations of building perfor-mance after hurricanes and earthquakesindicates that this method of evaluatingshear wall performance generally providesa conservative basis for building design,when wood structural panels are used for wall sheathing or siding.

In high wind conditions, building wallsare subjected to in-plane shear loads (aswell as normal uniform loads) which actprimarily in one direction, with fluctua-tions in load due to the response of thestructure to wind gusts. In hurricane ortornado conditions, reversal of shearloads from the opposite direction mayoccur if the building happens to belocated in or near the “eye” of the storm.

In contrast, building walls are subjected toin-plane cyclic (reversed) loads in earth-quake/seismic conditions. For a number ofyears, structural engineers and buildingofficials have requested information on theperformance of shear walls under cyclic

loading, to evaluate whether code-recognized values are appropriate undersuch conditions. However, response tothese requests has been hampered by thelack of consensus on a test protocol forconducting cyclic load tests. Furthermore,until recently, no commercial or industrytesting laboratories in the United Stateshad equipment for conducting such tests,except on a rudimentary basis. Because ofthe expense of the test equipment, only afew university laboratories are equippedfor and capable of conducting cyclic load tests on shear walls.

After the Northridge, California earth-quake in January, 1994, evaluation ofshear wall performance by cyclic loadtests was encouraged by a local buildingcode change adopted by the City of LosAngeles, Department of Building andSafety, in response to recommendationsfrom the Structural Engineers Associationof Southern California (SEAOSC). Thischange reduced the allowable designshear loads by 25%, for shear walls fabri-cated with wood structural panels, untilcyclic load tests are conducted to confirmpreviously recognized values, or anotherappropriate reduction factor is proposed.

SEAOSC formed an Ad Hoc Committeeon Testing Standards for StructuralSystems and Components, to study

appropriate test protocols and evaluationcriteria for cyclic load tests on assembliessuch as shear walls, hold-down connec-tors, fasteners, and anchor connectorsinto concrete and masonry elements.Their deliberations have focused on acyclic load test protocol suggested formasonry construction during earthquakeresearch discussions in 1987 betweenthe United States and Japan.

To address these concerns, and providepreliminary information on shear wallperformance under cyclic load condi-tions, APA – The Engineered WoodAssociation sponsored a limited series ofcyclic load tests on seven 8-foot x 8-footshear walls, fabricated with wood struc-tural panel sheathing attached to woodframing with hand-driven nails. Oneadditional shear wall test, on a matchedconstruction but with wall sheathingattached with pneumatically driven nails,was sponsored by a fastener manufac-turer and results are included in thisreport. The tests were conducted at theStructural Laboratory of the Departmentof Civil and Environmental Engineeringat the University of California - Irvine(Irvine, California), and used the cyclicload test method developed by SEAOSC.

RR,W260,158.0 11/9/01 10:53 AM Page 5

4. Evaluate shear wall performance when wood structural panel sheathing isfastened with pneumatically driven com-mon “short” (diaphragm) nails, popularlyused as an alternate to hand driven common nails.

TEST SPECIMENS

AND MATERIALS

Shear wall test specimens were fabricatedusing an 8-foot x 8-foot frame with con-struction as shown in Figure 1 andTable 1. The wall construction complieswith recent local building code changesadopted by the City of Los Angeles, forshear walls when allowable design shearload exceeds 300 pounds per foot. Theallowable design shear loads for the shearwall test specimens are listed in Table 1,and are based on recognized values listedin Table 23-II-I-1 of the 1997 UniformBuilding Code, and in National EvaluationService, Inc. Report NER-108.

Framing

Lumber used for framing in all tests wasgreen Douglas-fir. At the time of test, thelumber moisture content was in therange of 12 - 18%, as determined by aresistance-type electric moisture meter.The wall sheathing was fastened to theframing within 48 hours of testing, whichminimized any effects of moisture seasoning (drying) on the performanceof the wall.

Intermediate wall studs consisted of 2x4sspaced 16 inches on center, and double2x4s were used for top plates. End studs(or posts) consisted of 4x4s. To conform

with emergency changes adopted in1994 to local building code requirementsfor the City of Los Angeles, 3x4s wereused for the bottom plate, and for thecenter stud where the wall sheathingpanels butt together, both of which arerequired when the allowable design shear load for the wall is higher than300 pounds per foot. A similar require-ment has been incorporated in the 1997Uniform Building Code, when allowableshear values exceed 350 pounds perfoot.a The increased lumber size for thebottom plate was deemed necessary toincrease bending capacity of the bottomplate when the wall is subjected to rack-ing shear and uplift forces from sheathingfasteners, and to increase resistance tosplitting along lines of foundation andhold-down bolts. The bottom plate mustresist both cross-grain bending and ten-sion perpendicular-to-grain forces, devel-oped by sheathing fasteners. For thecenter stud, the 3x4 member is requiredto provide sufficient lumber edge dis-tance for fasteners when the sheathingedge distance for fasteners is increasedfrom 3/8 inch to 1/2 inch, which wasanother building code change adoptedby the City of Los Angeles when theallowable design shear load for the wall is higher than 300 pounds per foot.

The above changes to framing require-ments were based on shear wall failuremodes observed in buildings damaged inthe Northridge, California earthquake. Inmany cases, the bottom plate of shearwalls was notched, or enlarged holeswere cut for installing foundation bolts,

aTable 23-II-I-1, Footnote 3

6

OBJECTIVE

The objectives of these shear walltests are to:

1. Obtain preliminary data on the load-displacement characteristics andload capacity of typical wood-framedwalls with wood structural panel sheath-ing, under cyclic load conditions. Woodstructural panel sheathing included bothplywood and oriented strand board(OSB) in three thicknesses and two panel grades.

2. Study two different cyclic load-displacement test protocols, in testsusing identical wall constructions, todetermine whether results adequatelycharacterize the performance of shearwall assemblies. One protocol, on whicha SEAOSC recommended test method is based, is referred to as the TCCMARsequential phased displacement proce-dure, which was proposed in 1987 by the Joint Technical CoordinatingCommittee on Masonry Research for the United States - Japan CoordinatedEarthquake Research Program. Thisprotocol was used in seven of eight testsin the series. The other protocol, used in one test, consists of “ramped” cyclicdisplacement increments without subse-quent “decay” cycles used in the TCCMAR procedure.

3. Evaluate whether gypsum wallboard,installed on one side of the wall, con-tributes to the stiffness and strength ofshear walls sheathed with wood struc-tural panels on the other side of the wall,under cyclic load conditions.

RR,W260,158.0 11/9/01 10:53 AM Page 6

7

TABLE 1

Description of APA cyclic load shear wall test specimens (tests conducted at University of California - Irvine, April 1995)

Wall Sheathing Sheathing Nail Schedule(1)Shear Wall

Test No. Ext. Int. Ext. Int. Design Load, lb/ft(2)

1A 15/32" Plywood – 10d Com. – 510(Structural I) @ 4" oc

2B 15/32" Plywood – 10 Com. – 510(Structural I) @ 4" oc

3A 15/32" Plywood 1/2" GWB 10d Com. 5d GWB 510(4)

(Structural I) @ 4" oc 7" oc

4A 15/32" OSB#1 – 10d Com. – 510(Structural I) @ 4" oc

5A 15/32" OSB#2 – 10d Com. – 510(Structural I) @4" oc

6A 3/8" Plywood – 8d Com. – 550(Structural I) @ 3" oc

7A 7/16" OSB#3 – 8d Com. – 490@ 3" oc

8A 15/32" OSB#1 – 10d Com., Short(3) – 510(Structural I) @ 4" oc

(1) 10d Com: 0.148" x 3" (head dia. – 0.312") – Tests 1A, 2B, 3A, 4A, 5A8d Com: 0.131 x 2-1/2" (head dia. – 0.281") – Tests 6A, 7A5d GWB (wallboard-smooth shank): 0.092" x 1-5/8" (head dia. – 0.297") – Test 3A

(2) Table 23-II-I-1 of 1997 Uniform Building Code (UBC )and Sec. 2302 – wood structural panel definition, and Sec. 2315.5.3. Also NER-108, Table 2.

(3) 10d Com. “short”: 0.148" x 2-1/8" (head dia. – 0.279") – Test 8A

(4) Does not include shear contribution from gypsum wallboard (GWB) per Sec. 2513.1 of 1997 UBC

FIGURE 1

Cyclic load shear wall test specimen

12" 12"28" 28"16"

Gypsum wallboard joints

Plywood or OSB wall sheathing

4x4 (typ.)

3x4

Hold-downs (typ.)

12" 28" 12"28"16"

2x4 (typ.)16" o.c.

Test wall (96" x 96")2 - 2x4

RR,W260,158.0 11/9/01 10:54 AM Page 7

8

wiring conduit or plumbing. These fieldpractices severely reduce the capacity ofthe bottom plate to resist uplift/bendingforces that develop through the wallsheathing fasteners, when the wallsheathing panels rotate in response tolateral shear forces on the wall. Further, itwas not uncommon to find that sheath-ing fasteners at panel edges missed theframing members, or were driven tooclose to the edges (or ends) of the panels,which diminished the shear resistance ofthe wall and “pull-through” resistance offasteners near panel edges. These defi-ciencies can be detected by inspection ofshear walls, and should be correctedduring construction.

Wall Sheathing

Plywood and oriented strand board(OSB) wood structural panels were usedfor wall sheathing in the eight shear walltests. For six of the tests, 15/32-inch APAStructural I Rated Sheathing 32/16,Exposure 1 was used (5-ply plywood andOSB – three tests each), while for onetest 3/8-inch APA Structural I RatedSheathing 24/0, Exposure 1 plywood(3-ply) was used. In another test,7/16-inch APA Rated Sheathing 24/16,Exposure 1 OSB was used.

Plywood panels were manufactured inaccordance with U.S. Product StandardPS 1-83b, Construction and IndustrialPlywood. The OSB panels were manufac-tured to comply with the performance

requirements of APA Standard PRP-108,Performance Standards and Policies forStructural-Use Panelsc.

In all tests, wood structural panels wereapplied vertically, with their face grain orstrength axis parallel to studs. Paneledges were spaced 1/8 inch apart alongthe vertical joint at the center stud, toconform with APA recommendations forpanel installation, which provides spacefor panels to expand if subjected to mois-ture under on-site field conditions.

In one test, 1/2-inch gypsum wallboardwas installed on one side of the wall, oppo-site the wood structural panel sheathing.The gypsum wallboard was installed hori-zontally, as is typical in practice, with anunblocked horizontal joint at mid-height ofthe wall. The upper panels were installedin two 4-foot x 4-foot panels with a verticaljoint located at the center stud, while thelower panel consisted of a single 4-foot x8-foot panel. The joints between panelswere butted to a light contact. No tape orjoint compound was used over gypsumwallboard joints, to minimize the assemblyand testing schedule.

Fasteners

Wood structural panel sheathing wasfastened to framing with hand driven,U.S.-made 8d and 10d common steelnails for all but one of the tests. Samplesof fasteners were measured and com-pared (Table 2) with standard dimen-sions in Federal Specification FF-N-105,Nails, Brads, Staples and Spikes: Wire,Cut and Wroughtd.

In one test, pneumatically driven 10dcommon “short” (diaphragm) nails wereused. When used to fasten 15/32-inchwood structural panel sheathing, thesenominal 0.148-inch-diameter x2-1/8-inch-long nails provide penetrationinto framing of 1.659 inch, or slightlymore than 1-5/8-inch minimum penetra-tion prescribed in Table 23-II-I-1 of the1997 Uniform Building Code. The Uni-form Building Code requirement is basedon minimum fastener penetration intoframing of approximately 11 times thenail shank diameter (= 11 x 0.148 inch= 1.628 inch penetration). The pneu-matic nailing tool for driving these nailswas equipped with an adjustable deviceto control the depth to which the fas-tener is installed, so that the top of thenail heads were set flush with the surfaceof the panel. For diaphragm (and shearwall) applications, countersinking offasteners below the surface of the panel isnot permitted by Section 2315.1 of the1997 Uniform Building Code.

bNow PS 1-95.

cU.S. Product Standard PS 2-92, PerformanceStandard for Wood-Based Structural-Use Panels,establishes comparable requirements.

dSuperceded by ASTM Standard F1667, StandardSpecification for Driven Fasteners: Nails, Spikes, and Staples.

