ATC 40 Vol. 2

"

Seismic evaluation and retrofit. of concrete buildings Volume 2-Appendices

Applied Technology Counr:;;i CALIFORf\IIt1. SEISMIC SAFETY COMMISSION Proposition 122 Seisn lie Retrofit Practice8 impr
sse 96-01 "

.

'oh·

4.183~ G8 ~ '"

,'. .j

/

• .....,

I-:';':~::,;~::;:;"1

\ /"U" I 1 of Roorkee I ~llivel'5ity

I \ \

I I Call No ........... · .. ·· ..• ·····•·····

unci}

California Seismic Safety Commission

) is a non:d in 1971 ineers Asso1 Board of ppointed by , the StrucI, the Western ,ssociations, oed with the }irector

\

. :. .=. .\ i

In practitiodesign spe.1quake) in vely using lentifies and _ • s consensus opinions on structural engineering issues in a nonproprietary fonnal. ATC thereby fulfills a unique role in funded infonnation transfer.

I'

L~:c~:.=~.=

The California Seismic Safety Commission consists of fifteen members appointed by the Governor and two members representing the State Senate and State Assembly. Disciplines represented on the Commission include seismology, engineering, geology, fire protection, emergency services, public utilities, insurance, social services, local government, building code enforcement, planning and architecture. As a nonpartisan, single-purpose body, the mission of the Commission is to improve the well being of the people of California through cost-effective measures that lower earthquake risks to life and property. It sponsors legislation and advocates building code changes to improve buildings and other facilities, provides a forum for representatives of all public and private interests and academic disciplines related to earthquakes, and publishes reports, policy recommendations, and guides to improve public safety in earthquakes.

It works toward long-term improvements in all areas

Project management and administration are carried out by a full-time Executive Director and support staff. Project work is conducted by a wide range of highly qualified consulting professionals, thus incorporating the experience of many individuals from academia, research, and professional practice who would not be available from any single organization. Funding for ATC projects is obtained from government agencies and from the private sector in the form of tax-deductible contributions.

affecting seismic safety by: encouraging and assisting local governments, state agencies, and businesses to implement mitigation measures to make sure that they will be able to operate after earthquakes; establishing priorities for action to reduce earthquake risks; identifying needs for earthquake education, research, and legislation; and reviewing emergency response, recovery, and reconstruction efforts after damaging earthquakes so that lessons learned can be applied to future earthquakes.

1996-1997 Board of Direetors

Current (1996) Commission Members

John C. Theiss, President C. Mark Saunders, Vice President

Lloyd S. Cluff, Chairman James E. Slosson, Vice Chairman Alfred E. Alquist. State Senator Dominic L. Cortese. State Assemblyman Hal Bernson Jerry C. Chang Robert Downer Frederick M. Herman Jeffrey Johnson Corliss Lee Gary L. McGavin Daniel Shapiro Lowell E. Shields Patricia Snyder Keither M. Wheeler H. Robert Wirtz

Bijan Mohraz, Secretaryffreasurer Edwin T. Huston, Past President Arthur N. L. Chiu John M. Coil Edwin T. Dean Robert G. Dean Douglas A. Foutch James R. Libby Kenneth A. Luttrell Andrew T. Merovich Scott A. Stedman Jonathan G. Shipp , S:;harles Thornton

Disclaimer While the information presented in this report is believed to be correct, the Applied Technology Council and the California Seismic Safety Commission assume no responsibility for its accuracy or for the opinions expressed herein. The material presented in this publication should not be used or relied upon for any specific application without competent examination and verification of its accuracy, suitability, and applicability by qualified professionals. Users of information from this publication aSSume all liability arising from such use. Cover IllustratIOn: SLate Office Bldg, 12'h and N St.. Sacramento. CA, provided by Chris Arnold.

b

lmissio: n consists morand ! and State Cornmis,ogy. fire

ities, insUl

ATC-GO

,ilding cod ~

mission c ng of the ! measures Jerty. It . g code cilities. public anI related to recommer

ory in earth in all areas md assisti" iinesses to ure thaI Ihl !stablishinl risks; iden! !arch. and )onse, remaging ! applied I(

bers

Seismic Evaluation and Retrofit of Concrete Buildings volume 2-Appendices

~~D TECHN~LOGY

COUNCIL 555 Twin Dolphin Drive, Suite 550 Redwood City, California 94065 Funded by

SEISMIC SAFETY COMMISSION State of California Products 1.2 and 1.3 of the Proposition 122 Seismic Retrofit Practices Improvement Program

n

lblyman

PRINCIPAL INVESTIGATOR Craig D. Comartin CO-PRINCIPAL INVESTIGATOR PROJECT DIRECTOR Richard W. Niewiarowski SENIOR ADVISOR Christopher Rojahn

logy lracyor ,sed or its accu; publi-

Report No. SSC 96-01 November 1996

preface Proposition 122 passed by California's voters in 1990, created the Earthquake Safety and Public Buildings Rehabilitation Fund of 1990, supported by a $300 million general obligation bond program for the seismic retrofit of state and local government buildings. As a part of the program, Proposition 122 authorizes the California Seismic Safety Commission (CSSC) to use up to 1% of the proceeds of the bonds, or approximately $3 million, to carry out a range of activities that will capitalize on the seismic retrofit experience in the private sector to improve seismic retrofit practices for government buildings. The purpose of California's Proposition 122 research and development program is to develop state-of-the-practice recommendations to address current needs for seismic retrofit provisions and seismic risk decision tools. It is focused specifically on vulnerable concrete structures consistent with the types of concrete buildings that make up a significant portion of California's state and local government inventories. [n 1994, as part of the Proposition 122 Seismic Retrofit Practices Improvement Program, the Commission awarded the Applied Technology Council (ATC) a contract to develop a recommended methodology and commentary for the •eismic evaluation and retrofit of existing con~rete buildings (Product 1.2). In 1995 the :::ommission awarded a second, related contract :0 ATC to expand the Product 1.2 effort to in:lude effects of foundations on the seismic per'ormance of existing concrete buildings Product 1.3). The results of the two projects lave been combined and are presented in this \TC-40 Report (also known as SSC-96-01). rwo other reports recently published by the :a1ifornia Seismic Safety Commission, the 'rovisional Commentary for Seismic Retrofit 1994) and the Review of Seismic Research Re'ults on Existing Buildings (1994), are Products .. 1 and 3.1 of the Proposition 122 Program, re.pectively. These two previous reports provide he primary basis for the development of the ecommended methodology and commentary :ontained in this document.

This document is organized into two volumes. Volume One contains the main body of the evaluation and retrofit methodology, presented in 13 chapters, with a glossary and a list of references. This volume contains all of the parts of the document required for the evaluation and retrofit of buildings. Volume Two consists of Appendices containing supporting materials related to the methodology: four example building case study reports, a cost effectiveness study related to the four building studies, and a review of research on the effects of foundation conditions on the seismic performance of concrete buildings. This report was prepared under the direction of A TC Senior Consultant Craig Comartin, who served as Principal Investigator, and Richard W. Niewiarowski, who served as Co-Principal Investigator and Project Director. Fred Turner served as CSSC Project Manager. Overview and guidance were provided by the Proposition 122 Oversight Panel consisting of Frederick M. Herman (Chair), Richard Conrad, Ross Cranmer, Wilfred Iwan, Roy Johnston, Frank McClure, Gary McGavin, Joel McRonald, Joseph P. Nicoletti, Stanley Scott, and Lowell Shields. The Product 1.2 methodology and commentary were prepared by Sigmund A. Freeman, Ronald O. Hamburger, William T . Holmes, Charles Kircher, Jack P. Moehle, Thomas A. Sabol, and Nabih Youssef (Product 1.2 Senior Advisory Panel). The Product 1.3 Geotechnical/Structural Working Group consisted of Sunil Gupta, Geoffrey Martin, Marshall Lew, and Lelio Mejia. William T. Holmes, Y oshi Moriwaki, Maurice Power and Nabili Youssef served on the Product 1.3 Senior Advisory Panel. Gregory P. Luth and Tom H. Hale, respectively, served as the Quality Assurance Consultant and the Cost Effectiveness Study Consultant. Wendy Rule served as Technical Editor, and Gail Hynes Shea served as Publications Consultant. Richard McCarthy CSSC Executive Director Christopher Rojalm ATC Executive Director & ATC-40 Senior Advisor

III

Oversight Panel for proposition 122 Seismic Retrofit Practices Improvement program Frederick M. Herman, Chair Seismic Safety Commission Local Government/Building Official

Richard Conrad Building Standards Commission

Ross Cranmer Building Official Structural Engineer

Roy Johnston Structural Engineer

Frank McClure Structural Engineer

Joel McRonald Division of the State Architect

Joseph P. Nicoletti Structural Engineer

Dr. Wilfred Iwan Mechanical Engineer Gary McGavin Seismic Safety Commission Architect Stanley Scott Research Political Scientist

Lowell E. Shields Seismic Safety Commission Mechanical Engineer

Seismic Safety commission Staff Richard McCarthy Executive Director Karen Cogan Deborah Penny Carmen Marquez

Iv

Fred Turner Project Manager Chris Lindstrom Ed Hensley Teri DeVriend Kathy Goodell

S

c

V Ii

s 1 ~

1

c

Product 1.2 Senior Advisory Panel IS

Sigmund A. Freeman Wiss. Janney. Elstner & Associates Charles Kircher Charles Kircher & Assocates

Ronald O. Hamburger EQE International

William T. Holmes Rutherford & Chekene

Jack Moehle Earthquake Engineering Research Center

Thomas A. Sabol Engelkirk & Sabol

Nabih F . Youssef Nabih Youssef & Associates

Product 1.3 Senior Advisory Panel mission

William T. Holmes Rutherford & Chekene

Maurice Power Geomatrix Consultants. Inc.

Yoshi Moriwaki Woodward-Clyde Consultants

Nabih F. Youssef Nabih Youssef & Associates

Product 1.3 Geotechnical/structural working Group Sunil Gupta £Q Tech Consultants

Geoffrey R. Martin University of Southern California

Marshall Lew Law/Crandall. Inc.