RR,W260,158.0 11/9/01 10:54 AM Page 8

9

TABLE 2

Description of nails used for attachment of wood structural panel sheathing and gypsum wallboard

Dimensions, in.

Length Shank Diameter Head Diameter

Nail Type Spec.(a) Actual Spec.(a) Actual Spec.(a) Actual (avg.)

8d common 2-1/2 2-1/2 0.131 0.140 0.281 0.267(±1/16) (±0.004) (±10%)

10d common 3 3 0.148 0.150 0.312 0.292(±3/32) (±0.004) (±10%)

10d common short (diaphragm) – 2-1/8 – 0.147 – 0.279

5d gypsum wallboard (smooth) 1-3/4 1-5/8 0.092 0.092 0.375 0.301(±1/16) (±0.004) (±10%)

(a) Federal Specification FF-N-105B (superceded by ASTM Standard F1667)

FIGURE 2

Shear wall hold-down connector details

3/4"

1-1/2"

7"

3-1/2"

A

1-15/16"

3"

1-1/2"

2-1/4"

15/16" hole (for 7/8" bolt)

MC 3 x 7.1

Section A

3/16" weld(top and bottom)

1/2" Steel

6-1/2" above3x4 bottom plate (ref.)

25/32" holes(for 3/4" bolts)

RR,W260,158.0 11/9/01 10:54 AM Page 9

Hold-Down Connectors

Specially designed welded steel hold-downs were used on both 4x4 end poststo prevent overturning and uplift of thewall when subjected to in-plane lateralforces (Figure 2). The hold-downs weredesigned by California structural engineerBen L. Schmid, SE for another series oftests conducted on similar shear walls atthe University of California - Irvine (seeAppendix B), and were re-used in thesetests. The hold-down design incorporatestwo 2-5/8-inch-diameter steel shear plateconnectors, described in Part X of theNational Design Specification for WoodConstruction (NDS)3; and also in ASTMD5933, Standard Specification for 2-5/8 in.and 4-in. Diameter Metal Shear Plates forUse in Wood Constructions. The shearplates distribute bolt loads and minimizebolt deformation or slip in the end posts.Also, 1/4-inch-thick x 2-1/2-inch squaresteel plate washers were placed undernuts used with 3/4-inch-diameter hold-down bolts through the end posts, todistribute bolt loads without deformingthe washer. If standard steel round cutwashers had been used (instead of platewashers), field observations and experi-ence indicates that they could deformand precipitate splitting failures in the end posts.

Hold-down deformation and bolt slipwere minimized by using tight-fittingshear plate connectors, and carefuldrilling of bolt holes in the end postsusing a template and drill guide. Thediameter of the bolt holes was limited toa maximum of 1/16 inch oversize, also tominimize deformation, in accordancewith NDS provisions. The groove and

dap for the shear plate connector wereaccurately prebored in the end postsbefore the wall framing was assembled,using a template and drill guide to accu-rately locate the holes, and a specialcutter head obtained from the manufac-turer of the connector.

Deformation or slip of the hold-downconnectors contributes directly to walldeflection, in proportion to the aspect(height/length) ratio of the wall. It alsogenerates an added uplift component forlateral shear forces for sheathing fastenersalong the bottom and top plates of thewall, which can contribute to splitting ofthe plates from cross-grain tension, andpremature fastener fatigue failures undercyclic load conditions, thereby reducingthe shear capacity of the wall.

TEST SET-UP

AND PROCEDURE

The shear wall test specimens wereattached to the base of the test fixture with four 5/8-inch-diameter anchor bolts,located as shown in Figure 1. The bottomplate of the shear wall was supported on a3-inch-wide steel channel. The edges ofthe channel were inset from the nailingsurfaces of the wall framing, to allow thesheathing panels to rotate when the shearwall is laterally loaded, without bearing onthe test fixture at the bottom corners of thepanels. Bolt holes in the 3x4 bottom platewere carefully located and predrilled usinga template and drill guide, to minimizebolt deformation and slip when the shearwall is laterally loaded in the plane of thewall. The diameter of bolt holes in thebottom plate was limited to a maximum of

1/16 inch oversize, for reasons as notedabove. Steel square plate washers, 1/4 inchthick x 2-1/2 inches square, were usedunder nuts for the bottom plate bolts.

Racking shear loads were applied hori-zontally to the top of the shear wall testspecimens through a steel H-beam lag-screwed (with 3/8-inch-diameter lagscrews spaced 16 inches on center) tothe double top plate of the wall. Thebeam was placed on a 3/4-inch-thick x3-1/4-inch-wide plywood spacer, sand-wiched between the beam and the topplate, which allows the wall sheathingpanels to rotate when the shear wall islaterally loaded, without bearing on thebeam at the top corners of the panels.The beam was restrained laterally (nor-mal to the plane of the wall) by pairs ofTeflon® plastic pads fastened to the beamand to adjacent steel bracing frames, toallow in-plane displacement of the wallduring the test, while preventing the topof the wall from twisting outside of theload application plane.

Racking shear loads were applied with a55,000-pound-capacity programmabledouble-acting hydraulic actuator (MTSSystems Corporation) which was boltedto a stiff vertical cantilever columnhinged at the base to act as a lever arm, as shown in Figure 3. The hingedcantilever column arrangement was necessary to “multiply” the actuatordisplacement, which was limited to±3 inches in either direction, to achievegreater displacement at the top of thewall. The horizontal load at the top of thewall was measured with a hydraulic loadcell in the horizontal plane of the topplate and steel H-beam.

10

RR,W260,158.0 11/9/01 10:54 AM Page 10

11

FIGURE 3

Test setup for cyclic load shear wall tests (University of California-Irvine)

FIGURE 4

TCCMAR sequential phased displacement load-displacement procedure (Protocol A) for shear walls

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75

Cycle Number

Spec

imen

Dis

pla

cem

ent

(% o

f FM

E)

400375350325300275250225200175150125

755025

0–25–50–75

–100–125–150–175–200–225–250–275–300–325–350–375–400

Load cell

Load beam Test wall (96" x 96")

Hinge

Hydraulic actuator(double action)

RR,W260,158.0 11/9/01 10:54 AM Page 11

12

TABLE 3

TCCMAR sequential phased displacement load-displacement procedure (Protocol A) for shear walls

Cycle No. % FME Displ. (in.)

0 0 01 25 0.202 25 0.203 25 0.20

4 50 0.405 50 0.406 50 0.40

7 75 0.608 75 0.609 75 0.60

10 100 0.8011 75 0.6012 50 0.4013 25 0.2014 100 0.8015 100 0.8016 100 0.80

17 125 1.0018 94 0.7519 63 0.5020 31 0.2521 125 1.0022 125 1.0023 125 1.00

24 150 1.2025 113 0.9026 75 0.6027 38 0.3028 150 1.2029 150 1.2030 150 1.20

31 175 1.4032 131 1.0533 88 0.7034 44 0.3535 175 1.4036 175 1.4037 175 1.40


38 200 1.6039 150 1.2040 100 0.8041 50 0.4042 200 1.6043 200 1.6044 200 1.60

45 250 2.0046 188 1.5047 125 1.0048 63 0.5049 250 2.0050 250 2.0051 250 2.00

52 300 2.4053 225 1.8054 150 1.2055 75 0.6056 300 2.4057 300 2.4058 300 2.40

59 350 2.8060 263 2.1061 175 1.4062 88 0.7063 350 2.8064 350 2.8065 350 2.80

66 400 3.2067 300 2.4068 200 1.6069 100 0.8070 400 3.2071 400 3.2072 400 3.20

RR,W260,158.0 11/9/01 10:54 AM Page 12

13

Seven of the eight shear walls (i.e., alltests except Test 2B) were subjected tofully reversing cyclic displacements fol-lowing the TCCMAR sequential phaseddisplacement procedure (Protocol A)described in Figure 4 and Table 3. Thistest procedure has been reviewed andrefined over a 2-1/2-year period, and afinalized version was published in 1997by the Structural Engineers Associationof Southern California (SEAOSC) asStandard Method of Cyclic (Reversed) LoadTest for Shear Resistance of Framed Walls for Buildings.9

The TCCMAR procedure introduces theconcept of the First Major Event (FME),which is defined as the first significantlimit state which occurs during the test.A limit state is an event which marks thedemarcation between two behavior states.When a limit state occurs, some struc-tural behavior of the element or system isaltered. FME occurs when the loadcapacity of the wall, upon recycling ofload to the same wall displacement incre-ment, first drops noticeably from theoriginal load and displacement. However,the effect of FME for wood-framedassemblies is not as pronounced as forrigid concrete or masonry assemblies,where cracking or bond failures aroundreinforcing steel may result in a substan-tial change in the stiffness or strength ofthe assembly. FME can be determinedfrom preliminary cyclic load tests on anidentical test specimen. For the wood-framed shear wall test specimens in this

series, FME for all tests was estimatedfrom prior cyclic load tests to occur at adisplacement of about 0.8 inch.Subsequent analysis of test results for theeleven shear wall constructions describedin this report revealed that FME (e.g.,Yield Limit State) occurred at an averagedisplacement of 0.48 inch (0.5% ofheight) (see discussion on pages 24-26).

The TCCMAR procedure consists ofapplying three cycles of fully reversing load(displacement controlled) at each walldisplacement increment representing 25%,50%, and 75% of FME. Then, wall dis-placement is increased for one load cycleto 100% of FME. Next, “decay” cycles ofdisplacement for one cycle each at 75%,50% and 25% of the maximum displace-ment (100% of FME) are applied, followedby three cycles of displacement at maxi-mum displacement (100% of FME) tostabilize the load-displacement responseof the wall. Then, the next increment ofincreased displacement (125% of FME) isapplied, followed by similar decay andstabilization cycles of loading. This incre-mental cyclic load-displacement sequenceis continued to 150%, 175%, 200%,250%, 300%, 350% and 400% of FME,or until the wall exhibits excessive displace-ment (in these tests, at about 350% to400% of FME), at which point shear loadcapacity is greatly diminished.

The other cyclic load procedure(Protocol B), used for Test 2B only, consistsof applying three cycles of displacement,with increments of 0.2 inch for each cycle(see Figure 5 and Table 4); no decay cycles

of displacement are applied. This proce-dure is similar to one developed by theApplied Technology Council (ATC) forseismic testing of steel structures.7 Theincremental cyclic load-displacementsequence is continued, with the same endpoint conditions as described above for theTCCMAR procedure (Protocol A). Thissimpler “ramped” loading procedure wasincluded as an alternate to the TCCMARprocedure, to evaluate if wall performanceis influenced by the test procedure.

The wall displacements were input to acomputer which controlled the actuatordisplacement. A cyclic frequency of0.5 Hz (e.g., one cycle per 2 seconds)was used to avoid inertial effects of themass of the wall and test fixture hardwareduring cyclic loading, which could affectsystem response to cyclic loading. Duringlater stages of the cyclic loading sequence,when wall displacement exceeded1.5 inches, the cyclic frequency wasslowed to a constant rate of displacement(about 0.25 Hz) to properly control thehydraulic system with the instrumenteddisplacement input.

Instrumentation included measurementof horizontal displacement of the wall atthe top plate, vertical displacement at thebottom end of both end posts relative totest base (uplift and compression), hori-zontal displacement of the bottom platerelative to the test base (lateral in-planesliding), and vertical displacement of thehold-down connectors relative to the endposts (displacement/bolt slip).

RR,W260,158.0 11/9/01 10:54 AM Page 13

14

TABLE 4

“Ramped” cyclic load-displacement procedure (Protocol B) for shear walls

FIGURE 5

“Ramped” cyclic load-displacement procedure (Protocol B) for shear walls


1 25 0.202 25 0.203 25 0.20

4 50 0.405 50 0.406 50 0.40

7 75 0.608 75 0.609 75 0.60

10 100 0.8011 100 0.8012 100 0.80

13 125 1.0014 125 1.0015 125 1.00

16 150 1.2017 150 1.2018 150 1.20


19 175 1.4020 175 1.4021 175 1.40

22 200 1.6023 200 1.6024 200 1.60

25 250 2.0026 250 2.0027 250 2.00

28 300 2.4029 300 2.4030 300 2.40

31 350 2.8032 350 2.8033 350 2.80

34 400 3.2035 400 3.2036 400 3.20

0 5 10 15 20 25 30 35 40

Cycle Number

Spec

imen

Dis

pla

cem

ent

(% o

f FM

E)

400375350325300275250225200175150125

755025

0–25–50–75

–100–125–150–175–200–225–250–275–300–325–350–375–400

RR,W260,158.0 11/9/01 10:54 AM Page 14

15

during lateral loading of the shear walls is summarized in Table 6.

Results of cyclic load tests on three similarlyconstructed shear walls, conducted earlierat the University of California - Irvine forCalifornia structural engineer Ben Schmid,SE, are summarized for comparisonin Appendix B.

In most tests with 10d common nails,the observed failure mode was fastenerfatigue, with nails breaking within thelumber framing about 3/8 inch to 1/2 inchbelow the framing surface. Fastener fatiguefailures started near the corners of thepanels after near-maximum shear loadswere reached (at wall displacements ofabout 2 inches), and progressed furtheralong the top and bottom and verticaledges of the panels, away from the corners,as the load-displacement cycles continued.

When wall sheathing panels are orientedvertically, with their face grain or strengthaxis parallel to wall studs, the highest shearforces on fasteners occur near the corners ofthe panels, which are furthest removedfrom the centroid of rotation of the panel.The 2-inch displacement level was reachedin Cycles 45-51 in the TCCMAR procedure(Protocol A), and in Cycles 25-27 in theramped loading procedure (Protocol B).

For these tests, a 3/4-inch-thick plywoodspacer was placed on the top plate of thewall, and the bottom plate of the wallwas supported on a 3-inch-wide steelchannel, so that panel sheathing panelswere “free” to rotate without restraint(bearing) at their top and bottom cor-ners. This condition duplicates test configurations used by APA when wall bracing and shear wall tests areconducted monotonically in accordance

TEST RESULTS

AND DISCUSSION

Test results are summarized in Table 5,and a typical load-displacement curve,for Test 1A, is shown in Figure 6. Load-displacement curves for the remainingtests are shown in Appendix A. Table 5shows shear loads and shear modulusvalues for wall displacements of0.48 inch (0.5% of wall height, orH/200), 0.96 inch (1% of wall height,H/100), and 1.44 inch (1.5% of wallheight, or H/67), and are based on cycledtest results, recorded in the fourth cycleof displacement. Evaluation could bebased on displacements in other cycles of loading; further comments are pro-vided in the discussion under “DesignAnalysis.” Information on the slip (dis-placement) of the hold-down connectors

TABLE 5Results of cyclic load shear wall tests (APA/UC-Irvine, 1995)

Design At � = 0.48 in.(a) At � = 0.96 in.(a) At � = 1.44 in.(a)Avg. Max. Load,

Test No. Description lb/ft lb � in. Load, lb G1(b) Load, lb G2

(b) Load, lb G3(b) lb (L.F.)

1A 15/32" Str. I Plywood 510 4,080 0.28 6,200 12.9 7,900 8.2 8,300 5.8 10,15010d com. @ 4" o.c. (2.5)

2B 15/32" Str. I Plywood 510 4,080 0.21 7,300 15.2 8,750 9.1 9,300 6.5 10,90010d com. @ 4" o.c. (2.7)


1/2" GWB5d gwb @ 7" o.c.

4A 15/32" Str. I OSB #1 510 4,080 0.24 6,300 13.1 7,050 7.3 6,850 4.8 8,30010d com. @ 4" o.c. (2.0)

5A 15/32" Str. I OSB #2 510 4,080 0.28 6,150 12.8 7,150 7.4 7,450 5.2 8,75010d com. @ 4" o.c. (2.1)


7A 7/16" OSB #3 490 3,920 0.23 6,600 13.8 7,450 7.8 6,800 4.7 8,7008d com. @ 3" o.c. (2.2)

8A 15/32" Str. I OSB #1 510 4,080 0.32 5,300 11.0 6,000 6.3 6,350 4.4 7,70010d com. Short @ 4" o.c. (1.9)

(a) Tabulated data is average of ± cycles. Displacement (∆) and shear loads – after fourth cycle of loading.

(b) lb/in. ÷ 1,000 (= kips/in.)

RR,W260,158.0 11/9/01 10:54 AM Page 15

16

TABLE 6

Hold-down connector slip, �h

Test Shear Wall Avg. �h Shear Wall Load, Avg. �h Shear Wall Load, Avg. �h

No. Design Load, lb(b) in.(a) in./1,000 lb lb(b)(�=0.96 in.) in.(a) in./1,000 lb lb(b) (�=1.44 in.) in.(a) in./1,000 lb

1A 4,080 0.020 0.005 8,000 0.049 0.006 8,600 0.063 0.007

2B 4,080 NR – 8,900 NR – 9,200 NR –

3A 4,080 0.008 0.002 9,100 0.033 0.004 9,000 0.045 0.005

4A 4,080 0.006 0.001 7,100 0.017 0.002 7,000 0.030 0.004

5A 4,080 0.028 0.007 7,500 0.064 0.009 7,400 0.070 0.009

6A 4,400 0.013 0.003 8,900 0.041 0.005 8,900 0.052 0.006

7A 3,920 0.008 0.002 7,400 0.025 0.003 6,800 0.040 0.006

8A 4,080 0.026 0.006 6,200 0.047 0.008 6,400 0.057 0.009

Avg. 0.004 Avg. 0.005 Avg. 0.007

NR = Not Recorded

(a) = Tabulated data is average of ± cycles. Slip (�h) – after cycled loading

(b) = From prior cyclic load shear wall tests on similar test specimens, lateral (shear) load on hold-down connector is approx. 80% of shear wall load (see text)

FIGURE 6

Load-displacement curve for typical cyclic load shear wall test (Test 1A)

Ap

plie

d S

hea

r Lo

ad

, p

ou

nd

s

Displacement, inches

14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

RR,W260,158.0 11/9/01 10:55 AM Page 16

17

research in the United Kingdom hasshown that the application of axial load-ing increases the shear load capacity ofshear walls, since it reduces uplift on endposts and intermediate studs to whichedges of sheathing panels are attached.This minimizes vertical components ofshear forces on fasteners, particularly atthe corners of panels. Section 6.1 ofBritish Standards Institution StandardBS 5268: Part 6, Code of Practice forTimber Frame Walls includes provisions forincreasing the allowable design shear loadof walls based on the magnitude of axialloading that occurs. Axial loading wouldbe especially effective for increasing shearload capacity of shear or bracing wallswhen hold-downs are not incorporatedfor overturning or racking resistance.However, axial loading also can increaseshear capacity of walls where hold-downsare located at ends of the wall. In thelatter case, axial loading provides forresistance against uplift of framing wherehold-downs are omitted at intermediatelocations, such as at vertical edges ofadjacent sheathing panels, and at edges ofwall openings such as windows or doors.The influence of axial loading on theperformance of cyclically loaded shearwalls deserves further study.

Comparison of Calculated vs.Measured Shear Wall Displacement

Displacement of shear walls can be calculated in accordance with the follow-ing formula in Section 23.223 of 1997Uniform Building Code Standard 23-2,and associated tables therein:

� = 8vh3+ vh + 0.75hen + h da

____ ___ _EAb Gt b

where:

� = calculated shear wall displacement, in.

v = maximum shear load due to designloads at the top of the wall, lb per ft

h = shear wall height, ft

E = modulus of elasticity of verticalboundary element (end post or studs atshear wall boundary), lb per sq. in.

A = area of cross section of verticalboundary element, sq. in.

b = shear wall length, ft

G = modulus of rigidity of wall sheathing, lb per sq. in.

t = effective shear thickness of wallsheathing, in. (Note: For plywood wallsheathing, values for G and t are tabu-lated in APA’s Plywood DesignSpecification, Tables 1-3. For orientedstrand board wall sheathing, use Gvtvvalues in APA Technical Note N375,Tables 2.4 and 2.4.1. Values for plywoodwall sheathing also are tabulated inTechnical Note N375.)

en = nail displacement, in. (1997 UBCStandard 23-2, Table 23-2-K)

da = shear wall hold-down displacementincluding fastener slip, in. (from tests ormanufacturer’s information)

For this report, shear wall displacementswere calculated at two applied load levels(allowable design shear load, and1.4 times allowable load) for the construc-tions tested, and compared with shearwall displacements measured in the tests.

One component of shear wall displace-ment is sheathing nail slip (displacement).Since sheathing nail slip values inTable 23-2-K are based on monotonic(one direction) fastener lateral load tests,shear wall displacements measured in theinitial load-displacement application (e.g.,prior to subsequent load-displacementcycling) were used for this comparison.

with ASTM E72.1 Bearing of wall sheath-ing panels at the top and/or bottom endscan increase shear wall stiffness andstrength,4 since such bearing limits thedisplacement of the vertical panel edgesrelative to the framing at the end postsand at the center stud. This improvesperformance by reducing the displace-ment and resulting fatigue of the sheath-ing fasteners along vertical framing nearthe corners of the panels, and along topand bottom plates at the perimeter of thepanel sheathing.

Two of the test specimens (Tests 1A and2B) duplicated wall construction used inpreliminary cyclic load shear wall testsconducted earlier at the University ofCalifornia - Irvine for California structuralengineer Ben Schmid, SE (Appendix B).For Schmid’s tests, the steel loadingbeam on top of the wall was fasteneddirectly to the top plate of the wall. Thepanel sheathing was permitted to bear onthe steel loading beam at the top of thewall, but not on the test jig at the bottomof the wall. This condition restrictedpanel displacement along vertical paneledges at end posts and the center stud,as the panels rotated under applied lat-eral shear loading. This detail led tohigher maximum strength and load fac-tors (Table B1) than achieved in theseries of cyclic load tests conducted forAPA. The data from Schmid’s tests isindicative of performance when the panelsheathing is partially restrained by bear-ing at one end of the panels. Even betterwall stiffness and strength are anticipatedif both the top and bottom ends of wallsheathing panels are permitted to bear on other panels, floor surfaces orceiling/roof framing.

The cyclic load tests were conductedwithout axial loading on the walls. Past

RR,W260,158.0 11/9/01 10:55 AM Page 17

18

Table 7 summarizes calculated vs. measured shear wall displacement for the two load levels. Agreement betweenthese values generally was within 1/8 in.,and 13 of 22 values were within 1/16 in.Only 4 of 22 values differed more than1/8 in. (the maximum difference was0.15 in.). In two tests (Test Nos. 5A and8A), the applied shear load of 1.4 timesthe allowable load, was approximatelyequal to or greater than the Yield LimitState (see discussion on page 24), yet thecalculated and measured displacementswere in close agreement. In the remainingnine tests, the applied shear load of 1.4times the allowable load, was less thanthe Yield Limit State. These comparisonssupport the validity of the shear walldisplacement formulas, for shear loads upto at least 40% greater than the allowable load.

Load-Displacement Relationships

Allowable design shear loads for the testedconstructions are based on the 1997Uniform Building Code and NER-108. Atthe allowable design shear load, measuredwall displacement averaged 0.21 inch(range of 0.14 to 0.34 inch) for eleven testspecimens (Table 7). Also, load-displace-ment curves were nearly linear, and hys-teresis loops were repeatable for all testspecimens after four cycles of repeatedloading in this range (Figure 6 andAppendix A and B), demonstrating thatthe shear load capacity of the walls wasundiminished. Within the allowabledesign shear load range, wall displace-ments were consistent with observedresults in monotonic tests, in whichvertical steel tiedown rods are used toresist overturning forces, when walls aretested in accordance with ASTM E72.1,4

At a wall displacement of 0.48 inch,equivalent to 0.5% of height (H/200) forthe 8-foot wall height used in these tests,the load-displacement curves and hys-teresis loops continued to be nearlylinear and repeatable for all test speci-mens (Figure 6 and Appendix A). Afterthree cycles of repeated loading to thisdisplacement, the shear load capacity of the walls diminished only slightly. The shear loads (pounds per foot) at0.48-inch displacement averaged1.7 times the allowable design shear loadlisted in the code, for Tests 1A through7A (range 1.5 to 1.9), and 1.9 for Test 3A

with gypsum wallboard in combinationwith plywood sheathing. For Test 8A withpneumatically driven fasteners, the valuewas 1.3 times the allowable design shearload. Overall, the maximum shearstrength of the walls averaged about1.4 times the shear load at 0.48-inchdisplacement. In general, these compar-isons confirm the conservative nature ofthe allowable design shear load.