Lelio Mejia Woodward-Clyde Consultants

Quality Assurance Consultant Jregory P. Luth 'Jregory P. Luth & Associates

Technical Editor Wendy Rule Richmond. CA

:ost Effectiveness study Consultant rom H. Hale fimmy R. Yee Consulting Engineers

Publications Consultant Gail Hynes Shea Albany. CA

v

p

Seismic Evaluation and Retrofit of Concrete Buildings products 1.2 and 1.3 of the proposition 122 seismic Retrofit Practices Improvement Program

Table of Contents Volume 1

Preface ................................................................................................... iii Glossary ................................................................................................. xi Executive Summary ................................................................................... xv Chapter 1 Introduction ........................................................................... \-1 1.1 Purpose ........................................................................ \-\ 1.2 Scope .......................................................................... 1-2 1.3 Organization and Contents ................................................. 1-5 Chapter 2 Overview .............................................................................. 2-1 2.1 Introduction ............................................................ : ..... 2-\ 2.2 Changes in Perspective ..................................................... 2-3 2.3 Getting Started ............................................................... 2-6 2.4 Basic Evaluation and Retrofit Strategy ................................. 2-11 2.5 Evaluation and Retrofit Concept ........................................ 2-14 2.6 Final Design and Construction .......................................... 2-19 Chapter 3 Performance Objectives ............................................................. 3-1 3.1 Introduction .................................................................. 3-1 3.2 Performance Levels ......................................................... 3-\ 3.3 Earthquake Ground Motion ................................................ 3-8 3.4 Performance Objectives .................................................... 3-9 3.5 Assignment of Performance Objectives ................................ 3-12 Chapter 4 Seismic Hazard ...................................................................... .4-1 4.1 Scope ......................................................................... .4-1 4.2 Earthquake Ground Shaking Hazard Levels ............................ .4-1 4.3 Ground Failure .............................................................. .4-2 4.4 Primary Ground Shaking Criteria ........................................ .4-5 4.5 Specification of Supplementary Criteria ............................... 4-12 Chapter 5 Determination of Deficiencies ..................................................... 5-1 5.1 Introduction .................................................................. 5-1

·able Of

contents

vii

SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

Chapter 6

Chapter 7

Chapter 8

Chapter 9

Chapter 10

Chapter 11

Chapter 12

viii

5.2 Description: Typical Layouts and Details ............................... 5-1 5.3 Seismic Performance ....................................................... 5-5 5.4 Data Collection ............................................................ 5-12 5.5 Review of Seismic Hazard ............................................... 5-17 5.6 Identification of Potential Deficiencies ................................ 5-18 5.7 Preliminary Evaluation of Anticipated Seismic Performance ...... 5-20 5.8 Preliminary Evaluation Conclusions and Recommendations ....... 5-21 Retrofit Strategies .................................................................... 6-1 6.1 Introduction ................................................... ~ .............. 6-1 6.2 Alternative Retrofit Strategies ............................................. 6-4 6.3 Design Constraints and Considerations ................................ 6-24 6.4 Strategy Selection ......................................................... 6-27 6.5 Preliminary Design ....................................................... 6-30 Quality Assurance Procedures ..................................................... 7-1 7.1 General. ....................................................................... 7~1 7.2 Peer Review .................................................................. 7-2 7.3 Plan Check ................................................................... 7-8 7.4 Construction Quality Assurance ........................................ 7-10 Nonlinear Static Analysis Procedures ............................................ 8-1 8.1 Introduction .................................................................. 8-1 8.2 Methods to Perform Simplified Nonlinear Analysis ................... 8-3 8.3 Illustrative Example ....................................................... 8-34 8.4 Other Analysis Methods .................................................. 8-54 8.5 Basics of Structural Dynamics .......................................... 8-57 Modeling Rules ....................................................................... 9-1 9.1 General ......................................................................... 9-1 9.2 Loads ............. : ............................................................ 9-2 9.3 Global Building Considerations ........................................... 9-4 9.4 Element Models ............................................................. 9-7 9.5 Component Models ....................................................... 9-19 9.6 Notations .................................................................... 9-46 Foundation Effects ................................................................. 10-1 10.1 General. . .. .... . . . . . . . .. .. .. . . .. . . . . .. . . . .. .. . .. . . . .. . . .. . . .. . . . . . .. . . . . . .. .. 10-1 10.2 Foundation System and Global Structural Model .................... 10-2 10.3 Foundation Elements ..................................................... 10-7 10.4 Properties of Geotechnical Components .............................. 10-12 10.5 Characterization of Site Soils ........................................... 10-20 10.6 Response Limits and Acceptability Criteria .......................... 10-28 10.7 Modifications to Foundation Systems ................................. 10-29 Response Limits .................................................................... 11-1 11.1 General. ..................................................................... 11-1 11.2 Descriptive Limits of Expected Performance ......................... 11-2 11.3 Global Building Acceptability Limits ........... '" .................... 11-2 11.4 Element and Component Acceptability Limits ........................ 11-5 Nonstructural Components ....................................................... 12-1

Table of Contents

\

T

~S


12.1 Introduction ................................................................ 12.2 Acceptability Criteria ..................................................... Chapter 13 Conclusions and Future Directions .............................................. 13.1 Introduction ................................................................ 13.2 Additional Data ............................................................ 13.3 Potential BenefIts .......................................................... 13.4 Major Challenges .......................................................... 13.5 Recommended Action Plan .............................................. References ...........................................................................................

12-1 12-1 13-1 13-1 13-1 13-4 13-5 13-6 14-1

volume 2-Appendlces Appendix A Escondido Village Midrise, Stanford, California .............................. A-I Appendix B Barrington Medical Center, Los Angeles, California ......................... B-1 Appendix C Administration Building, California State University at Northridge, Northridge, California .. : .......................................................... C-l Appendix D Holiday Inn, Van Nuys, California .............................................. D-l Appendix E Cost Effectiveness Study ........................................................... E-l Appendix F Supplemental Information on Foundation Effects ............................. F-l Appendix G Applied Technology Council Projects and Report Information .............. G-l

I

Of contenl'able Of contents

Ix

f SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

APpendix A

Example Building Study Escondido village Mldrlse stanford, California prepared by EQE International 44 Montgomery Street, Suite 3200 San Francisco, California 94104

'1Il1endlx A, Escondido village Mldrlse

A-'


Table of Contents 1. Introduction ................................................................................................. A-5 1.1 Purpose ...................................................................................... A-5 1.2 Scope of Example Building Study ....................................................... A-5 1.3 Summary of Findings ..................................................................... A-5 2. Building and Site Description ............................................................................ A-7 2.1 General ...................................................................................... A-7 2.2 Structural Systems and Members ....................................................... A-8 2.3 Soil and Seismicity ........................................................................ A-9 2.4 Building Performance During the Lorna Prieta Earthquake ........................ A-9 3. Preliminary Evaluation ................................................................................... A-9 3.1 Summary .................................................................................... A-9 3.2 FEMA-178 Evaluation Statements ..................................................... A-II 3.3 Elastic Analysis ........................................................................... A-I4 4. Evaluation by Product 1.2 Methodology .............................................................. A-IS 4.1 Introduction ................................................................................ A-IS 4.2 Analysis Methodology ................................................................... A-IS 4.3 Structure ryIodeling ....................................................................... A-IS 4.4 Pushover Analysis ........................................................................ A-22 4.5 Performance Point. ....................................................................... A-27 4.6 Performance Assessment ................................................................ A-31 5. Conceptual Retrofit Designs .......................................................... , ................. A-33 5.1 Performance Objectives ................................................................. A-33 5.2 Retrofit Strategies ........................................................................ A-33 5.3 Retrofit Systems .......................................................................... A-34 6. Assessment of the Product 1.2 Methodology ......................................................... A-36 6.1 Damage Prediction ....................................................................... A-36 6.2 Comparison with Preliminary Evaluation Findings ................................. A-36 6.3 Comparison with Inelastic Time-History Analysis .................................. A-37 6.4 Conclusions ................................................................................ A-37 7. Foundation Analysis ...................................................................................... A-38 7.1 Introduction ................................................................................ A-38 7.2 Varying Soil Parameters ................................................................. A-38 7.3 Comparisons with Inelastic Time-History Analysis ................................. A-42 7.4 Conclusions ................................................................................ A-43 8. References ................................................................................................. A-43

I!Ipendlx A. Escondido Village Mldrlse

--SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

APpendix A

Example Building Study Escondido village Midrise stanford, California 1.

Introduction

1.1

purpose

The purpose of this example building study is to illustrate and evaluate the techniques outlined in products 1.2 and 1.3 of Proposition 122 as a tool for the evaluation and retrofit of existing concrete buildings. Titled Seismic Evaluation and Retrofit of Existing Concrete Buildings, Volume 1, the document is referred to herein as the . Methodology.

1.2

scope of Example Building study

This study presents the evaluation and conceptual retrofit design of a concrete building located on the Stanford University campus, following the recommendations of the Methodology. This study was performed ;oincident with the various draft stages of :ievelopment of the Methodology and feedback from this study was used to affect final nodifications of the Methodology. Our scope neluded: • Preliminary evaluation (Section 3 of this report) •

Modeling, analysis, and assessment by nonlinear pushover analysis (Section 4)

•

Conceptual retrofit (Section 5)

• Assessment of the Methodology (Section 6) • Foundation analysis (Section 7) 1.3

summary of Findings

The tools currently available to the structural ngineer for seismic evaluation and retrofit of

Ippendlx A, Escondido Village Mldrlse

existing concrete structures are essentially limited to the building codes for new construction and the FEMA-178 document. In comparison with these existing tools, the Methodology appears to represent a significant enhancement in the state of practice. Based on the Escondido Village Midrise (EVM) case study, the Methodology appears to provide a realistic and conservative, if not completely accurate, approach to seismic evaluation of complex reinforced concrete structures yet also permi ts the engineer to develop retrofit strategies that are significantly more cost effective than were traditionally utilized in the past. FEMA-178 evaluations of the EVM buildings indicate an inability to satisfy the life safety performance level for the design earthquake, due to a lateral force resisting system comprised of discontinuous shear walls, with inadequate shear capacity. Prior to development of the Methodology, the standard approach for mitigation of such deficiencies would have been the introduction of an extensive number of supplemental shear walls to the structure. This would have great architectural and economic impact on the building. In comparison, the Methodology identified that the existing walls essentially provide adequate drift control for the structure, but that several other vulnerabilities related to shear capacity of the lower story columns and punching shear capacity of the floor slabs exist. Retrofit of these vulnerabilities, which were not specifically identified by the FEMA-178 approach, was found to be possible with much less architectural impact on the buildings and at

A-5


Table 1.$·1. ComparIson Of flOOf DlsPlacemen.;;;ts~ _ _

significantly reduced cost compared to alternative approaches suggested by the FEMA-178 evaluation. These retrofit modifications have actually been constructed, within a time period of approximately 3 months and while the buildings remained nearly completely occupied. Compared to existing approaches, the Methodology does require more complex and time consuming work on the part of the structural designer. However, the additional level of effort required is well within the capability of the average practicing engineer in California, who has the familiarity with the basic concepts of structural dynamics and inelastic behavior of structures that is essential to being able to design effective seismic resistant systems, either for new or existing buildings. In the case of the EVM buildings, the additional effort and cost invested in the evaluation and analysis of the structure resulted in a very substantial reduction in retrofit construction costs, and consequently in overall project costs. Notwithstanding the above, it can not be overemphasized that this Methodology does not provide an "exact" tool for the seismic evaluation of structures, and that in fact, such an "exact" tool does not exist within our current technological capabilities. In the EVM case study, target displacements were determined by two alternative methods encompassed by the Methodology, the Displacement Coefficient Method and the

A-a

Capacity-Spectrum approach; as well as by two other approaches that are commonly cited in the literature - the so called "Equal Displacement Approximation" and non-linear response history analysis, in which the average result for 20 different response histories is shown. Table 1.3-1 indicates the range of computed roof displacement obtained by these alternatives methods, and also provides a normalized index that consists of the ratio of the displacement computed by each method to the maximum displacement predicted by the nonlinear response history analyses. As can be seen by evaluating the data contained in Table 1.3-1, the various approaches for estimating the maximum roof displacement produced in the building vary by as much as + 25 %, to 35 %. The method with the largest variation, and the least conservative estimate, is actually the use of the average of the series of non-linear response history analyses. The two methods contained in the methodology; the displacement coefficient approach and capacity spectrum approach, produce the most conservative estimates. This apparent conservatism would appear to be the result of the way in which the various approaches treat the pinched hysteretic response. The equal displacement rule and response history analyses both neglect the effects of hysteretic pinching. Both the displacement coefficient and capacity spectrum techniques account for this effect. Although the research

Appendix A, EscondidO Village Midrise

C

il b \I

A~

-


'oof (zed,to 'fIum onse 'ory

- - ,-

/8

34 JO

r-

1

..

..

~.

1

·[X

33

25

'c>ZIl" I

I·

.,

: by two · I· ...•... ·1 ... iii' ... , . ~ ed in the 10'-7" ement .... C ",",L_',.~. . . .:. . . - - - . - - - - " - - - . ..............-"'1 8'-3e history .1.. · L ...... ~ .. - - - . B : 20 I . . . . . . . . . " .... ' . s·..•. W · ~ ... l ' .. I able 1.3-1 12' . 12' 12'_7" 12' 12' : 12' : splacemen '2' '2' and also ,ts of the Figure 2.1,1. Typical Floor Plan !ach )redicted 1:community is currently divided with regard to the 2. Building and Site importance of pinched hysteresis to overall Description Ita building response, it would seem pruden~ given the pproaches wide range of variation in the response history 2.1 Ceneral Icement analyses to take the conservative approach as has The Escondido Village Midrise buildings are a lch as been done by the methodology. Such conservatism set of five, similar, reinforced concrete shear wall ,argest is further warranted, given that our ability to structures. The buildings were constructed in two timate, is Iccurately estimate the ground motions that a phases. The first phase, designed in 1961, !ries of milding will be subjected to is quite limited. consisted of three structurally identical buildings 'he two As noted earlier, although the Methodology Abrams, Barnes, and Hulme. The second phase, ; the Ippears to provide conservative estimates of designed in 1964, consists of Hoskins ll?d . capacity lUilding response, compared to other approaches, McFarland, which are also structurally Identical to ;onservativetrofit designs developed using the Methodology each other. The two phases of construction were would ICtually appear to be quite cost effective and designed by the same designers and have nearly hich the :conomical relative to the designs commonly identical floor plans. The primary difference ysteretic Iroduced in the past using more traditional between the two phases is in the layout of and pproaches. basement areas. the effects • The buildings have overall plan dimensions of cement 65 feet by 109 feet, and are approximately niques rectangular in plan (Figure 2.1-1). They are esearch

IX

:

"W

T.

lIIage Mldr!bpendlx A. Escondido Village Mldrlse

A·7


,!

.. ':'.:i

, , " : . ' , " " " " " " .' ,'.j

•

t;.:.·_':.":~.::.:~::_·:

j•

..;.'.:":.:; ;.... .'_ .. _.

•

{COO: -060000000 0 ' :0000

A

Figure 2.2,1. Typical Floor Slab constructfon

arranged in random directions on the Stanford campus, but are all located at the northeast corner of the site, near EI Camino Real and Stanford Avenue. Each building is 8 stories tall, with a mechanical equipment penthouse and a full basement. The typical story height is 9'-1" (the basement story height is 12'-7"). The basements are only partially embedded within the ground, with the first floor located about 4 feet above adjacent grade.

2.2

structural systems and Members

Figure 2.2-2. Typical Connection Of Floor to wall

•

Continuous strip footings support walls

•

Isolated spread footings support columns

LaterallDad-reslstlng system •

Materials •

Per original design drawings, specified 28-day concrete strength: 3000 psi for slabs, beams, and walls; 3750 psi for columns

•

Per test program conducted in 1989, tested concrete strength: 2470 psi for slabs

•

Concrete strength used in analysis: 2470 psi for slabs, beams, walls; 3000 psi for columns

Gravity IDad-reslstlng system •

12" one-way concrete core slabs (7" diameter hollow cores spaced 9" apart) carry floor loads to walls and columns (Figure 2.2·1)

•

Strips of slabs aligned with column lines are solid and provide a beam-like element at the columns

•

10" concrete walls at stairs, elevators, and perimeter of typical floors

•

12" concrete walls at basement

•

15x24 interior concrete columns, 15x22 re-entrant corner columns, i Ix II balcony columns

A-a

Load-path: rigid slabs, through shear walls, to foundation

•

Specified steel reinforcing: "intermediate" (40 ksi) grade for slabs, beams, and walls; "hard" (60 ksi) grade for columns Concrete shear walls are typically reinforced with two curtains of reinforcing steel. Vertical steel is lap spliced at each floor level. Floor slabs are doweled to the wall, as indicated in Figure 2.2-2. Above the first floor level, the walls are of uniform layout in all of the buildings, as shown in Figure 2.1-1. There is a substantially larger

Appendix A, Escondida Village Midrlse

;S


umber of walls in the basements of the buildings

~an there is in the upper stories, and the two

phases of construction have slightly different arrangements of basement walls. Figure 2.2-3 shows the arrangement of typical basement walls in the first increment of buildings.

5011 and seismicity The Escondido Village is underlain be approximately 200 feet of alluvial soils over Franciscan formation bedrock. As reported in various project geotechnical reports'·'·J ..·" the alluvial soils are generally dense interbedded layers of clayey sands, sandy clays, sands, and gravels. Woodward-Clyde Consultants" developed estimates of the force-deformation relationships for shallow spread foundations, like those for the valls Escondido Village Midrise buildings, founded on these soils. These force-deformation properties, lumns that were evaluated at loading rates similar to those expected during an earthquake, are presented II in Figure 2-3.1. As indicated in the figure, an effective :ar wa s, I subgrade modulus of 800/B tons/ft2/ft is estimated. Initial stiffness of footings founded on this material is estimated as being 4K,A, where A is the area of ified 28-da the footing and J(, is the subgrade modulus. )s, beams, Ultimate permissible bearing pressures are estimated by Woodward-Clyde as being on the 9, tested order of 15 tons/ft'. It is projected that the )s foundation conditions could vary from 2/3 to . 2470 . 150 percent of the stiffness projected in the figure. : I pSI The Escondido Village Midrise buildings are orcouIDIlJlocated on the Stan.or ~ dU" . mverslty campus In nediate" (4northern California. The western border of the alls; "hardcampus along Junipero Serra Boulevard is approximately 4.0 miles northeast of the reinforcedmid-peninsula segment of the San Andreas fault, Vertical and the eastern border along El Camino Real is Floor slababout 5.5 miles northeast of the fault.

2.3

n

2.4

Building Performance During the Lama Prieta Earthquake lls are of as shown i The Escondido Village Midrise buildings were larger jamaged during the October 17, 1989 Lorna Prieta

lIIage Mldril'ppendlx A. Escondido Village Mldrlse

Earthquake. This included moderate but widespread cracking of the cast-in-place concrete walls, including both shear cracking in classic diagonal "x" patterns, flexural cracking consisting of cracks that were approximately horizontal near the bases of the walls, and horizontal cracking along the construction joints present at floor levels. The walls around the stair towers experienced the heaviest damage. Most damage to the walls was repaired shortly after the earthquake with the injection of epoxy grout.

3.

Preliminary Evaluation

3.1

summary As recommended in Chapter 5 of the Methodology, a preliminary seismic evaluation of the Escondido Village Midrise buildings was conducted using the procedures contained in FEMA-178" to determine if nonlinear analysis is warranted. The FEMA-178 evaluation procedure was developed with national consensus of the engineering community and is intended to serve as a preliminary screening tool to determine if a building is a potential unacceptable risk to life. The procedure contains a series of checklists, organized by model building type, that guide the evaluator through examination of important structural features of the building, relative to earthquake performance. In some cases, rapid approximate calculations of capacity are performed. The premise of the procedure is that most building failures in earthquakes can be traced to a relatively limited number of critical flaws, that the checklists are designed to specifically explore. Failure of a building to pass the screening test of the checklist does not necessarily indicate that a life safety hazard exists. It is expected that some buildings that fail the checklist screening can be demonstrated to be adequate to a substantial life safety performance objective upon more detailed evaluation.

A·9


- -, - ,~

-

.

---

---

-,

.

.. ,---------. --

-

---

- - -

-.~.-

r I -- - -- --r---......,I--"""i

....-t- -- J - _11_ --

~I~ _J -+-+-.~

~

.

..

10'-7~

rlr --II

... :-,' ..

....._

12'

12'

. )1:

. . .~_l_______ _ 12'

12'

12'

'2'

Figure 2.2·/1. Basement Floor Plan

16

B_

feet

.. Jolm

I'

12

i2'

.[

!

•

10 8

Note - "B" is the footing width

£, " 6

·cc

.I!

2 0

0

O.OS

0.10

O.IS

Foundation DisD]acement rfeetl

Figure 2./1·1. $011 Force·Deformatlon Relationships

A-10

Appendix A. Escondido Village Mldrlse

f

,

,

~---------------------------------------

-


G)

, I"'"W20

(0 0

(0

0

G)

G)

W23

-

I

...1

~1~r~1

-.

><

,

-

;..

W1.1:.

-

-

.

L ___~.....I_wiii,·..'_ _j,---~,___ . ., '

12'

- -

12'.r

12"

12'

'

12'

1

W26

10'·7"

iI

-ill

-

-

--0

W41

Stair #2 .

-. - - .. - - -iII- -ill - -. --

GD

G)

_J

J. ,

i5
W31

12'

0

- - -iI

_ _ __ " _ 12'_7" '

'2'

'2'

Figure 3.2·1. Typical Floor Plan

It should be noted that the ground motion criteria used in FEMA-178 is substantially less than projected by Woodward-Clyde for the Stanford campus. Therefore, FEMA-178 may not be suitable for use as a life safety hazard screening tool at this site. This evaluation was performed for the Escondido Village Midrise buildings primarily to ensure that all important critical flaws were identified, prior to proceeding with more detailed analyses. For the Escondido Village Midrise buildings, the FEMAc 178 Evaluation Statements identified a number of deficiencies in the original design. Primary concerns include: • Vertical discontinuity in major shear resisting elements

•

Inadequate boundary reinforcing in shear walls

•

Inadequate overturning resistance of foundations.