The next wall displacement of interest is0.96 inch, or 1% of height (H/100). Atthis displacement, the load-displacementcurves are non-linear, but the hysteresisloops continued to be repeatable for all

TABLE 7

Comparison of calculated vs. measured shear wall displacement

Applied Shear Avg. �Test Load � Calc., Meas., Difference,No. lb/ft(a) lb(b) in. in.(c) in.

1A 510 4,080 0.21 0.22 +0.01714 5,712 0.39 0.38 – 0.01

2B 510 4,080 0.21 0.15 – 0.06714 5,712 0.39 0.25 – 0.14

3A(d) 510 4,080 0.21 0.14 – 0.07714 5,712 0.39 0.24 – 0.15

4A 510 4,080 0.18 0.16 – 0.02714 5,712 0.34 0.27 – 0.07

5A 510 4,080 0.18 0.20 +0.02714 5,712(f) 0.34 0.34 0

6A 550 4,400 0.25 0.23 – 0.02770 6,160 0.42 0.37 – 0.05

7A 490 3,920 0.15 0.15 0686 5,488 0.26 0.27 +0.01

8A 510 4,080 0.18 0.22 +0.04714 5,712(g) 0.34 0.45 +0.11

BLS.1A(e) 510 4,080 0.21 0.31 +0.10714 5,712 0.39 0.48 +0.09

BLS.1B(e) 510 4,080 0.21 0.34 +0.13714 5,712 0.39 0.54 +0.15

BLS.2A(e) 550 4,400 0.21 0.22 +0.01770 6,160 0.37 0.35 – 0.02

(a)First value listed for each test is code-recognized allowable design shear load for tested construction (1997 UBC Table 23–II–I–1). Second value is 1.4 times allowable load.(b)Corresponding applied shear load for 8 ft. long wall segment as tested(c)Displacement measured during initial load-displacement application; average of +/– cycles(d)Gypsum wallboard installed on opposite side of wall(e)See Appendix B for details and discussion(f)Approximate Yield Limit State (YLS) observed in test (FYLS = 5,700 lb @ �YLS = 0.35 in.). See text for discussion.(g)Exceeds Yield Limit State observed in test (FYLS = 5,100 lb @ �YLS = 0.36 in.).

RR,W260,158.0 11/9/01 10:55 AM Page 18

19

R = Numerical coefficient representativeof overstrength and ductility of lateral-force-resisting system (Table 16-N of the1997 Uniform Building Code).

�S = Design Level ResponseDisplacement that occurs when thestructure is subjected to design seismicforces; based on static, elastic analysis ofthe structural system.

The R factor is 5.5 for bearing wall systems with light-framed walls sheathedwith wood structural panels, when build-ing heights are three stories or less; or 4.5for all other light-framed wall systems.Accordingly, maximum inelastic displace-ment calculated by the above formula is3.15 to 3.85 times the wall displacementat allowable design load (static force). Inthe summary of test results in Table 7,calculated wall displacement at allowabledesign load averaged 0.20 in. (measureddisplacement averaged 0.21 in.). Accord-ingly, the maximum inelastic wall dis-placement would be 0.63 in. to 0.77 in.For comparison, shear wall displacementat Yield Limit State in the eleven tests wasin the range of 0.34 in. to 0.76 in. (see Table 8).

Section 1630.10.2 of the 1997 UniformBuilding Code limits maximum storydrift (wall displacement) to 2% to 2.5%of height. For an 8-foot wall height asused in these tests, maximum story driftwould be limited to 1.92 in. (2%) or2.40 in. (2.5%). All walls described inthis report attained maximum shearcapacity (Strength Limit State) beforereaching 2.5% story drift (see Table 8).

However, in discussions of the StructuralEngineers Association of SouthernCalifornia Ad Hoc Committee for TestingStandards for Structural Systems andComponents, a maximum story drift limitof 1% of height (H/100) was suggested to limit structural damage.

For brittle exterior finishes such as stuccoor brick veneer, or interior finishes such asgypsum wallboard, it is recognized thatwall displacement, the magnitude ofwhich is undetermined at present, mayresult in minor cosmetic cracking of thefinish around window and door openingsin shear wall segments, during an earth-quake. When brittle wall finishes are used,a reduced story drift limit, such as 0.5% ofheight (H/200), may be necessary to limitdamage and repair costs.

Shear Wall Load Capacity – Cyclic vs. Monotonic Tests

The shear strength of cyclically loadedwalls, in these tests, was lower than typi-cally obtained in monotonic tests.5 Theload factor (average maximum shearload/allowable design shear load) averaged2.4 (range 2.0 to 2.7) for seven tests withplywood and OSB sheathing fastened withhand-driven 8d or 10d common nails. Aload factor such as 2.5 may be appropriatewhen performance of seismic-resistingassemblies is evaluated on the basis ofcyclic (reversed) load tests.

Since matching monotonic tests were notconducted on these shear wall construc-tions, it is not possible to evaluate thepotential reduction in shear stiffness ormaximum shear capacity that mightoccur under repeated cyclic (reversed)loading. However, when the maximumcycled shear loads (at 1.44 inches of dis-placement) are compared with the averagemaximum shear loads of the walls(Table 5), these preliminary results indicatethat the shear strength for repeated cyclicloading averages about 18% lower than themaximum shear strength reached duringthese cyclic load tests.

In Yasumura’s tests,6 the constructiondetails used in fabricating test specimensare unknown. However, Yasumura reported

test specimens. However, the maximumcycled shear strength of the walls “flat-tened out” at nearly the same shear loadreached at 0.96-inch displacement, whensubjected to repeated cycles of loadingup to a displacement of 1.6 inches. This“expected maximum shear strength”provides a basis for evaluating the perfor-mance of shear walls subjected torepeated cyclic loading, in terms of stiffness and strength.

Examination of the load-displacementcurves for all test specimens (Appendix A)revealed that the hysteresis loops continuedto be repeatable up to displacement in therange of 1.4-1.6 inches, at which point themaximum shear strength (e.g., strengthlimit state) was reached. Beyond this dis-placement, yielding and resulting fatiguefailures of fasteners resulted in decreasingshear load capacity at each repeat cycle ofdisplacement. The hysteresis loops werenot repeatable (at least for three replicationsof loading), and the shear load was notstabilized upon repeated loadings at eachdisplacement increment. It is uncertainwhether additional replications of loadingwould stabilize the hysteresis loops, orinstead cause earlier fatigue failures offasteners due to the large displacementswhich occur at these stages.

The 1997 Uniform Building Code(Sections 1630.9.1 and 1630.9.2) pro-vides a basis for establishing a maximumvalue for inelastic displacement of struc-tural systems such as shear walls. Walldisplacement is derived from calculatedstatic force displacement, which isincreased by factors related to over-strength and ductility capacity of the assembly as follows:

�M = 0.7 R �S

where:

�M = Maximum Inelastic ResponseDisplacement that occurs when the struc-ture is subjected to design ground motion.

RR,W260,158.0 11/9/01 10:55 AM Page 19

20

only a 10% reduction in shear wall stiffnessand strength, when results of cyclic testswere compared to monotonic tests.

In 1994, the City of Los AngelesDepartment of Building and Safety imple-mented an emergency code change call-ing for 25% reduction in code-recognizedallowable design shear loads for shearwalls, since these values were based onmonotonic tests. This emergency codechange was later finalized in 1995. Basedon the preliminary test results in thisseries, this interim measure appears to beconservative, for seismic design of build-ings, until more comprehensive monot-onic and cyclic load testing of matchedtest specimens is conducted to confirmor modify this adjustment.

The City of Los Angeles also implemented acode change to limit the allowable designshear load to 200 pounds per foot, for shearwalls constructed with 3-ply plywoodsheathing. This limitation applies toplywood of 5/8-inch thickness or less, whenmanufactured in accordance with provisionsof U.S. Product Standard PS 1-95. Test 6A,with 3/8-inch Structural I plywood (3-ply)had an allowable design shear load of550 pounds per foot. The load factor forTest 6A was 2.3, which was consistent withresults of other cyclic load shear wall testswith 15/32-inch 5-ply plywood. Thus, thelimitation on allowable design shear load for3-ply plywood sheathing does not appear tobe warranted, based on the results of thispreliminary test. It was observed that a

greater proportion of 8d common nails,used to fasten the 3/8-inch wall sheathing inTest 6A, pulled through the sheathing andremained in the framing. This contrasts withother tests with 15/32-inch sheathing fas-tened with 10d common nails, wherefatigue failure of the nails was typical.

Plywood vs. OSB Wall Sheathing

The load factor for four walls tested withplywood sheathing averaged 2.5 (range2.3 to 2.7). For three walls tested withOSB sheathing (excluding Test 8A withpneumatically driven nails), wall displace-ment within the design shear load rangewas consistent with values obtained intests with plywood sheathing. However,

FIGURE 7

Load-displacement curve for typical cyclic load shear wall test with pneumatically driven 10d common “short” diaphragm nails (Test 8A)

Ap

plie

d S

hea

r Lo

ad

, p

ou

nd

s


14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000

–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

RR,W260,158.0 11/9/01 10:55 AM Page 20

21

than a matching wall constructed withfull-length, hand driven nails (Test 4A,load factor 2.0). A greater proportion ofthe shorter pneumatically driven nailswithdrew from the framing, but only afew fractured in fatigue as was typical ofhand driven nails. The load-displacementcurve in Figure 7 shows that the walldisplacement continued to increase up to2 inches without a significant reductionin cycled shear strength. Thus, theenergy dissipation of the shear wall withsheathing fastened with shorter nails was greater than in Test 4A (compareFigures A4 vs. A8 in Appendix A). At thepossible expense of somewhat lowershear strength, a potential increase inenergy dissipation might be an advantageunder certain situations. However, con-sideration also would have to be given tothe amount of wall displacement that canbe tolerated without resulting in struc-tural damage that would be uneconomi-cal to repair. Further study is needed todetermine whether minimum fastenerpenetration permitted by the NDS3 hasan effect on lateral load capacity of fas-teners when loaded cyclically.

TCCMAR vs. “Ramped” CyclicLoading Procedure

Seven of the eight tests (i.e., all testsexcept 2B) used the TCCMAR sequentialphased displacement test procedure. Thisprocedure has been used by SEAOSC fordevelopment of a cyclic load test methodfor shear walls, and also for a proposedASTM fastener cyclic load test procedure.Incremental increases in load/displace-ment during cyclic loading are followedby a “decay” cycle in which cyclic loadsare applied to reduced displacementlevels before increasing load/displacementto the next higher increment.

For Test 2B, a “ramped” cyclic loadingprocedure was used, in which incremen-tal increases in load/displacement arerepeated for three cycles at each incre-ment, before increasing load/displace-ment to the next higher increment; nodecay cycle is used.7 The results ofTests 1A (TCCMAR procedure) and 2B(ramped procedure), with matched testspecimens, were in good agreement(compare Figures A1 vs. A2 in Appen-dix A). Thus, the ramped procedureappears to offer an alternate, simplerload/displacement procedure for con-ducting cyclic loading tests. However,some researchers or structural engineersmay need complete load-displacementhistory information from the TCCMARprocedure, to permit refined dynamic orfinite element analysis of buildings forearthquake analysis. Further, the TCC-MAR procedure has established prece-dent for cyclic load testing of assembliesand fasteners. Since the results obtainedby the two methods were similar, thereappears to be no substantive reason atthis time to suggest consideration of anycyclic load procedure other than theTCCMAR procedure.

Shear Wall Hold-Down Slip

As noted previously under “Hold-downConnectors,” shear wall hold-down slipshould be minimized. This minimizesdisplacement of the shear wall, and over-stressing of fasteners attaching sheathing toframing, especially in the critical areas nearthe corners of the sheathing panels. Inthese tests, a special welded steel platehold-down connector (Figure 2) was usedin conjunction with shear plates, to mini-mize eccentricity of the hold-down bolt anddistribute lateral shear forces from the boltsinto the end posts with minimum slip.

the load factor at maximum shear loadaveraged 2.1 (range 2.0 to 2.2), or about15% lower than plywood (2.1/2.5 =0.84). There seemed to be earlier andmore extensive fastener fatigue failures inthe cyclic load shear wall tests with OSBthan with plywood. This may be due tohigher density and perhaps higher dowelbearing capacity for OSB than plywood. Itis theorized that denser OSB sheathingdeforms less along the nail shank in con-tact with the sheathing, creating a “fixedend” condition on the fastener. As a result,higher internal bending stresses in the nailshank may develop at locations below thesurface of the wood framing, thus causingearlier fatigue failures of the fastenersunder cyclic loading. Further study ofthis possibility is needed, however.

Contribution of Gypsum Wallboard

Two tests (Tests 1A and 3A) were identical except that 1/2-inch gypsumwallboard was installed on one side ofthe wall in Test 3A. The results were asanticipated; the gypsum wallboard addedto early stiffness but not to shear strengthof the wall. This would seem to be appro-priate for reversed cyclic loading as mightoccur during an earthquake. For windloading, where cyclic (reversed) loading isunlikely, the resistance of the two materi-als appears to be additive, as shown inother monotonic tests.8

Pneumatically Driven Nails

For Test 8A, 15/32-inch Structural IOSB sheathing was fastened with pneu-matically driven 10d common short(diaphragm) nails. The nails providedslightly more than 1-5/8 in. penetration(about 11 diameters) into the framing.The load factor for this wall was 1.9, witha maximum shear load of about 5% less

RR,W260,158.0 11/9/01 10:55 AM Page 21

22

Cyclic load shear wall tests conducted for Schmid indicated that the measureduplift force on the hold-down bolts wasabout 20% less than ordinarily calculated(corresponding hold-down slip was notrecorded) – see Figure B4 in Appendix B.This is believed to be due to the addi-tional overturning resistance provided bythe fasteners which attach wood struc-tural panel sheathing to the top andbottom plates of the wall; such resistanceis not normally taken into account by theshear wall analysis procedure. The differ-ence between the calculated and actualhold-down forces depends on the lengthof the shear wall, the distance betweenhold-down bolts, and the shear capacityof the wall as determined by the sheath-ing type, thickness and grade, framingspecies, and sheathing fastener schedule.

In the cyclic load shear wall testsconducted for APA, uplift force on thehold-down bolts was not recorded, buthold-down slip, relative to the end posts,was measured (data for Test 2B was notrecorded due to a malfunction of thedata acquisition system). The hold-downslip is shown in Table 6. At the allowabledesign shear load for the walls, hold-down slip averaged only 0.016 inch; andat the expected maximum shear strengthof the walls (at 0.96-inch displacement),only 0.039 inch (range 0.017 to0.064 inch) after load cycling.

Corresponding hold-down slip, based on applied lateral load (approximatelyequivalent to hold-down uplift force),averaged 0.004 in. per 1,000 lb at shearwall design load, and slightly greater athigher loads (see Table 6).

DESIGN AND

CONSTRUCTION

CONSIDERATIONS

When designing shear walls, a number offactors should be considered in additionto the typical specifications based onsheathing thickness and grade, framingspecies and size, and fastener type andschedule which determine the allowabledesign shear load for the wall.

If multiple studs are used in lieu of solidone-piece wood framing at panel edges,the studs must be fastened together totransfer the shear forces that the shearwall is designed for. This may call forclosely spaced face nailing of studs with10d or 16d common nails, or perhapsbolts or lag screws. When multiple studsare used, sheathing fastening should beapportioned between these members.

When walls are subjected to rackingshear forces, the end posts act alternatelyas tension and compression “chords” of the vertical diaphragm (shear wall).When the end posts are in compression,the bearing capacity of the end posts onthe top and bottom plates of the wallshould be checked. This design checkshould consider not only axial designload from tributary dead and live loadson the structure, but also the additivecomponent of chord forces at the maxi-mum shear strength for the shear wall.The maximum shear strength could be asmuch as 2.5 or more times the allowabledesign shear load for the assembly. If thebearing stress is exceeded, larger or mul-tiple end posts are needed to minimize orprevent crushing of the top and bottomplates at the expected maximum shearstrength. If such crushing occurs, it canhave an adverse effect on the displace-ment or expected maximum shearstrength of the assembly.

In situations where the shear wall issupported on a concrete slab floor, it maybe possible to detail the shear wall sothat the bottom end of end posts bearsdirectly on the concrete instead of onwood, to utilize higher lumber parallel-to-grain compression stresses andincrease the bearing capacity of the endposts. However, for wood-framed walls,the upper end of end posts typicallybears on the wood top plate of the wall.In this situation, the bearing capacity ofthe end posts would be governed by theallowable stress for compression perpen-dicular to grain for the end post bearingon the top plate.

The net section of the end post at thehold-down connector bolt holes (ifapplicable) should be checked to insurethat the end posts have sufficient tensioncapacity to resist the uplift loads imposedat this location. If shear plate connectors(Part X of NDS3) are used in conjunctionwith bolts for attaching hold-down con-nectors to the end posts, the net sectionof the end post is further reduced by thedapped slot cut into the end post toaccommodate the shear plate connector.

In situations where the hold-down bolt isembedded near the edge of a concreteslab floor or foundation wall, the amountof embedment and uplift capacity of thehold-down bolt in the concrete shouldbe checked in accordance with buildingcode design provisions. If the “shearcone” from the bolt intersects the side ofthe slab or foundation wall, the upliftresistance of the bolt is reduced. Themass of the tributary area of the slab,foundation, footing and overlying earthbackfill (if applicable) must be sufficientto resist the uplift load acting on thehold-down bolts.

RR,W260,158.0 11/9/01 10:55 AM Page 22

23

Shear wall displacement should bechecked for conformance with modelbuilding code provisions for maximuminelastic displacement and story drift.Calculation of shear wall displacement isdescribed on pages 17-18 of this report.In the future, it may be possible to use ashear modulus (G), as described inASTM E5642 and derived from cyclicload tests, for calculating shear wall dis-placement. For further information, seethe following discussion under“Design Analysis.”

During fabrication, hold-down deforma-tion and fastener slip should be mini-mized, to avoid additional shear forces onsheathing fasteners, and to minimize thepotential for splitting the horizontal topand bottom plates when sheathing panelsrotate or uplift due to racking (shear)forces acting on the wall. When bolts areused to attach hold-down connectors toend posts, the bolt holes should be care-fully located and drilled, preferably with atemplate and drill guide. A maximum of1/16-inch oversize hole is specified by theNDS.3 Installation of shear plate connec-tors for hold-down bolts in end posts, asmentioned previously, will reduce slip ofthese connections when subjected tohigh lateral (uplift) forces. Cyclic loadshear wall tests (in progress by others)indicate that hold-down connectors thatare fastened to end posts with nailed orlag screw connections can minimize walldisplacement due to hold-down slip.

These tests also emphasized the need for carefully locating and installing hold-down bolts in foundations or slabs, or atfloor intersections. Structural or architec-tural drawings should include detailsshowing the type of hold-downs requiredand dimensions locating where hold-

down connectors are to be installed. Foron-site construction, jigs or templates aresuggested for accurately locating anddrilling holes for hold-down bolts, andfor bolts to connect the bottom plate ofshear walls to slabs and foundation walls.

Hold-downs may extend through oracross floor intersections. Dry lumber,with a moisture content of 19% or less,should be specified for floor framingmembers located beneath shear walls, tominimize lumber shrinkage which couldaffect the effectiveness (e.g., displacement)of the hold-down. For wood in protectedapplications, 9% to 12% moisture contentis the usual in-service range. Dry woodframing products for floor rim joists, suchas APA Rim Board,® APA PRI® woodI-joists and blocking panels, or structuralcomposite lumber, are now available fromseveral engineered wood product manu-facturers in the United States and Canada.

DESIGN ANALYSIS

Load-displacement curves for all testspecimens have similar shape charac-teristics (see Appendix A). A potentialmethod of analyzing shear wall perfor-mance in terms of stiffness and shearstrength combines features of a designapproach suggested by John Kariotis, SE(Kariotis & Associates, Sierra Madre, CA).It is based on the static force procedurefor seismic design of structures in accor-dance with Section 1630.2 of the 1997Uniform Building Code.

Fundamental to this method is thedetermination of a shear modulus (G')for the wall. The shear modulus for shearwalls also is evaluated in ASTM E564.2 Inthat standard, the shear modulus isdetermined at one shear load value (33%

of the maximum shear strength of thewall). The shear modulus can be derivedfrom the following formula:

�= (F/G')(H/L) [1]

where �= Displacement, in.

F = Applied load, lb

G'= Shear modulus, lb/in.

H = Height of wall, ft

L = Length (width) of wall, ft

Re-arranging terms, G' = (F/�)(H/L),which is the slope of the load-displacementcurve or wall stiffness, reported in terms ofpounds per inch (or kips per inch, where1 kip = 1,000 pounds) at any desiredshear load. An average (F/�) value for cyclicstiffness can be determined for any desiredshear load or displacement increment bydetermining the slope of a line drawnbetween the maximum positive and nega-tive values of load or displacement on initialloading or stabilized hysteresis loops. Sincethe load-displacement curve is not linear,except at low shear loads, the wall stiffnesschanges continuously. Therefore, to accu-rately characterize the performance of theshear wall, G' must be determined for key wall displacements of interest.

From the hysteresis loops recorded duringthe cyclic load tests, an envelope or “back-bone” curve can be constructed throughthe maximum shear load recorded at eachdisplacement increment (Figure 8 forTest 1A). This curve can be used to checkthe initial shear stiffness and the maximumshear strength of the wall; the latter valuealso can be used to check the strength ofthe hold-down connections for the wall.Similarly, a “cycled” backbone curve can beconstructed through the cycled shear loadrecorded for the stabilized hysteresis loopsat each displacement increment (in the

RR,W260,158.0 11/9/01 10:55 AM Page 23

24

fourth cycle or any other desired cycle afterthe initial cycle), as shown in Figure 8. Thecycled backbone curve represents the shearstiffness and expected maximum shearstrength of the shear wall after having beensubjected to repeated cyclic (reversed)displacements such as typically occurduring an earthquake. It is emphasized thatthe many repeated cycles of load-displace-ment in the TCCMAR (SEAOSC) testmethod, especially beyond the allowabledesign load and Yield Limit State (FME),provide a very conservative basis for evalu-ating shear wall stiffness and maximumshear strength. In effect, this protocolevaluates a building’s shear wall perfor-mance when subjected to numerous cyclesof reversed loading, not only for “design

level” but also higher magnitude earth-quakes over the life of the building. Anaccurate representation of shear wall per-formance requires development of back-bone curves for both positive and negativedisplacements, and for both initial andcycled backbone curves, using the averagevalues from these curves for analysis.

The shape of the backbone curve can beapproximated by bilinear segments, suchas shown in Figure 9. Key wall displace-ments occur at Yield Limit State (YLS) andStrength Limit State (SLS). The resultingbilinear load-displacement curve repre-sents a reasonable approximation ofshear wall displacement and shearstrength from initial loading to YLS, and from YLS to SLS.

The following terms have been proposedby SEAOSC to describe various character-istics of the bilinear load-displacementcurve, which are important in evaluatingdesign of shear walls.9

Yield Limit State (YLS): The point inthe force-displacement relationshipwhere the difference in forces betweenthe first and fourth cycle, at the samedisplacement, does not exceed 5%.

Strength Limit State (SLS): The pointin the force-displacement relationshipcorresponding to maximum displace-ment for the peak force attained by theelement or system.

�yls, Fyls: Displacement (�yls) and shearload (Fyls) at Yield Limit State (FME).

�sls, Fsls: Displacement (�sls) and shearload (Fsls) at Strength Limit State (maxi-mum shear capacity).