3.2

FEMA·178 Evaluation statements The FEMA-178 methodology utilizes a series of Evaluation Statements that can be answered as true or false. Typically, these statements are based on qualitative issues regarding the building's construction. In some instances, limited calculations are performed to determine the appropriateness of a statement. An answer of false typically indicates a potential seismic deficiency. For this study, FEMA-178 Evaluation Statements are used as a preliminary evaluation tool in accordance with Chapter 5 of the Methodology. No detailed analysis was performed to verify potential deficiencies. No testing of materials was performed. Non-structural elements were not considered. This evaluation is based on review of the original structural drawings and a site walkdown. False FEMA-178 Evaluation Statements indicating potential seismic deficiencies include:

General Weak Story. The story strength at each story is at least 80 percent of the stories above,

Ullage MldrllPpendlx A. Escondido Village Mldrlse

A-n

SEISMIC EVALUATION ANIi RETROFIT OF CONCRETE BUILDINGS

8tbfloor

7tbfloor

-i!-6th floor

TypicallrUllverK wall (Walll3t and 41)

sthnoor Typical comer wall

4tbfloor

3rdfloor ~-

n'....'tinu·"" at lhear 41

2ndfl"",

Buemcnt

n... ....... Figure S.2·2. Discontinuity at Transverse Shear Walls (Walls S1 and 41J

however, there are local discontinuities in some of the vertical elements of the lateral force resisting system. These are located at stairways #1 and #2 and at the primary shear walls along lines 1 and 10, designated as W31 and W41, respectively in Figure 3.2-1. Figure 3.2-2 presents an elevation of walls 31 and 41, indicating the discontinuity condition that occurs at the first floor level in these walls. The effect of this discontinuity is to create a severe condition for the boundary elements of these walls. Figure 3.2-3 presents partial plans of stairway #1 at the basement, first floor and typical floors. The primary lateral load resisting components of this stairwell core are designated as walls "a", "b", "en, "d" and "e". Wall "a" is offset below the second floor and walls "CO, "d" and "e" have large door openings in the mid length of each wall at the basement level. The discontinuity of wall "a" is not believed to represent a severe problem because the return walls "c" and "d", that serve as the flanges of wall "a" under flexural behavior, are continuously connected to the wall above, and are continuous themselves through the zone of

A-12

discontinuity. Therefore, it is believed that an adequate load path exists across this discontinuity. The openings in walls "c", "d", and We" are not considered significant because there are an extensive number of additional shear walls present in the basement, and the portions of these walls that are removed are not critical to the flexural behavior of this element. Figure 3.2-4 is a plan of stair way #2. Primary walls resisting lateral load are indicated as walls "f", "gH, "h", "iH and "j". Wall "f", along column line 6 is discontinuous at the first story, where it is replaced by a column at grid coordinate D-6. This represents both a shear and flexural discontinuity, but is primarily a concern because of the flexural condition. The column at D-6 and boundary element at B-6 must resist all of the overturning demands delivered by wall "f" above. • Vertical Discontinuities. As described above, under "Weak Stories", there are three conditions of vertical discontinuity - the transverse walls (W31 and W41), stairway #1, and stairway #2. •

Deterioration of Concrete. Many of the floor slabs have horizontal cracks present. These cracks appear to be a result of drying shrinkage of concrete that was cast too wet.

•

Concrete Wall Cracks. The buildings experienced significant cracking in the Lorna Prieta Earthquake of 1989. Nearly all such cracks have been repaired with epoxy injection, except at the basement, where some cracks with widths as much as 4mm width were observed. It is not believed that these cracks are detrimental to the building's future behavior, however.

•

Complete Frames. The concrete shear walls resist a significant portion of the building's total weight.

•

Shear Walls Shearing Stress Check. The maximum calculated stress in the walls, when the

Appendix A. Escondido Village Midrise

App.

, ~

--------------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

,"', ',;!)

an mtinuity, are not

it

,i'

Wall"b"

C~) .

(~l

Wall"b"

(~;

Wall 'd'

:j':

'd'

Wall

'e'

Wall"a"

Wall

an

an

'e'

an

e'

c'

c'

Wall "a"

(~l

i,~)'

Wall "a"

~

1.,

T~!:'

is walls' ong t story, coordinru ' !xural because D-6 and of the ;'f" above Jed above 'ee . the :airway #1

Basement

Aoor

Figure ~,2-~. Partial Floor Plans at stairway NO. 1

(if

~

(~

(7 All .".

Wall

'r

Wall 'g'

~.

I

Wall

Wall "g"

'h'

Wall "h'

all . "f'

,~.
Jf the flrnJ t. These 19 too wet.

TYPical Floor

Wall 'I'

~

all

g'

Wall "h'

'1' .-I' .,,'1 LL-J. Wall"i"

Basement

'"

FI(ItI(

Figure ~.2·4 partial Floor Plans at Stairway NO.2

building is evaluated in accordance with the FEMA-178 Quick Check procedure is.125 psi, which is substantially in excess of the 50 psi specified as the limiting value. However, the walls are well reinforced for shear and the computed value is well within ACI 318 limits.

•

;hear walls ,uilding's

,imum

\!: :,~,'

Wall

'c'

t. Primar

ngs the Lorna all such ,xy Nhere som n width lat these ing's futu~

+

Wall"b"

Wall

Is presen ! walls !xural

"

f,',~;

Wall 'd'

In

/-',

(~)

Overturning, Many of the building's walls have slenderness ratios that substantially exceed the indicated amount, with the ratio on some walls approaching 10: 1. Major walls of the building in the transverse direction have a ratio of 5: I, while walls in the longitudinal direction have a ratio of 3.5: 1. Walls are

IlIlage Mldr/Ppendlx A. Escondido Village Mldrlse

provided with special boundary reinforcing for overturning demands. •

Coupling Beams. Coupling beams are generally poorly reinforced and have no stirrups.

•

Column Splices, Longitudinal reinforcing steel in wall boundary elements are spliced with 24 diameter lap lengths. These are not adequate to develop the reinforcing steel. Bar splices are staggered with not more than 50 percent of the bars spliced at a given location. Therefore, the effective bar splice in boundary elements is equivalent to the strength

A-15


of 50 percent of the longitudinal reinforcing provided. (This was an assumption made during the initial evaluation, see Section 4.3.4 for additional discussion.) •

•

•

•

Reinforcing Steel. The typical reinforcing pattern for walls provides a ratio of 0.0023 times the gross cross sectional area. Reinforcing at Openings. Trim bars are typically provided at openings, however, these are not confined with special ties.

Plan Irregularities. Re-entrant corners occur at column locations B-3, B-7, G-4, and G-8. Special chord bars have not been provided in these areas. However, the distribution of shear walls throughout the building is such that diaphragin flexural demands are moderate and slab reinforcing is generally adequate to handle corner stresses. Transfer to Shear Walls. Dowels provided between the floor slabs and walls are not adequately embedded to fully develop their yield strength. Consequently, the connection of diaphragms to walls cannot develop the diaphragm strength. In addition, most walls do not extend the full length of the diaphragm, and collector reinforcing has not been provided to drag diaphragm loads into the walls.

Vertlcalcor.nponents •

•

Conrmement Reinforcing. Ties provided at boundary elements of shear walls are #3 at 12 inch spacing. However, ties are provided with 1350 hooks, so that confinement could be considered of intermediate quality.

Diaphragms •

capacity of Foundations

Shear Wall Boundary Columns. As previously described, the lap splice of wall boundary reinforcing is not adequate to develop the reinforcing strength.

A·'4

3.3

Overturning. The ratio of the effective horizontal dimension, at the foundation level of the seismic force resisting system, to the building height exceeds 1.4Av . Neglecting near-source effects, A. for the Stanford campus is 0.4 and the ratio is 0.56. In the transverse direction, the ratio of foundation width to building height is 0.38.

he thl bu

thl

on

m:

an re

Elastic Analysis

Elastic analysis is the conventional method of evaluating the seismic demands on elements of a structure used in both design of new structures and detailed evaluation of existing structures. For this project, a dynamic response spectrum method analysis was performed. Using the ETABS21 software package, a three-dimensional computer model was constructed and analyzed. The resulting displacements are a reasonable estimate of those that the real structure would see, if it remained elastic. Forces calculated for individual elements by this technique are also a reasonable estimate of the maximum demands on these elements if the structure were to remain elastic. The primary benefits of the elastic analysis is that it provides a rapid method of determining the strength of the building relative to current code requirements, the distribution and locations of large strength demands on the structure, and the overall level of lateral displacement the building would experience in the design earthquake. Buildings with limited displacement demands, well distributed elastic strength demands, and relatively moderate conditions of strength deficiency relative to current code can generally be judged to provide acceptable performance. On the basis of the elastic analysis, using cracked section properties and accounting for elastic flexibility of the foundation system, the Escondido Village Midrise buildings are demonstrated to have strength in the longitudinal direction, comparable to that required by the current UBC. Strength in the transverse direction,


4.

Iy m

bl M

is d)

In fo

pI

e~

W su

d(

m m

of

stl

ei

m

bI di th di

th su Sl

in

n(

df

~

-------------------------------------------------------------------------------SEISMIC EVALUATION ANO RETROFIT OF CONCRETE BUILDINGS

--------------------------------------------

e

level the :ting d the lation 1

.0

however, is substantially less than that required by the current UBC. The effective periods of the building are 0.83 and 0.71 seconds respectively in the transverse and longitudinal directions. Based on the elastic response spectrum analysis, maximum elastic interstory drifts of l. 2 percent and I percent are estimated in these directions respectively.

4. ethod .of nts of a :tures ali For this :thod IS" ,mputer e resultin )f those nained

4.1

Evaluation by Product 1.2 Methodology Introduction

A series of simplified inelastic analyses of the type known as static pushovers were performed to more accurately evaluate the behavior of the buildings in response to strong ground motion. Most design of buildings for earthquake resistance is based on an elastic analysis of the building's dynamic response to the expected ground motion. In such analyses, it is assumed that the amount of force induced in an element is directly ~lements stimate (j proportional to the amount of deformation it .s if the experiences in response to the ground motion . While all buildings behave in this manner when nalysis is subjected to low levels of loading, most structures nining the do not have adequate strength to respond in this nt code manner when subjected to intense levels of ground .ons of motion. In reality, when subjected to such levels , and the of ground motion, individual elements of the building structures will be stressed to a point at which they Ike. either yield - that is continue to deform while nands, we maintaining a relatively constant stress state, or ,d relativel break. Following such yielding or breaking, the ncy relatil distribution of both deformations and stresses I to provid throughout the structure can be significantly different than predicted by an elastic analysis. . using Elastic design and analysis procedures, such as those contained in FEMA-178 incorporate ing for [em, the substantial factors of safety in the permissible re stress states and configuration limits they specify, mgitudinal in recognition of the fact that the elastic analysis is by the not accurately predicting the distribution of ;e directiordemands at high load levels. Inelastic analyses,

'lIIage Mld~pendlx A, Escondido Village Mldrlse

such as that outlined in the Methodology allow for more accurate prediction of the demands on individual elements of the building and therefore permit lower factors of safety to be used in evaluating the adequacy of specific structural components. Many buildings that appear to be highly deficient when evaluated by elastic analysis methods can be demonstrated to be only modestly deficient, or perhaps completely adequate, when evaluated to these more accurate approaches.