FIGURE 8

“Backbone curve” and “cycled backbone curve” for cyclic load-displacement shear wall (Test 1A)

Ap

plie

d S

hea

r Lo

ad

, p

ou

nd

s


–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

Allowable shear load (UBC)

Yield Limit State (FME)

14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000

Strength Limit State(maximum shear strength)

“Cycled”backbonecurve

Backbonecurve

RR,W260,158.0 11/9/01 10:55 AM Page 24

25

Overstrength Factor (Ro): Ro = Fsls/Fyls

Ductility Factor (Rd)10,11: Rd = µ =�sls/�yls (for T>Ts); or (2µ – 1)0.5

(for T<Ts)

R Factor: Numerical coefficient repre-sentative of the inherent overstrength andglobal ductility capacity of the lateral-force-resisting system (e.g., shear wall);per Section 1628 of the 1997 UniformBuilding Code.

Data for these factors, calculated fromeight tests conducted for APA and threetests conducted for Schmid (AppendixB), are summarized in Tables 8 and 9.Data for effective stiffnesses (Ke = F/�)for the eleven walls also are shown inTable 9, for information. The effectivestiffnesses are applicable to the testedwalls which had an aspect (height/length) ratio of 1.0.

The following procedure has been suggested for determining the requiredlength of shear wall(s), based on shearstrength and displacement, in accor-dance with the Static Force Proceduredescribed in Section 1630.2 of the 1997Uniform Building Code (UBC). In thefollowing discussion, all references to theUBC are based on the 1997 edition ofthe code. Total design base shear is deter-mined from Formula (30-4) in Section1630.2.1 of the UBC:

V = (Cv/T)(I)(W) / R [2]= (2.5Ca)(I)(W) / R; maximum Vwhen Cv/T = 2.5 Ca; or [2a]= (3.0Ca) (W) / R for simplified struc-tures (for example, single-familydwellings) per Section 1630.2.3 of the UBC. [2b]

where:

V = Total design base shear.

Cv, Ca = Seismic Coefficients based onSeismic Zone Factor Z ( = 0.075 to 0.4),Soil Profile Type S, and Near-SourceFactor N ( = 1.0 to 2.0).

I = Building importance factor ( =1.0 for single-family dwellings).

W = Total seismic dead load ofstructure, plus other applicable loads perSection 1630.1.1 of the UBC.

R = Coefficient representative ofoverstrength and ductility of structure.See discussion on page 26 and Table 16-N of the UBC for code-assigned val-ues ( = 4.5 to 5.5 for buildings withshear panels on light-frame walls). Foradditional background information, see References 10 and 11.

FIGURE 9

Bilinear load-displacement relationship for shear wall Test 1A

–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2


Ap

plie

d S

hea

r Lo

ad

, p

ou

nd

s

Strength LImit State

Yield LimitState (FME)

Bilinear Segments

Bilinear Segments 14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000

RR,W260,158.0 11/9/01 10:55 AM Page 25

26

T = Elastic fundamental period ofvibration (seconds) of the structure.

To determine the required length of shearwall(s) based on shear strength, the firststep is to calculate a fundamental periodof vibration (T) for the structure, whichcan be estimated from Method A andFormula (30-8) in Section 1630.2.2.1 ofthe UBC:

T = (0.020)(hn)0.75, where hn is theheight (ft) above the base to level n. [3]

Next, a spectral response modificationfactor for peak ground acceleration (cor-responding to Cv/T) can be determinedper Figure 16-3 of the UBC, based onthe estimated elastic fundamental periodof vibration.e

The design shear load for the wall (lb orkips) is then calculated from Formula [2],by multiplying the assigned seismic load(lb or kips) by the spectral response mod-ification factor, and dividing by R.

Finally, the length of shear wall(s) (ft)required for adequate strength is deter-mined as follows. (a) Divide the designshear load (V) by the unit shear load atYield Limit State (vyls). (b) Divide thedesign shear load (V) by the product ofthe Resistance Factor φ (sometimesreferred to as the Capacity ReductionFactor) for Load and Resistance Factor(LRFD) design, and the unit shear load atStrength Limit State. For shear walls andconnections, φ = 0.65 in accordancewith ASCE 16-95.11 The greater of theshear wall lengths determined by (a) or(b) determines the length of wallrequired for strength.

eThe maximum value of the spectral responsemodification factor (Cv/T) is 2.5Ca, per Figure 16-3and Sec. 1630.2.1 of the 1997 Uniform BuildingCode. Therefore, it is conservative to use the maxi-mum value of 2.5 for the spectral response modifica-tion factor. For example, in Seismic Zone 4 with Ca= 0.4, the spectral response modification factorwould be (2.5)(0.4) = 1.0. For a more exact solu-tion, use values of Cv, Ca as determined from Tables16-Q, 16-R, 16-S and 16-T of the 1997 UniformBuilding Code for applicable soil profile type.

TABLE 8

Calculated R factors (Ro-overstrength and Rd-ductility) from wood structural panel shearwall tests(a)

Test Fyls �yls Fsls �sls Ro Rd-L = µ Rd-SNo. (lb) (in.) (lb) (in.) =(2µ-1)0.5

1A 6,950 0.51 10,150 2.00 1.5 3.9 2.6

2B 7,900 0.47 10,900 1.99 1.4 4.2 2.7

3A 8,250 0.47 10,600 1.74 1.3 3.7 2.5

4A 6,200 0.34 8,300 1.16 1.3 3.4 2.4

5A 5,700 0.35 8,750 1.82 1.5 5.2 3.1

6A 7,700 0.51 10,300 2.02 1.3 4.0 2.6

7A 6,150 0.35 8,700 1.16 1.4 3.3 2.4

8A 5,100 0.36 7,700 2.06 1.5 5.7 3.2

BLS.1A 6,050 0.55 11,100 2.33 1.8 4.2 2.7

BLS.1B 7,150 0.76 10,900 2.32 1.5 3.1 2.3

BLS.2A 8,550 0.62 12,400 1.80 1.5 2.9 2.2

(a)Based on average bilinear load-displacement relationship determined from Figures A1-A8 and B1-B3. Fordescription of wall test specimens, see Tables 1 and B1.

= �sls____�yls

= Fsls____Fyls

TABLE 9

Summary of unit shear resistance at yield limit state (vyls), 1% story drift (v0.01H) andstrength limit state (vsls), and corresponding effective stiffnesses (Ke=F/�) from woodstructural panel shear wall tests(a)

Yield Limit State (yls) 1% Story Drift (H/100) Strength Limit State (sls)

Test vyls Ke-yls v0.01H Ke-0.01H vsls Ke-slsNo. (Kip/ft) (Kip/in.) (Kip/ft) (Kip/in.) (Kip/ft) (Kip/in.)

1A 0.87 13.6 0.99 8.3 1.27 5.1

2B 0.99 16.8 1.11 9.3 1.36 5.5

3A 1.03 17.6 1.15 9.6 1.33 6.1

4A 0.78 18.2 0.98 8.2 1.04 7.2

5A 0.71 16.3 0.88 7.3 1.09 4.8

6A 0.96 15.1 1.06 8.9 1.29 5.1

7A 0.77 17.6 1.01 8.4 1.09 7.5

8A 0.64 14.2 0.75 6.3 0.96 3.7

BLS.1A 0.76 11.0 0.91 7.6 1.39 4.8

BLS.1B 0.89 9.4 0.96 8.0 1.36 4.7

BLS.2A 1.07 13.8 1.22 10.2 1.55 6.9

(a)Based on average bilinear load-displacement relationship determined from Figures A1-A8 and B1-B3, andTables 5 and B1. For description of wall test specimens, see Tables 1 and B1.

RR,W260,158.0 11/9/01 10:55 AM Page 26

27

Next, it is necessary to recheck the shearwall length based on the shear modulus(G') of the assembly. A refined value forthe elastic fundamental period of vibra-tion can be determined from Formula [5]below, which is based on Method B and

Formula (30-10) described in Section1630.2.2.2 of the UBC. The length ofshear wall(s) determined above, and theshear modulus of the shear wall assem-bly, are used in this calculation. Since theshear wall length determined above isbased on shear load at Yield Limit State(Fyls), the shear modulus correspondingto the shear wall displacement at thisload [G'yls = (Fyls / �yls)(H / L)] is usedin this analysis. Values for G'yls, deter-mined from these cyclic load shear walltests, are summarized in Table 9 (notethat Ke-yls = G'yls when H / L = 1.0).

T � (2�)[ (W)(�2)]0.5[4]_______

(g)(F)(�)

For � = (F)(H) and g = 386.4 in. per______(G')(L)

second2 (= 32.2 ft per second2), thisformula reduces to:

(W)(H) 0.5____________T � (2�)[(386.4)(G')(L)] [5]

After determining T per Formula [5],recheck the spectral response modificationfactor per Figure 16-3 of the UBC, and thedesign shear load for the wall(s). If thisvalue has changed, repeat the processdescribed above until closure (e.g., nochange in the design shear load) results.

When the required length of shear wallhas been determined, the shear walldisplacement can be calculated perFormula [1].

Also, the process described above can be used to check the design for otherserviceability limits, such as limited dam-age at �0.01H. In this case, the unit shearload and shear modulus at the applicable

limit state (for example, see Table 9 forv0.01H, Ke-0.01H) are substituted whereappropriate in the calculations for deter-mining T per Formula [5] to recheck thedesign shear load based on the refinedspectral response modification factor, theshear capacity of walls required forstrength, and wall displacement.

A design example to demonstrate theanalysis procedures described above isincluded in Appendix C of this report.

The shear load should not exceed the“expected maximum shear strength”(page 24); and wall displacement shouldbe limited to a value which can be toler-ated without substantial damage to thestructure or attached finishes. If neces-sary, the shear wall design (or length) isrevised until a suitable solution is found.

CONCLUSIONS

The following conclusions are based onthe results of a series of eight cyclic loadshear wall tests conducted at theUniversity of California - Irvine. Sinceonly one test was conducted of eachshear wall configuration, and no matchingmonotonic shear wall tests were con-ducted, the conclusions are consideredpreliminary until additional shear walltests are conducted to confirm or modifythese conclusions. More extensive cyclicand monotonic shear wall tests are inprogress at the Research Center of APA – The Engineered Wood Association(Tacoma, WA), using recently acquiredcyclic load test equipment and apparatus.

1. The load-displacement curves andhysteresis loops for the test walls werenearly linear and repeatable for all cyclesof loading up to a story drift limit (dis-placement) of 0.48 inch, or 0.5% of wall

height. The shear load at this displace-ment averaged 1.7 times the allowabledesign shear load recognized in the 1997Uniform Building Code and NationalEvaluation Service, Inc. Report NER-108.The maximum shear strength of wallsaveraged about 1.4 times the shear loadattained at this displacement level.

2. The load factor (maximum shearload/allowable design shear load) aver-aged 2.5 for four test walls with 3/8-inchor 15/32-inch plywood sheathing fas-tened with 8d or 10d common hand-driven nails, respectively. The testedperformance appears to confirm theconservative nature of the allowabledesign shear loads for shear walls withplywood sheathing, which are based on monotonic tests and recognized inmodel building codes in the U.S.

3. The shear modulus of similarlyfabricated shear walls with orientedstrand board (OSB) or plywood sheath-ing appears to be equivalent. However,the maximum shear strength of shearwalls with OSB sheathing appears toaverage about 15% less than walls with acomparable thickness and grade of ply-wood sheathing. This appeared to resultfrom earlier fatigue failures of the fasten-ers used to attach higher density OSBsheathing, rather than to the sheathingitself. More cyclic load tests on matchedshear wall constructions are in progressto confirm or modify this relationship.

4. Gypsum wallboard contributed toincreased stiffness of a wall sheathed withwood structural panels in the range ofallowable design shear loads, but did notincrease the maximum shear strength ofthe wall, when subjected to cyclic(reversed) loads.

RR,W260,158.0 11/9/01 10:55 AM Page 27

28

5. One shear wall with sheathing fas-tened with common short (“diaphragm”)nails exhibited about 5% less maximumshear load than obtained in a matchingtest with full-length common nails.However, the energy dissipation of thewall with short nails was substantiallymore than obtained in the matching testwith full-length nails.

6. The expected maximum shearstrength, obtained after repeated(reversed) cycles of displacement, aver-aged about 18% less than the maximumshear strength reached in the initial dis-placement cycle. More cyclic (andmonotonic) load tests on matched shearwall constructions are in progress tofurther study this relationship.