4.2

Analysis Methodology

The static pushover technique is one of the simpler types of inelastic analyses. Essentially it consists of a series of elastic analyses of successive models of the building that have been progressively modified to represent the stiffness of the structure at a given stage of lateral deformation. In other words, as structural components yield, the stiffness of the structure is reduced to reflect that yielding. For the example building study, tl1e following basic steps were implemented based on the Metl1odology: • Structure modeling (Section 4.3 of this report) •

Pushover analysis (Section 4.4)

•

Performance point (Section 4.5)

•

Performance assessment (Section 4.6)

4.3

Structure Modeling

4.$.1 Software Limitations The static pushover analyses of tl1e Escondido Village Midrise buildings were performed using DRAIN-2DX software. As with any software package, limitations can significantly affect the nature of the analysis. Some of the limitations imposed by the DRAIN-2DX software include: • No Inelastic Panel Elements. Walls subject to potential flexural and shear yielding were modeled as column elements. See Section 4.3.4.

A·15


W23 Stair #1 W16

W20

I>"?J • • •

I W31

WIO

W41

• • •

I

•

•

WI7

Stalr#Z

WI3

FIgure 4.$-1. wall Element Numben

•

No Degrading Elements. All yielding elements maintain their strength, but the Methodology requires degrading for elements with high ductility demands. See Section 4.4.2.

•

Two-Dimensional Modeling. The program allows for two·dimensional modeling only; resulting in the loss of torsional effects.

•

No Graphics or Post-Processing. This limits the efficiency of the analysis.

4.$.2 Materials The same material properties used in the elastic analysis (see Section 2.2) were used for the nonlinear static analysis: • Existing Concrete Strength. 2470 psi for slabs, beams, walls; 3000 psi for columns •

Existing Steel Reinforcing Strength. 40 ksi for slabs, beams, walls; 60 ksi for columns

4.$.$

structural systems

DRAIN 2DX is capable only of analyzing two dimensional structures. Therefore, independent analyses of the building response were performed for the longitudinal and transverse building axes, using different models. Figure 4.3-1 is a typical floor plan for the building, indicating the numbering scheme used for various walls

A-'.

contained in the buildings. Figure 4.3-2 schematically represents the model developed for the longitudinal axis of the building. The principal disadvantage of using two-dimensional models to represent the building is that torsional effects are lost, as are the combination of effects from simultaneous loading in different directions. The elastic analysis, previously performed, demonstrated that the building is torsionally quite regular. Therefore, it was not felt necessary to model its torsional response characteristics. Modeling of the effects of combined response in two directions on those elements of the lateral system which participate in both directions could not be captured. In addition to the inability of the two dimensional approach to capture this behavior, it was not possible to develop constitutive models (force-deformation curves) for the infinite number of combinations of loadings about the two axes of these walls that are possible. As seen in Figure 4.3-2, the longitudinal model essentially consisted of 7 stick type sub-models interconnected at each floor level by rigid translational links. Each stick represents one or more vertical elements of the lateral force resisting system. Individual sticks were provided to represent each of the major shear wall


co

cal

sti, W the reI f01

co an 26

we

sui ch the #2

re~

reI e1c fie in be

•

--------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

i:,"

.

W13

U .

;- W23 _

..

WLO , W20

WL7 W26

~~-l!~ ~-

~,

flexural elements

'-1

shear elements

, J• •-' "

W70',

•~;,' ,

Roof

Frame

8th 7th 6th

5th

p,

1c"

4th 3rd 2nd 1st

- Foundation soil springs

Figure 4.~·2. Non-linear Model-Longitudinal Direction

JUilding

configurations contained in the buildings, categorized by response direction. Thus, a single loading stick was provided to model both walls W13 and s, W23, the long rectangular walls along the sides of the the building (Figure 4.3-1). Since the stick efore, it represented two identical walls, the Lal effects d force-deformation relationship for the stick consisted of a composite of both walls stiffness lose cipate in and capacity characteristics. Walls 10, 17, 20, and lddition 26, all of which are identical "L" shaped walls, )roach to were combined into two different stick sub-models, each sub-model representing the to of these walls when pushed in either characteristics lation the positive or negative direction. Stair #1, Stair ations of. ; that are #2, and the elevator core (W70) each have unique response characteristics and were provided with separate models. nal As shown in Figure 4.3-2, each of the sticks representing the shear walls is comprised of two :vel by elements at each story. One element represents the :ents one~ of the walls and is infinitely rigid flexural behavior ,rce in shear. The second element represents the shear rovided behavior of the walls and is infinitely rigid in

Ie Mldrl'

AIIpendlx A. Escondido Village Mldrlse

flexure. Each stick is also provided with a rigid beam a,t its base, supported by a series of inelastic soil springs. The soil springs are preloaded with the calculated dead load soil pressure under the foundation and are set with compressive spring rates. The springs have null tensile stiffness. Stair #2 has the additional complication of the vertical irregularity at the first story, previously described in Section 3. This was modeled by using altered flexural stiffness properties for this wall at the first story. A horizontal linear translational spring is attached to the model at the level of the first floor. This spring represents the shear stiffness of the numerous additional concrete wails present in the basement story of the buildings (Figure 2.2-3). The value of this spring was calculated as the difference in stiffness of the basement story of the building in the linear elastic ETABS model, and the DRAIN model constructed without this spring. A final sub-model stick was provided to represent the stiffness of the concrete frame (beams and floor slabs) and the smaller walls within the

A-17


Roof

8th 7th 6th 5th 4th 3rd 2nd

1st

- Foundation soil springs

Figure 4.8·8. Nonlinear Model· Transverse DirectIon

building that do not contribute significantly to the lateral load resistance. The initial stiffness of this stick was chosen such that the initial stiffness of the entire inodel matched that of the elastic analysis model. Based on evaluation of the elastic analysis results, a lateral deformation was selected for each story at which flexural yielding of the frame would commence. This information was than used to construct an elastic-purely plastic representation of the frame stiffness at each story. The model for the transverse building response was constructed in a similar manner to that for the longitudinal direction. A schematic diagram for that model is presented as Figure 4.3-3. An important difference between the two models is the way in which the discontinuity at the base of the main transverse walls was handled in the transverse model. This problem was previously discussed in Section 3.2 and illustrated in Figure 3.2-2. At this discontinuity, the wall boundary elements are the only continuous components. The behavior at this discontinuity

A-'8

was modeled by running two columns, each representing the boundary element properties of the wall, through the basement and first stories of the building. These boundary element columns were linked together by a rigid beam at the underside of the second story. The rigid beam was provided at the underside of the second story since the first story wall would not be completely effective due to the discontinuity below. To illustrate, Figure 4.3-4 shows an assumed effective axial zone of the first story wall panel relative to the door and louvre openings at the basement level. Because of the modeling of this discontinuity, the transverse model was judged to be slightly more flexible than the real structure, but of adequate accuracy to investigate the concentrations of demands likely to occur in the real structure at this area of discontinuity. It should be noted that coupling beams between the main transverse walls and comer walls were not modeled. From the elastic analysis, it was

fo an Cc

co re

4.

cb us pe

m se sp df

to se bf

w aI Vl c(

fr

Pl c(

Appendix A, Escondido Village Midrlse

•,

--....

---

--....

-------------------------------------------------------------------------------------


transverse wall effective '. " area of wall -t------H~Jl...

/

i,

\ :

2nd floor

comer wall, typo 1st floor

basement wall

Foundation louvre opening

door opening

vertical shear crack forms at wall, typo

Figure 4.S·4. Effective Axial zone at Discontinuous Transverse wall

i1 .es of lries of mns

found that these beams would be highly stressed and would fail early during a severe earthquake. Consequently, these beams were judged to contribute minimally to the building's lateral-load resistance.

4.So4

structural Elements and components

Wall Element Flexural Properties. Flexural ,am was: characteristics of wall elements were determined Iry since, using the software package BIAXI7. This software permits the development of non-linear y moment-curvature relationships for concrete o effective' sections of arbitrary cross section, subjected to specified axial load. The program was actually tive to developed for use in analyzing columns as opposed ent to walls and incorporates the Euler assumption that sections that are plane prior to initiation of dged to·. bending remain plane after bending. Since the cture, walls in the Escondido Village. Midrise buildings are quite slender, this assumption is thought to be in the valid. The program has several concrete compressive behavior models programmed into it, s bet'lVeri. from which the user may choose. These include s were parabolic stress-strain distributions for both was confined and unconfined models.

Ie Hllrlrl,;' Appendix A, Escondido village Mldrlse

For this project, moment-curvature curves were generated based on an unconfined model with an ultimate compressive strength (f' c) of 2470 psi, matching the findings from previous testing conducted at the buildings. The comer L-shaped walls, elevator C-shaped walls, and stairwell walls were each modeled as complete walls with entire flanges assumed effective. All concrete was assumed to be unconfined. As previously described in Section 3.2, the splices of boundary reinforcing for the shear wailS are inadequate to develop the tensile strength of the bars. General notes on the original construction documents indicate that lap splices in continuous bars should be staggered. It was judged, therefore, that 50 percent of the longitudinal boundary bars would be fully effective in tension at any horizontal section through the walls. Therefore, in the BIAX models, only 50 percent of the boundary steel was incorporated. Assumed strength of the steel is 40 ksi, based on the notes contained in the drawings. It should be noted that there was some uncertainty with regard to these assumptions. Lap splice details for chord reinforcing in walls are not

A-1.


300000 -+-Iongitudinal Mlls ~ comer .:lis

1-.

-M- comer 'Mll1-X .....,&- elevator -4-ltair#l +:1

_ 200000

-+-Itair 4H -x -B-Itair #2 +x ___ .tair #2 -x

"L.~!"

~

;! 150000

Looooo '0000

o~--~-----+----~----~----+---~-----+----~

O.OOBofOO

2.00E-04

4.00&.04

6.00E-04

8.00B-04

1.00E·03

1.20E'()3

1.40E-03

1.60E·03

Curl'8ture [radlaa5llncb}

Figure 4.11-5. longitudinal Moment-curvature Relations lOr First Floor wails

specifically shown on the drawings, while column splices are. In details for column reinforcing splices, all of the bars in a column are lap spliced just above each floor level. There was some possibility that the boundary steel for the shear walls was spliced in a similar manner. This would result in lower flexural capacity for these walls. There was also some uncertainty with regard to the strength of the reinforcing used for the boundary elements of walls. The general notes on the . construction drawings indicate that Intermediate Grade steel, with a yield strength of 40 ksi, was to be used for all reinforcing except longitudinal column bars, where Hard Grade steel, with a yield strength of 60 ksi was specified. It was possible that the Hard Grade steel was also used for the boundary elements of walls. In such a case, the lap splices provided for the bars would be even less adequate. After the completion of our analyses, it was subsequently learned, through x-ray photography and minor destructive testing, that the boundary element reinforcing had lap splices just above the floor level. Chemical and tensile testing also confirmed the reinforcing to be Intermediate Grade steel. It should also be noted that BIAX tends to under-estimate the flexural stiffness of elements

A-20

with minimal steel reinforcing. Professor Jack Moehle at the University of California, Berkeley, recommended that the initial effective stiffness of the wall elements be one-half of the gross sectional value. Based on the moment-curvature relations from BIAX, the initial effective stiffnesses were generally on the order of 25 percent of the gross sectional value. Figures 4.3-5 and 4.3-6 present the moment curvature relationships for each of the major walls of the building, acting respectively in the longitudinal and transverse directions of the building. These curves are based on the assumption of 40 ksi boundary steel with staggered lap splices. The curves were computed for the dead load axial stress condition at the base of the walls. They have been terminated at peak concrete compressive strains of 0.005, as suggested in the Commentary of Section 9.5.4.2 of.the Methodology. Examination of the curves for the longitudinal direction (Figure 4.3-5) indicates that the primary lateral load resistance for the structures in this direction is provided by the main longitudinal walls (W13 and W23, Figure 4.3-1), and the walls around stairways # 1 and #2 and at the elevator core. The wall at stairway #2 has substantially greater strength and deformation capacity in the

Appendix A, Escondido Village Mldrlse

+ di: de pr rei

60 pr m:

the co ea on stl

is the \ir ac

Ie' 10

w,

ca as:

fu 51 afl

-----

•

------------------------------------------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS ------------------------------------------------------------------------------------

.......... transverse walls -+- comer "'03.1.15 +y ",","*,,-comer walls-y -6-eievator +y

400000 350000

- . - elevator-y

-+- stair #1 -+- stair -#1

t~:..:-a-a-~ d2S0000~ 1300000

~

200000

~

150000

-y +y

-B-nair-#2+y _5tair#2 -y

1l

:Ii

o~----~----~------+_----~----~------+_----~ O.OOE..oo

2.00E-04

4.()(JE.Q4

6.00E-04

8.00E-Q4

l.OOE-03

l.20E·03

1.40E·Q3

Curwture [radlansllncb.]

Figure 4.~·6. Transverse Moment·curvature Relations for First Floor Walls

ack :keley, less of ;ectional .tions were gross

+X direction than in the -X because of the discontinuity in the first story, previously described. The two main walls (W13 and W23) provide more than 50 percent of the total lateral resistance in the + X direction and more than 60 percent in the -X direction. In the transverse direction (Figure 4.3-6) the primary lateral resistance is provided by the two )ment or walls main walls (W31 and W41, Figure 4.3-1) and by the walls around the two stair wells. The configuration of the stairwell walls is such that Ie each stairwell has substantially more resistance in ;taggered one direction than the other. Stairwell # 1 is strongest in the + Y direction, while Stairwell #2 : the is strongest in the -Y direction. l of the Manual calculations of the shear capacity of concreW :I in the the walls indicated that they are, in general, limited by the shear friction capacity of the walls across the construction joints present at each floor gitudinal level. Typically, this capacity is approximately primary 10 percent less than the nominal capacity of the walls derived using UBC formulas without 1 this capacity reduction factors (cp). It was arbitrarily !inal the walls assumed that a 114 inch displacement is required to fully mobilize the shear friction strength. A ,vator . 5 percent strain hardening factor was permitted ltially r in the after attainment of the 114 inch initial slip.

Ige Mldrll', Appel1C1lx A, Escondido Village MIClrise

It should also be noted that by using BIAX, the modeling rules for flexural properties in shear walls presented in Chapter 9 of the Methodology were ignored. This is allowed per Commentary in Section 9.5.1 of the Methodology. Wall Element Shear Properties. As noted above, each wall element in the DRAIN model was built with two elements - a flexural element and a shear element. Shear properties used in the DRAIN model were computed based on the shear capacity of the wall as calculated per ACI 318. Although shear friction capacities per ACI 318 were typically less than wall shear capacities, shear walls generally do not fail at their construction joints when sufficient dead loads are applied to the walls. Strain-hardening was not included in the modeling of these shear elements. The inclusion of strain-hardening would have slightly increased the overall shear capacity of the building, but not the deformation capacity. By not including strain-hardening, we could more easily account for shear degradation in the wall elements. (See Section 4.4.2 for discussion on shear degradation.) Foundation Rotational Stiffness. Non-linear springs were used to model the rotational stiffness of foundations beneath the major shear walls. Initially, the stiffness assumptions provided by

A-21


, Linear !tOil tprinp

.. "",......on LoDgIh

~;=Jm.. DdI ..........

FIgure 4.$-7. DetermlnatiDfl OF Effective Foundation length

Woodward-Clyde were used as the basis of the model. Finite vertical translational soil springs were incorporated into the model. Each spring represented the stiffness of a length of the wall foundation, equivalent to its width. Beams, with springs representing the foundation stiffness were provided beyond the width of the shear walls, with linear elastic properties corresponding to a section comprised of the basement walls, the strip foundations and a portion of the first floor slab. The effective width of the first floor slab was taken based on the limitations for flange widths in "T" beams contained in the ACI code. The length of the foundation systems effective in resisting shear wall overturning was taken based on independent, beam-on-elastic foundation type analyses. These analyses are schematically represented in Figure 4.3-7. The effective foundation length was taken as the point at which foundation uplift was produced beyond the compressive side of the shear wall. In performing visual surveys of the buildings, it was noted that some of the basement walls have vertical cracks through them. It was surmised that these cracks may be the result of shear failures, induced in the walls by the Lorna Prieta Earthquake, as they attempted to spread overturning demands from the shear walls.

A-22

Therefore, the computed shear capacity of these basement walls was programmed into the DRAIN model to simulate this failure mode. As determined by the DRAIN analyses, initial inelastic behavior of the structure was dominated by foundation rotation and liftoff effects. It was expressed by Stanford Facilities Management that the stiffness suggested by Woodward-Clyde for the soil springs appeared to be significantly larger than revealed by previous plate load test data for various locations on the campus. Therefore, a series of sensitivity analyses were performed in which the spring stiffness and ultimate capacities of the soil compression springs were evaluated for 150 percent, 67 percent and 25 percent respectively of the values suggested by Woodward-Clyde. It was found that these assumptions had negligible effect on the overall behavior of the model. The predominant factor in the inelastic behavior of the foundation, as predicted by the model is the liftoff of the foundation on the tension side. This appeared to be independent of the compression spring stiffness assumed. The total effect on structural elements of the model, for the various assumed soil stiffness properties, was a change in demands of approximately 2 percent. Therefore, it was concluded that the structure's behavior is insensitive to the spring stiffness of the soils beneath the foundations, but is quite sensitive to the ability of the foundations to rock about their bases.

4.4

pushover Analysis

T d

cl 4 tl d

P CI

b

~

tl d u b a: fl IT

tJ n d p

fc sl

P b c

c

d TI

Deriving and Applying Pushover FOl'Ces Per Section 8.4 of the Methodology, the Escondido Village Midrise buildings were evaluated based on a Level 3 pushover analysis. Level 3 is prescribed as the basic level of analysis for the Methodology. Lateral forces are applied in proportion to the product of story masses and first mode shape of the elastic model of the structure.

4.4.1


C SI

o TI

[

s· p

c c

--

~-~==~-----------------------


--

This distribution is obtained from the initial dynamic DRAIN-2DX model in each direction. Although the fundamental mode shape will change due to changing stiffnesses (see Section 4.4.2 below), the initial load distribution was used throughout the analyses for each respective direction. Changing the load distribution would probably yield a slightly more accurate pushover curve. However, the overall effect on the building behavior and evaluation results was judged to be negligible. The effects of higher modes on the structure were also ignored in this study. Evaluation of these higher modes may result in additional damage to the building that was not discovered by using just the fundamental mode. However, because the building is relatively regular in plan and stiffness (with the exception of the basement floor), higher mode effects are anticipated to be minimal.

~se

AIN nitial ated vas

t that for the ~r than

a I in :ities ~ed for

:rall ;tor in

ed to be ness Jents of Jness

Is ive to : their

Shover, : he alysis. , analysis( .pplied in; and first; ructure.! }

Model Degradation 4.4.2 Element properties can be characterized by a typical elastic-plastic force-deformation relationship with strength degradation at high ductility demands as shown in Figure 4.4-1. As previously indicated, the flexural force-deformation relationships for the concrete shear walls were obtained using the software program BIAX. For walls exhibiting ductile behavior with strain-hardening, force-deformation curves were terminated at a peak concrete compressive strain of 0.005. For walls with degrading strength at large rotations prior to reaching a concrete compressive strain of 0.005, curves were terminated at the point of initial strength degradation. As a result, we constructed Our own component force-deformation relationships that were implemented into the DRAIN-2DX models. Unfortunately, DRAIN-2DX does not have strength degradation capabilities built into the program. Consequently, the continuous pushover curves shown in Figures 4.4-5 and 4.4-6 were constructed from a series of incremental pushover

ge Mldrt!'; Appendix A. Escondido Village Mldrlse

~Yield

r

residual strength

Deformation

FIgure 4.4,1. TYpical Force-Deformation Relationship for Model Elements

curves. Each increment was defined at the displacement that a critical element reached its degradation point. The degraded element would be replaced by a similar, weaker element (with a new yield strength that was 20 percent of the original yield strength per Methodology Table 9- 10). With this new element, the pushover analysis would then be started again and continue until the next critical element reached its degradation point. In addition, it should be noted that not only was the strength of the degraded element reduced to 20 percent of the initial undegraded element, but the degraded stiffness was also similarly reduced to 20 percent of the initial. The Methodology provides no quantitative guidance with respect to post-yield shear stiffness, axial strength, axial stiffness, or degradation rate as a function of ductility demand. In the case of our building models. consideration of a ductile model with no strength degradation would have overestimated the maximum pushover base shear by less than 10 percent. Although this is not significant, the implementation of a degraded model, per requirements of the Methodology, would more accurately determine a building's seismic behavior. In some buildings, the effect of degradation may be significant.

A-25


r

2!iOO

2000

-

hingefi ~tion

at floor

7

~

. /

,

o

/ \

/

"-

- hmge fonnation at basement waDs

,

o

20

10

bor DI.pJ.cement. d [Incher]

Figure 11.11,2. LDngltudlnal PushDver Curve fDr EXlst/ng structure 2000

lr

ex

1>00

I 1000

/

! '00

o

I

I

I

~

~

,

I

;inge !ormati n'l floor beams

\

I,i

~ hinse fomation ~lstnoor and buem:nt w:

I

1/ o

II 10

I'

20

25

to su th pt sh re re.