7. Results of other shear wall tests indi-cate that the shear stiffness and shearstrength of shear walls can be increasedwhen sheathing panel ends are restrainedby bearing on the test fixture, simulatingbearing on adjoining panels, floors, ceil-ing framing or foundations at the topand/or bottom of the wall. The buildingcode in the United Kingdom also pro-vides for increased allowable design shearload when shear walls are axially loaded.

8. In matching cyclic load shear walltests, the “ramped” cyclic load test procedure provided results similar to theTCCMAR (Technical CoordinatingCommittee for Masonry Research)sequential phased displacement cyclicload test procedure, which incorporatesdecay cycles of load after ramped incre-ments of increased cyclic load. The TCC-MAR procedure has precedent for cyclicload testing of assemblies and fasteners,and is considered appropriate as a testprotocol for conducting cyclic load tests.

9. Shear wall performance as representedby these tests is contingent on use ofhold-down connectors which minimizeslip and deformation. If commerciallyavailable hold-down connectors are usedwhich permit more slip or deformation,the shear wall stiffness and shear strengthwill be compromised. In this case, result-ing shear wall performance should beverified by cyclic load tests onappropriate wall assemblies.

10. A method is presented for analyzingthe test results of cyclically loaded shearwalls to determine shear wall stiffness(shear modulus) at Yield Limit State andmaximum shear strength (Strength LimitState), and expected maximum shearstrength (e.g., cycled shear strength). Use of this information in analyzing astructure is described.

RR,W260,158.0 11/9/01 10:55 AM Page 28

29

LITERATURE CITED

1. ASTM. 1996. Standard Test Methods ofConducting Strength Tests of Panels forBuilding Construction (Sections 14 and15). ASTM E72-95. American Society forTesting and Materials, WestConshohocken, PA 19428.

2. ASTM. 1996. Standard Practice for StaticLoad Test for Shear Resistance of FramedWalls for Buildings. ASTM E564-95.American Society for Testing and Materials,West Conshohocken, PA 19428.

3. AF&PA. 1991. National DesignSpecification for Wood Construction.American Forest and Paper Association,Washington, DC 20036.

4. Adams, Noel R. 1988. Plywood ShearWalls. APA Research Report 105. APA –The Engineered Wood Association,Tacoma, WA 98411.

5. Tissell, John R. 1993. Structural PanelShear Walls. APA Research Report 154.APA – The Engineered Wood Association,Tacoma, WA 98411.

6. Yasumura, M. 1992. MechanicalProperties of Wood-Framed Shear WallsSubjected to Reversed Cyclic LateralLoading. CIB-W18/25-15-3. Proceedingsof International Council for BuildingResearch Studies and Documentation,Meeting 25 (Ahus, Sweden).

7. ATC. 1992. Guidelines for Cyclic SeismicTesting of Components of Steel Structures.ATC-24. Applied Technology Council,Redwood City, CA 94065.

8. Rose, J.D. and Keith, E.L. 1996. WoodStructural Panel Shear Walls with GypsumWallboard and Window/Door Openings. APAResearch Report 157. APA – The EngineeredWood Association, Tacoma, WA 98411.

9. SEAOSC. 1997. Standard Method ofCyclic (Reversed) Load Test for ShearResistance of Framed Walls for Buildings.Structural Engineers Association ofSouthern California, Whittier, CA 90601.

10. Newmark, N.M. and Hall, W.J. 1982.Earthquake Spectra and Design. MonographMNO-3. Earthquake EngineeringResearch Institute, Oakland, CA 94612.

11. ASCE. 1995. Standard for Load andResistance Factor Design (LRFD) forEngineered Wood Construction. AF&PA/ASCE 16-95. American Society for CivilEngineers, New York, NY 10017.

ACKNOWLEDGMENT

Robert Kazanjy, PE, President of PMREnterprises, Inc. (P.O. Box 10934, SantaAna, CA 92711) is thanked for providinginput and close cooperation in arrangingfor and conducting these cyclic loadshear wall tests for APA.

One of the cyclic load shear wall tests in this series of eight tests was sponsoredby Halstead Enterprises/Stan-Tech Divi-sion of Stanley-Bostitch Inc. (RanchoCucamonga, CA).f Their pneumaticallydriven 10d common short (“diaphragm”)nails were used in attaching OSB sheath-ing to framing in Test 8A. A Halsteadpneumatic nailing tool was used to drive

the nails; it was equipped with a depthcontrol adjustment device that limits thedepth of fastener embedment, so that thenail head was driven flush with the sur-face of the panel. Code recognition forHalstead fasteners for shear wall anddiaphragm applications was providedthrough ICBO Evaluation Service, Inc.Report No. 4296 (cancelled 1996).

Outside observers at the cyclic load shearwall tests conducted in April, 1995 at theUniversity of California - Irvine included:

John Kariotis, SE/Kariotis & Associates(Sierra Madre, CA)

Ben Schmid, SE/Consulting Engineer(Balboa Island, CA)

Robert Harder/City of Los Angeles,Dept. of Building & Safety

(Los Angeles, CA)

Chuck Shubnell/Halstead Enterprises(Rancho Cucamonga, CA)

Prof. Donald Breyer/California StatePolytechnic University (Pomona, CA)

Dr. Robin Shepherd, SE/Forensic ExpertAdvisers, Inc. (Santa Ana, CA)

Kurt Katsumata, SE/State of California,Div. of State Architect (Sacramento, CA)

fThis plant is now closed. Current fastener produc-tion by Stanley-Bostitch Inc. (East Greenwich, RI).Code recognition per National Evaluation Service,Inc. Report NER-272.

RR,W260,158.0 11/9/01 10:55 AM Page 29

30

APPENDIX A – LOAD-DISPLACEMENT CURVES

(APA/UNIVERSITY OF CALIFORNIA – IRVINE TESTS)

The following pages present load-displacement curves for the eight shear walls tested at the University of California - Irvine and discussed in the body of this report.

FIGURE A1

Load-displacement curve for cyclic load shear wall test (Test 1A)

Ap

plie

d S

hea

r Lo

ad

, p

ou

nd

s


FIGURE A2

Load-displacement curve for cyclic load shear wall test (Test 2B)

14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000

–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

Ap

plie

d S

hea

r Lo

ad

, p

ou

nd

s


14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

RR,W260,158.0 11/9/01 10:56 AM Page 30

31

FIGURE A3


FIGURE A4


Ap

plie

d S

hea

r Lo

ad

, p

ou

nd

s


14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

Ap

plie

d S

hea

r Lo

ad

, p

ou

nd

s


14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

RR,W260,158.0 11/9/01 10:56 AM Page 31

32

FIGURE A5


FIGURE A6


Ap

plied

Sh

ea

r Lo

ad

, p

ou

nd

s


14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000

–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

Ap

plied

Sh

ea

r Lo

ad

, p

ou

nd

s


14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

RR,W260,158.0 11/9/01 10:57 AM Page 32

33

FIGURE A7


FIGURE A8


Ap

plied

Sh

ea

r Lo

ad

, p

ou

nd

s


14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000

–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

Ap

plied

Sh

ea

r Lo

ad

, p

ou

nd

s


14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000

–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

RR,W260,158.0 11/9/01 10:57 AM Page 33

34

APPENDIX B – TEST

RESULTS AND LOAD-

DISPLACEMENT CURVES

(SCHMID/UNIVERSITY

OF CALIFORNIA –

IRVINE TESTS)

Three cyclic load tests on 8-foot x 8-footshear walls were conducted in Februaryand March, 1995 at the University ofCalifornia - Irvine, for California struc-tural engineer Ben L. Schmid, SE (BalboaIsland, CA). Permission has been given to include results of these testsin this report.

The construction of the shear walls, testsetup and TCCMAR sequential phaseddisplacement cyclic load test procedurein these tests were similar to the testsconducted at the University of California- Irvine for APA, as described in the bodyof the report. All walls were framed withDouglas-fir lumber as described in thereport, with only minor variations infastening of studs and posts to the bot-tom plate of the wall. All walls

were sheathed with 15/32-inch APAStructural I Rated Sheathing 32/16,Exposure 1 plywood (5-ply).

In two of the tests, the sheathing wasfastened with 10d common, full-lengthhand-driven nails spaced 4 inches o.c.along panel ends and edges and12 inches o.c. on intermediate studs. Theallowable design shear load for these twowalls is 510 pounds per foot in accordancewith Table 23-II-I-1 of the 1997 UniformBuilding Code; these walls were basicallyidentical to Test 1A in the APA test series.

In the other test, the sheathing was fas-tened with 8d common, full-length hand-driven nails spaced 3 inches o.c. alongpanel ends and edges and 12 inches o.c.on intermediate studs. The allowabledesign shear load for this wall was550 pounds per foot in accordance withthe code. The results of the test on this wallcan be compared to Test 6A in the APA testseries. These walls were similar in construc-tion except for the use of 15/32-inchStructural I plywood in Schmid’s test,whereas 3/8-inch Structural I plywood(3-ply) was used in APA’s test.

Table B1 provides a summary of shearwall deflection and shear modulus, andmaximum shear strength for comparisonwith APA tests described in the body ofthe report. The load-displacement curvesfrom these three cyclic load tests areshown in Figures B1 through B3.Figure B4 compares shear load applied tothe wall versus the measured tensionforce in the bolt which attaches the hold-down connector to the rigid base ofthe test fixture.

Although results are similar, somewhatbetter shear wall performance wasobserved in Schmid’s tests than in the APAtests. One factor that probably led to thisdifference is that the sheathing at the topof the wall came into bearing on the load-ing beam as the panels rotated during thecyclic load tests. As discussed in the bodyof this report, such bearing reduces thedeformation of nails attaching the sheath-ing to the framing, leading to better wallstiffness and greater shear strength. Thisdifference also has been noted in pastmonotonic load tests on walls.4

TABLE B1

Results of cyclic load shear wall tests (Schmid/UC-Irvine, 1995)

Design At � = 0.48 in.(a) At � = 0.96 in.(a) At � = 1.44 in.(a)Avg. Max. Load,

Test No. Description lb/ft lb � in.(a) Load, lb G1(b) Load, lb G2

(b) Load, lb G3(b) lb (L.F.)

BLS.1A 15/32" Plywood(c) 510 4,080 0.34 5,400 11.3 7,300 7.6 8,150 5.7 11,10010d com. @ 4" o.c. (2.7)

BLS.1B 15/32" Plywood(c) 510 4,080 0.36 5,000 10.4 7,450 7.8 8,300 5.8 10,90010d com. @ 4" o.c. (2.7)

BLS.2A 15/32" Plywood(c) 550 4,400 0.26 7,150 14.9 9,550 9.9 9,950(d) 6.9 12,4008d com. @ 3" o.c. (2.8)

(a) Tabulated data is average of ± cycles. Displacement (�) and shear loads – after fourth cycle of loading.

(b) lb/in. ÷ 1,000 (= kips/in.)

(c) APA Structural I Rated Sheathing 32/16 (5-ply plywood): Douglas-fir framing (studs spaced 16" o.c.).

(d) Expected maximum (cycled) shear strength: 10,500 lbs. at � = 1.33 in. (G = 7.9 kips/in.)

RR,W260,158.0 11/9/01 10:57 AM Page 34

35

FIGURE B1

Load-displacement curve for cyclic load shear wall test (Test BLS.1A)

FIGURE B2

Load-displacement curve for cyclic load shear wall test (Test BLS.1B)

Ap

plie

d S

hea

r Lo

ad

, p

ou

nd

s


14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000

–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

Ap

plie

d S

hea

r Lo

ad

, p

ou

nd

s


14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000

–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

RR,W260,158.0 11/9/01 10:57 AM Page 35

36

FIGURE B3

Load-displacement curve for cyclic load shear wall test (Test BLS.2A)

FIGURE B4

Typical hold-down load vs. shear wall load (Test BLS.1A)

Ap

plie

d S

hea

r Lo

ad

, p

ou

nd

s


14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000

–3.2 –2.4 –1.6 –0.8 0 0.8 1.6 2.4 3.2

Ap

plie

d S

hea

r Lo

ad

, p

ou

nd

s

Holddown Bolt Load, pounds

14000

12000

10000

8000

6000

4000

2000

0

–2000

–4000

–6000

–8000

–10000

–12000

–14000

–2000 –0 2000 4000 6000 8000 10000 12000

RR,W260,158.0 11/9/01 10:58 AM Page 36

37

APPENDIX C

Design Example for Shear Wall (Basedon 1997 Uniform Building Code)

Problem: Calculate the length of shearwall (8 ft high) required to resist a totalseismic dead load of 30 kips (SeismicZone 4). Design shall be based on provi-sions of Chapters 16 and 23 in the 1997Uniform Building Code (UBC).