Roor DllpI.CUltDt. d [IDdlel]

Figure II.II·S, Transverse PushDver curve fDr EXisting structure

Pushover Force-Displacement Curve Figures 4.4-2 and 4.4-3 show the pushover curves for the existing (unstrengthened) Escondido Village Midrise buildings, when pushed in the longitudinal and transverse directions, respectively. As can be seen, the first critical events consist of hinging of floor beams throughout the frame. This is considered 4.4.S

A-24

potentially life threatening because of the lack of adequate development of the bottom reinforcing of the beams through the beam column joint. Hinging of the beams - first in positive flexure and on the return cycle in negative flexure - will result in formation of a vertical crack through the beam column joint. Following such behavior the floor systems would rely on the catenary behavior of the


sh ro po ob ca· tho co aI ex to be, di!

---------

r "

-----------------------------------------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS ------------------------------------------------------------------------------------

r 2500

hinge fonnatio at floor beams

-~

2000

~

~ 1500 :;.

j

!

1000

500

o

/ t! o

I

•

\

./

Y{J

'01

~"

.

compression failure at "toe" ~ ofbaserrent orner walls

/ \

compression failure

Lhing fonnation at bas, frent walls

baserrent stairs #1 a

10

5

/ I ~#2 J

15

20

Roof Displacement, d [inches]

Fll/Ure 4.4·4. Longitudinal Pushover curve For strengthened Building

top reinforcing steel in the beams for vertical support. However, because there are no stirrups in the beams, there is potential for this top steel to pull free of the slabs, resulting in floor collapse. In addition to the hinging of floor beams, shear failure of first floor columns occurs at relatively small roof displacements. This also results in significant collapse hazard. Because the beam hinging and the column shear failure mechanisms form at relatively small roof displacements (3.5" to 4.0"), a performance point as defined by the Methodology cannot be lack of obtained since the demand spectrum and the orcing oI capacity spectrum do not intersect. This indicates . Hinginl' that the structures, as they are, present significant Ion the collapse hazards when subjected to the demands of a large magnitude earthquake. Furthermore, the Iltin existing structure does not present a good example beam to evaluate the procedures of the Methodology e floor 'ior of tIt, because of the high collapse potential at small displacements. Consequently, for the purposes of


this example building study, it is more instrumental to follow the Methodology using the life-safety retrofit concept. To create a more stable structure and allow the pushover analysis a chance to develop some ductility, the problems of the hinging beams and shear critical columns were initially addressed. The retrofit concept is discussed in Section 5 of this report. For the purpose of continuing our discussion of the pushover curve, assume that the hinging of floor beams and the shear failure of first floor and basement columns are adequately addressed with structural upgrades. Figures 4.4-4 and 4.4-5 present the pushover curves for the strengthened building. Significant events in the progressive lateral response of the building are annotated on the figures, and more fully described in Tables 4.4-1 and 4.4-2. Critical events listed in the tables are indicated in italics.

A-25


2000

rocking oH undation at stair II 1

7

hinge fonmtion at floor bearna J500

v:r

-0; ""l"""'"

i

i

JOOO

! sao

a

1/ o

~

/'ompreSSion til ure at .•un"

mpression failure at

b~ement transverse" aDs

W~

\ \

ion failure at base_ t elevator core co~res

'-- shear failure of .. emont and 1st floor inteno coluIIIlS

/

II>

-----Ie

V-

hin e fonnation at bas e lOnt and 1st floor walls 5

JO

J5

20

25

Roof Dlaplac:ement, d [lnche.]

Figure 4.4·5. Transverse pushDver curve IDr strengthened Building

4 4

SI

11

compression failure at "toe" of basement corner walls

12.84

te

E S

A-2G


A

r

-------------------------------------------------------------------------------------

~


---- ------------------------------------------------------------------------

I

Table 4.4,2 Transverse Pushover Events

.... ...........

. ·····EVent<·. ....

4.5

•.

...

.

.Description·

. .

/toOl p/sP!acement (inCheS)< ..

I.

1

hinge formation at basement stair #2

1.55

2

hinge formation at basement stair #1

1.80

3

hinge formation at 1st floor elevator core

2.00

4

hinge formation at basement corner walls

2.29

5

hinge formation at 1st floor corner wailS

2.45

6

hinge formation at 1st floor stair #2

3.21

7

hinge formation at 3rd·8th floors beams

3.78

8

rocking at transverse walls

4.04

9

hinge formation at 2nd floor beams

4.10

10

shear failure of 15x241st floor colUmns

4.23

11

shear failure Of 15x221st floor colUmns

4.90

12

hinge formation at roof beams

5.10

13

shear failure of 15X24 basement COlumns

5.15

14

Shear failure of 15X22 basement COlumns

5.96

15

shear failure of 1st floor transverse wailS

6.10

16

hinge formation at basement transverse walls

6.81

17

compression failure of basement transverse walls

10.11

18

hinge formation at 2nd floor stair #2

10.11

19

rocking Of foundation at stair #2

12.01

20

shear failure of 11x11 basement colUmns

13.73

21

rocking of foundation at stair #1

14.90

22

shear failure of 11 X11 1st floor columns

15.58

23

compression failure of basement corner walls

16.81

24

compression failure of basement elevator core

23.11

Performance point

4.5.1 Perfol'l11ance Objectives Per Section 3.4 of the Methodology, various performance objectives can be selected in the ; evaluation of a structure. In this example building study, the owner selected a performance objective to satisfy Life Safety requirements for a Design ! Earthquake ground motion that is defined in , Section 4.5.3.

!.

MldrlSl~ AppenCllx A. EsconClIClo Village MIClrise

Capacity spectrum 4.5.2 The force·displacement pushover curves shown in Figures 4.4-4 and 4.4-5 are converted to spectral coordinates per Section 8.3.2 of the Methodology. The capacity spectra for the longitudinal and transverse directions are shown in Figures 4.5-1 and 4.5-2, respectively. Tables 4.5-1 and 4.5-2 show the conversion for the longitudinal and transverse direction pushover curves respectively. Since the loading function was

A-27


,.,r--------------,---------------,--------------,

rr 1'' 1/ ,

-

'.00 ~_-------_+----------__+_--_----___i SpeClnll DI.plaeemeat. Sd

"

"

(ID~UJ

Figure 4.S·1. Longitudinal capacity Spectrum

as afi

ef 0.30

:; 020

II'" ! '.00

··_--1

,-----

---

-

4. TI pa M bt

I

I,

I !

•

!,

/,

! i

I

"

SptdJ1ll DI.pI.ceraut. Sd (ladle.]

"

•

"

• Figure 4.S·2. TranSverse capacity spectrum

•

•

Table 4.S·1. conversion Of v and d,.., to So and Sd for Longitudinal Direction A

1633

2.07

0.138

1.449

0.653

0.211

1.43

0.83

B

1756

3.08

0.148

1.449

0.653

0.227

2.13

Q98

2011

11.37

0.170

1.449

0.653

0.260

7.85

o

2052

12.84

0.173

1.449

0.653

0.265

8.86

1.85

E

2011

18.08

0.170

1.449

0.653

0.260

12.48

2.H

A-28


sp

----

f

-----------------------------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

---

Table 4.5,2. conversIon 01 V and d,.., to Sa and Sd lor Transverse DIrection

.,j,/W

PIT"",

T,seCI '

A

1137

1.80

0.096

1.451

0.671

0.143

1.24

0.94

8

1258

2.29

0.106

1.451

0.671

0.158

1.58

1.01

C

1354

3.28

0.114

1.451

0.671

0.170

2.26

1.17

0

1478

6.10

0.125

1.451

0.671

0.186

4.20

1.52

E

1790

14.10

0.151

1.451

0.671

0.225

9.72

2.10

F

1824

17.02

0.154

1.451

0.671

0.229

11.73

2.29

G

1839

23.11

0.155

1.451

0.671

0.231

15.93

2.65

Demand Spectrum 4.5.S The 5 percent damped spectrum is derived from parameters described in Chapter 4 of the Methodology. For the Escondido Village Midrise buildings, the following parameters were used: •

Soil Profile Type = D for stiff soil (Methodology Table 4-3)

•

Seismic Zone, Z = 0.4 for seismic zone 4 (Methodology Table 4-4)

•

Near Source Factor, N = 1.18 for seismic source type A, linearly interpolated between 5 and 10 kIn (Methodology Table 4-5)

•

Seismic CoeffIcient, CA = 0.47 for shaking intensity larger than 0.4 (Methodology Table 4-7)

•

Seismic Coefficient, Cv = 0.76 for shaking intensity larger than 0.4 (Methodology Table 4-8) Based on the capacity spectra, the demand spectra can be reduced with the modification

i

Sdfln.1

d""" fln.1

assumed to remain constant throughout our analysis, the participation factor (PFroof) and the effective mass coefficient (CXm) remain constant.

'Idrlse

" 0/"'< 1"'5.'91'"

V'klpil

point "

Appendix A, Escondido Village Mid rise

factors SR. and SR, as calculated by the following relations (see Chapter 8 of the Methodology): d ' - d a) ]) SR, = - -1 - ( 3.21- 0.681n [63.7 K(a'P" pi + 5 2.12 ap;d p;

p; p.) SR, = - 1 ( 2.31- O.4l1n[63.7 K(a,d - d,a + 5] ) 1.65 ap;d p;

By guessing the maximum displacement of the capacity spectrum, the values of dp; and ap; (based on the capacity spectrum) can be calculated. These, in turn, effect the values of SR, and SR,. Through an iterative process of adjusting the value of dp; until the capacity spectra intersects the demand spectra at dp, a performance point can be determined. Figure 4.5-3 shows the relationship between api, ay, dp; and dy. For the longitudinal direction, the total spectral roof displacement at the performance point was Sdmox=9.51" (see Figure 4.5-4), which corresponds to a total roof displacement of d.,.,= 13.8". For the transverse direction, the total spectral roof displacement at the performance point was Sdmox= 11.2" (see Figure 4.5-5), which corresponds to a total roof displacement of d.,.,=16.2".

A-29


,.

r

I

I

api·

actual pushover curve

. _. _. _. _ .•J. _. _. _. _. _. _.:-................. ................ .. .. .. . .. .. .

ay - _._._.-

idealized pushover curve

dy

dpi Displacement

Figure 4.5-$. Idealized Bilinear RepresentatIon of MDt/al Pushover curve

1.0

de pc de of fo

0.0

CL

bl

0.'

~

4.

s~

0.1

r··

10 in

.ll

0.'

4.

-< 0.4

i

0.' 0.2

P(

0.1

ac

Ie

0.0 0

2

4

,

•

7

o

10

11

12

13

14

"

C(

M

C(

Figure 4.5-4. Demand vs. capacity Spectra Showing performance point for longitudinal Direction

PI PI

4

N

s:

A-SO

Appendix A. Escondido Village Midrlse

-

-

~--------------------------------------------

SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS ~--------------------------------------------

---7--··_·····_···_···_·_····

1.0 0.'

,

0.8

~

•

dpi=3.0"

i

0.7

~

I

0.'

<

0.'

l

0.3

~

~

i

0.'

0.2

Sd=l1 "

m

0.1 0.0 0

4

,.

6

1

8

9

10

II.