The 1997 UBC introduces design ofwood-framed structures using Load andResistance Factor Design (LRFD) analy-sis, per UBC Section 1605.1. For wood,LRFD provisions11 are referenced in UBCSection 2303.5 (Item 5.4). When analyz-ing a structure by LRFD, the full value ofearthquake load (1.0 E) is used per UBCSections 1605.1 and 1612.2.1.

Given:

a. Building design is assumed to complywith the criteria of UBC Section1630.1.1 for assigning a Reliability/Redundancy Factor ρ = 1.0. For shearwall design using wood structural panels,based on LRFD analysis per ASCEStandard 16-9511, the Time Effect Factorλ = 1.0 for earthquake (and wind) loads,and the Resistance Factor φ = 0.65(based on connections).

b. Shear wall consists of 15/32 in. APAStructural I Rated Sheathing 32/16,Exposure 1 (oriented strand board – OSBper U.S. Product Standard PS 2-92 orAPA Standard PRP-108); fastened with10d common nails spaced 4 in. o.c. atpanel edges/12 in. o.c. to intermediatestuds; framing of Douglas-fir, studsspaced 16 in. o.c. (4x end posts, 3xframing at common panel edges andbottom plate, 2x framing for intermediatestuds and double top plate); see Figure 1

in report for details of 8 ft x 8 ft test wall.Allowable design shear load for wall =510 lb/ft (0.51 kip/ft) (Ref: UBC Table 23-II-I-1).

c. Shear wall design is based on proper-ties determined by cyclic load testing ofwalls (average of Tests 4A and 5A in APAResearch Report 158). Average valuesfrom Tests 4A and 5A (Tables 8 and 9 in this report):

Gyls = Ke-yls (all shear modulus valuesbased on H/L = 1.0) = 17.3 kips/in.

G0.01H = Ke-0.01H = 7.8 kips/in.

Gsls = Ke-sls = 6.0 kips/in.

vyls = 0.74 kip/ft

v0.01H = 0.93 kip/ft

vsls = 1.07 kips/ft

Ro = 1.4

Rd-S = 2.8

1. Determine required length of shearwall based on shear strength.a. Calculate estimated elastic fundamental period of vibration.

Ref: UBC Sec. 1630.2.2.1, Method A.

T = (0.020)(hn)0.75 = (0.020)(8)0.75

= 0.10 sec.

b. Determine spectral accelerationcoefficient (Cv / T).

Ref: Fig. 16-3 and Tables 16-Q (Ca),16-R (Cv ), 16-S (Na ) and 16-T (Nv) ofUBC. Assumed: Seismic Zone 4, SoilProfile Type SD, Near-Source Factors Na = 1.0 and Nv = 1.0.

Ts = Cv / 2.5Ca = (0.64)(1.0) /(2.5)(0.44)(1.0) = 0.58 sec.

To = 0.2 Ts = (0.2)(0.58) = 0.12 sec.

Proportioning from Fig. 16-3 with abovevalues for Ts and To:

For T = 0.10 sec., Cv / T = Ca +[(2.5Ca – Ca)(0.10) / 0.12] = 2.29 Ca

c. Calculate design base shear (LRFD).

Ref: UBC Sec. 1630.2.1.

Vlrfd = (Cv / T)(I)(W) / R; assumeImportance Factor I = 1.0 and R = 5.5(Ref: UBC Tables 16-K and 16-N). Note:See following Discussion for AllowableStress Design (ASD) analysis. For ASD,earthquake load Easd = Elrfd / 1.4 perUBC Sections 1612.3.1 and 1612.3.2.

Vlrfd = (2.29 Ca)(W) / R =(2.29)(0.44)(30) / 5.5 = 5.5 kips

d. Calculate length of shear wallrequired for shear strength at YieldLimit State.

Llrfd = Vlrfd / vyls = 5.5 / 0.74 = 7.4 ft

e. Check length of shear wall requiredbased on shear strength at StrengthLimit State.

Ref: Resistance Factor φ = 0.65 (forconnections; also applies to shear walls)per ASCE 16-9511. See followingDiscussion regarding comparison of walllength as determined by vlrfd and vasd.

Llrfd = Vlrfd / (φ)(vsls) = 5.5 / (0.65)(1.07)= 7.9 ft (controls, > 7.4 ft)

f. Recheck elastic fundamental periodof vibration and shear wall length.

Ref: UBC Sec. 1630.2.2.2, Method B.

(W)(H)_____________T = (2)(π)[(386.4)(G')(L)]0.5

(30)(8)________________= (2)(3.14)[(386.4)(17.3)(7.9)]0.5

= 0.42 sec.[> (1.3)(T = 0.10 sec. per Method A;see 1.a above) = 0.13 sec. Therefore useT = 0.13 sec. per UBC Sec. 1630.2.2.2].

Note use of Gyls for G′. Length of shearwall is based on shear strength atStrength Limit State; and T is applicablefor shear loads up to Yield Limit State.

From Fig. 16-3 for T = 0.13 sec., Cv / T= 2.5 Ca (see 1.b above).

RR,W260,158.0 11/9/01 10:58 AM Page 37

38

Vlrfd = (2.5Ca)(W) / R = (2.5)(0.44)(30) / 5.5 = 6.0 kips

Llrfd = Vlrfd/ (φ)(vsls) = 6.0 / (0.65)(1.07)= 8.6 ft

Recheck T:

(30)(8)________________T = (2)(3.14)[(386.4)(17.3)(8.6)]0.5

= 0.41 sec. [use T = 0.13 sec. per 1.f above]

From Fig. 16-3 for T = 0.13 sec., Cv / T = 2.5Ca

V = 6.0 kips (same as above) (Closure – ok)

Llrfd = 8.6 ft (based on shear strength atStrength Limit State)

Note: Shear capacity at Yield Limit State = Vyls = (vyls)(Llrfd) = (0.74)(8.6)= 6.4 kips. This shear capacity corre-sponds to 6.4 / 6.0 = 1.1 times designbase shear.

g. Check shear wall displacement atdesign base shear.

∆ = (Vlrfd)(H) / (Gyls)(Llrfd) = (6.0)(8) / (17.3)(8.6) = 0.32 in.(corresponds to 0.003 H for 8-ft-high wall).

2. Determine shear capacity of shearwall based on limited structural damage(�max = 0.01 H).a. Length of shear wall required forshear strength.

Llrfd = 8.6 ft (see 1.a - 1.f above)

b. Recheck elastic fundamental periodof vibration.

(30)(8)______________T = (2)(3.14)[(386.4)(7.8)(8.6)]0.5

= 0.60 sec. [use T = 0.13 sec. per 1.f above]

Note use of G0.01H for G ′.

From Fig. 16-3 for T = 0.13 sec., Cv / T= 2.5 Ca (see 1.b above)

V = 6.0 kips (same as 1.f above)(Closure – ok)

c. Calculate shear capacity at � = 0.01H.

V0.01H = (v0.01H)(Llrfd) = (0.93)(8.6) = 8.0 kips

Note: This shear capacity corresponds to8.0 / 6.0 = 1.3 times design base shear.For limited damage, Ro = vsls / v0.01H

= 1.07 / 0.93 = 1.1.

d. Check shear wall displacement.

D = (Vlrfd)(H) / (G0.01H)(Llrfd) =(8.0)(8) / (7.8)(8.6) = 0.95 in.(corresponds to 0.01H for 8-ft-high wall).

Discussion:

Allowable Stress Design (ASD)A factor of 1.4 was introduced in the1997 UBC for converting calculatedearthquake loads based on LRFD to ASD.This can be seen by comparing earth-quake load (E) in UBC Section 1612.2.1for LRFD, with UBC Sections 1612.3.1or 1612.3.2 for ASD. Note that thedesign base shear is calculated by using aResponse Modification Factor R = 5.5for light-framed shear walls with woodstructural panel sheathing as specified inUBC Table 16-N (or 4.5 for light-framedwalls sheathed with materials other thanwood structural panels). These valuesapply to LRFD analysis and averageabout 1.4 times the design base sheardetermined with corresponding Rw fac-tors (8 or 6) for ASD analysis in previous editions of the code.

When the above provisions are taken into account, a comparable shear walldesign results when an ASD analysismethod is used. Allowable design shearloads for shear walls analyzed by ASD are contained in UBC Table 23-II-I-1.

To compare the above shear wall designexample using ASD instead of LRFDanalysis per UBC provisions, consider thelength of shear wall required to resist the

calculated design base shear of 6.0 kips,applicable to LRFD analysis with maxi-mum V = (2.5Ca)(W) / R as noted in 1.fof the design example above. For thisdesign example, shear wall length is con-trolled by required shear strength atStrength Limit State.

Shear wall length would be determinedas follows according to UBC provisionsnoted above, and based on the ASDallowable design shear vasd = 0.51 kip/ft:

Lasd = (Vlrfd / (1.4)(vasd) = (6.0) / (1.4)(0.51) = 8.4 ft

The resulting shear wall length is approximately the same as in 1.f in theabove analysis. It is noted that the ratio(vyls/vasd), which relates Yield Limit Stateto allowable design shear (ASD basis),has a value of 1.45 (=0.74/0.51) whichis nearly equivalent to the 1.4 factor used in the UBC to adjust from ASD to LRFD values.

For comparison, a check was made on therequired shear wall length determinedusing ASD design and the 1994 UBCrequirements for equivalent conditions:

Vasd = (Z)(C)(I)(W) / Rw =(0.40)(2.75)(1.0)(30) / 8 = 4.1 kips [ = 0.138W]

Lasd = 4.1 / 0.51 = 8.1 ft, or slightly lessthan determined above (shear wall lengthbased on 1997 UBC is 4% longer).

The test results reported herein (Tables 8and 9, on page 26 of this report) providepreliminary information on Ro and Rd

factors obtained from cyclic load testing ofshear walls with wood structural panels.More cyclic load shear wall testing is inprocess by APA to confirm these values.

RR,W260,158.0 11/9/01 10:58 AM Page 38

APA RESEARCH AND TESTING

APA – The Engineered Wood Association’s 37,000-square-foot Research

Center in Tacoma, Washington is the most sophisticated facility for basic

panel research and testing in the world. The center is staffed with an

experienced corps of engineers, wood scientists, and wood product techni-

cians. Their research and development assignments directly or indirectly

benefit all specifiers and users of structural wood panel products.

RR,W260,158.0 11/9/01 10:58 AM Page 39

We have field representatives in most major U.S. cities and in Canada who can help answer questions involving APA

trademarked products. For additional assistance in specifying APA engineered

wood products, contact us:

APA – THE ENGINEERED WOOD ASSOCIATION

HEADQUARTERS7011 So. 19th St. ■ P.O. Box 11700Tacoma, Washington 98411-0700

(253) 565-6600 ■ Fax: (253) 565-7265

(International Offices: Bournemouth,United Kingdom; Hamburg, Germany;

Mexico City, Mexico; Tokyo, Japan.)

PRODUCT SUPPORT HELP DESK(253) 620-7400

E-mail Address: [email protected]

The product use recommendations in thispublication are based on APA – TheEngineered Wood Association’s continuingprograms of laboratory testing, productresearch, and comprehensive field experi-ence. However, because the Association hasno control over quality of workmanship orthe conditions under which engineered woodproducts are used, it cannot accept responsi-bility for product performance or designs asactually constructed. Because engineeredwood product performance requirementsvary geographically, consult your local archi-tect, engineer or design professional toassure compliance with code, construction,and performance requirements.

Form No. W260ARevised June 1999/0300

www.apawood.org@Web Address:

THE ENGINEEREDWOOD ASSOCIATION

APA

RR,W260,158.0 11/9/01 10:58 AM Page 40

Documents

Preliminary Testing of Wood Structural Panel Shear ... - Bay Area … Cyclic Shear Wall Testing.pdf · Preliminary Testing of Wood Structural Panel Shear Walls Under Cyclic (Reversed)