12

13

14

IS

Spectul D11p1aeement,Sd [Inches]

Figure 4.5·5. Demand VS. capacity spectra showing performance point for Transverse Direction

4.5.4 Performance point The intersection point of the capacity and demand spectra is the performance point. This point represents the expected level of seismic demand on the structure. The spectral coordinates of the performance point can be converted back to force-displacement coordinates on the capacity curve. For the Escondido Village Midrise buildings, the performance point occurs at a base shear of 2010 kips and 13.8 inches in the longitudinal direction, and 1750 kips and 16.2 inches in the transverse direction.

4.6

Performance Assessment

Component deformations at the performance point displacements must be checked against acceptable limits. The acceptable deformation levels for various structural elements and components are presented in Chapter II of the Methodology. Individual evaluation of these components is required to determine not only the performance level of the component, but also the performance level of the entire structure.

4.6.1 Drift Limits Based on story drifts, the Escondido Village Midrise buildings (as strengthened with the Life Safety Objective scheme presented in Section 5 of

Ildrlse


this report) satisfy the Immediate Occupancy performance level criteria in the longitudinal direction and the Life Safety performance level criteria in the transverse direction.

Component Deformabllity 4.6.2 Walls. With the exception of basement walls, typical concrete shear walls are flexure critical. Adequacy of these walls are based on plastic hinge rotations. In general, existing shear walls satisfy immediate occupancy requirements (Methodology Table 11-7) with the exception of some first floor walls that satisfy the Life Safety performance level as shown in Table 4.6-1. Basement walls are checked by drift ratios shown in Methodology Table 11-8. Because of the large number of walls in the basement level, deformations are small and meet Immediate Occupancy requirements. The inability of the transverse walls to transfer required shears at the first floor level is due to the door and louver openings which reduce the number of dowels that make the required shear transfer. In the degraded model these walls are allowed to resist only 20 percent of their yield shears. As shown in the pushover curve, there is still substantial strength after the "failure" of these walls. Per requirements of Methodology

A-31


Table 11.6-1. NumerIcal Acceptance criterIa for Plastic Hlnlle Rotations In ReInforced concrete walls and wall sellments CDntrolled by Flexure

h( th c! M ot Zf

SI

sl T, Sf

n( /I

d, 1.

{A,-A,)f,+P t,..lwfr'

3.

; assume A.i

= Ax"

flY =

19.5". I", = 25'

.J.g

t•

•

N.A. (Not Applicable): Deformations remain elastic.

Table 11-9, the expected sliding displacement satisfies the Life Safety performance level. Columns. Columns were initially checked using simple SAP90 stick models. These models were displaced at each floor level based on the final building displacements at the performance point. Since columns were determined to be shear critical. only shear capacities were checked. Columns typically had adequate shear capacity to resist required demand which indicates that columns generally remain elastic. Nevertheless. first floor and basement interior 15xZ4 columns were found to have demand-to-capacity ratios ranging form 1.4 to 2.0. Consequently. these columns would not remain elastic and can be evaluated per the Methodology. As noted previously. the beam-column frame system does not contribute significantly to the lateral-load-resisting system. but rather goes along for the deformation "ride". Therefore. columns that undergo inelastic deformation were checked

A-32

•

v

2.

using the secondary member performance criteria. Because of the lack of adequate confining steel. these first floor and basement columns fall under the category of Columns controlled by shear, other cases in Methodology Table 11-4 and are judged to be unacceptable. Shown below is a simple calculation of the plastic hinge rotation at an interior 15x24 first floor column:

b, fa a( e1

th el

5 . = "t n LI tota I h'mge rotatlon =L

t

VL2 elastic hmge rotatIOn = 8. = L' = 12EI ..

.

LI

plastic hinge rotation = 8 = 6 -6 _ 1.148 pt. 109

since

~A gf.

s.

N 0.00997

P,

4~1.8) = 03n 0.1 and

al hi C

34.4(109)2 12(3490)(17280) 360 3.750


n(

AI

f' ~

---

-----------------------------------------

SEISMIC EVALUATION ANO RETROFIT OF CONCRETE BUILDINGS

--------------------------------------------------------------------.

hOoP spacmg

eria.

:1. der

other ~ed

!tic

22

11",

the column member falls in the other cases category for columns controlled by shear. Per Methodology Table 11-4 for secondary elements in other cases. the allowable plastic hinge rotation is zero for both the Life Safety and Structural Stability Performance Objectives. Beams. Floor beams are checked using the slab-column connection criteria in Methodology Table 11-6. As noted above. these beams are secondary elements. In general. these beams do not satisfy the Structural Stability requirements.

4.6.S

•

d

=12"~2=1l=

Summary of Deficiencies

LifeSSfetY oblectlile.:> Damag:e c(Jntrol ..... ObJectfve., '. . ~ : ~~i.V:: . ' ," ,," " :'In DesIgn Earthquake ,r:

in. DesIgn Earthquake ,.

Reinforcement of shear critical columns Floor beam supports Discontinuity at transverse Shear walls Shear wall boundary elements

Reinforcement of shear critical columns Floor beam supports New concrete shear walls New pile foundations

For a Life Safety Performance Object. deficiencies are summarized as follows: • Lack of confining steel in first floor and basement columns render them unacceptable for any level of plastic deformation.

of a large magnitude earthquake causing strong ground motion at the site. For the purposes of this study. only the Design Earthquake with soil type D was considered . The required structural work for the , two objectives is summarized in Table 5.1-\'

•

S.2

S.

Conceptual Retrofit Designs

5.1

performance Objectives

After determining that an existing structure is unable to resist design earthquake demands. the engineer often evaluates a number of alternative retrofit concepts to determine feasibility. applicability, and cost. Technical strategies, as well as management strategies. are employed to obtain the required seismic risk reduction. The advantage of using a nonlinear. pushover analysis is the ability to determine the potential failure mechanism of the building as it deforms. As a result. the engineer can focus his retrofit design solely on the elements that are deficient so that the building can reach the desired performance level without changing the entire behavior of the structure. Using conventional elastic evaluation techriiques, the retrofit of the Escondido Village Midrise structures would probably include the addition of new concrete shear walls. While this solution is included in the Damage Control objective retrofit work. the addition of shear walls would not have addressed the most critical structural deficiency. As shown in the Table 5.1-1.

Lack of adequate reinforcing in floor beams result in significant hinging and potential collapse of most beams above the first floor level. As previously mentioned. the hinging of beams and shearing of columns were significant failures. These were assumed to be included and addressed in retrofit schemes so that the inelastic evaluation could continue. It is interesting to note that these deficiencies were not apparent in the elastic FEMA-178 evaluation.

Retrofit designs for the Escondido Village Midrise buildings were developed for two Performance Objectives: the Life Safety objective already described and used for evaluation. and a higher Damage Control objective. The Damage Control objective is to limit structural and non-structural damage to the building in the event

drlse

Table 5.1-1. Required Retrofit Work for Different performance Objectives


Retrofit strategies

A-!!


even with the addition of shear walls, additional upgrade schemes must be incorporated to address the shear critical columns, and hinging floor beams.

5.3

Retrofit systems

A combination of technical issues and management issues were considered in the design and implementation of the seismic upgrade schemes. Some of those issues are discussed below.

5.5.1

RetroFit FDr the LIFe saFety Objective

Reinforcement of Shear Critical Columns. To retrofit the first floor and basement columns, either the shear capacity could be increased or addition confinement could be provided. The provision of bolted steel jackets would provide additional confinement as well as added shear capacity. However, Stanford required the building to remain functional during the entire construction process which meant drilling of concrete would be limited to keep noise levels at a minimum for residents. Consequently, the first floor columns were confined with fibrewrap reinforcement. Unfortunately, basement columns were typically built into adjacent concrete walls. The use of fibrewrap reinforcement is impossible. Therefore, bolted steel jackets are used in the basement area. Drilling of concrete would be required at only three column locations. Floor Beam Supports. To strengthen the floor beams at each level to resist expected earthquake demands would be impractical from an engineering, as well as a construction perspective. Since the performance criteria is for life safety, significant damage that required repair after a large earthquake would be acceptable as long as the structure did not collapse. Therefore, the approach was to allow the beams to hinge and form vertical cracks at the slab-column joints, but provide secondary vertical support to prevent the beams and slabs from collapsing.

A·54

A steel corbel around each interior column was devised to provide secondary vertical support. Initially, this was to be a series of welded steel angles bolted to the concrete column directly below the floor slab. However, this was judged to be unacceptable on two counts. This type of connection required a large number of anchors drilled into the existing concrete column. Installation of such anchors would have been extremely difficult since existing column reinforcing could not be damaged. In addition, Stanford had the low noise requirement previously mentioned. Nevertheless, the corbel idea was not abandoned. Instead of relying on tension and shear of bolts embedded into the concrete column, we decided to try friction collars. These collars will derive vertical support through friction between the steel tubes directly below the slab and the existing concrete columns. No drilling of concrete will be required. Since this friction collar concept is an unproved method, a testing program was setup to verify the adequacy of these restraints. Through the testing program, we were able to determine an appropriate friction coefficient that enabled the design of the final friction collars. Discontinuity at Transverse Shear Walls. The discontinuity created by the door and louver openings greatly reduce the effectiveness of the transverse shear walls. The transverse walls reached their shear capacity prior to the performance point. The limited displacement of the walls satisfied the Life Safety performance level. Nevertheless, in the interest of ensuring ductile behavior, bolted steel jackets at the columns will provide additional confinement and shear resistance. Shear Wail Boundary Elements. As mentioned in Section 4.3.4, the shear wall boundary elements were assumed to contain intermediate grade (40 ksi) steel reinforcement with staggered lap splices. Concrete at one of these boundary elements was chipped away to reveal the steel reinforcement and splices. After chemical and tensile testing, the steel was


\\

jl sl (

b tI P n n

\\\

II

I~ I

el n

5

I~

S

-----. umn SUppOrt, ;reel :ly dged to If 10rs

:n ion, :viously

ld shear 1,

we

swill ween he oncrete oncept ",as ints. , to It that

--

SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINCS

confirmed to be 40 ksi steel. However, lap splices were not staggered, but instead were all located just above the floor slab. In addition, these lap splices lacked adequate development length. Consequently, the actual building would not behave as modeled. After consideration, we felt that the modeled building behavior met performance objectives and the building should be retrofitted to behave as the model. Therefore, reinforcement in boundary elements are to be Iwelded or lengthened to provide adequate splices. In addition, boundary elements at the first floor level would be provided additional confinement to ensure ductile performance since this was the region of most significant wall hinging. Figure 5.3-1 shows a typical floor plan of this life safety level retrofit scheme. 5.~.2

New Concrete Shear Walls. New concrete shear walls provide additional stiffness and shear resistance to the buildings. As a result, deformations are reduced as well as demands on other shear walls.,' New Pile Foundations. New cast-in-place drilled concrete piers are required to support the new concrete shear walls for overrurning forces. Because the new walls are significantly more rigid than other walls, earthquake loads will tend to be resisted by these walls resulting in substantial overturning forces. The piers are anticipated to be on the order of 30" diameter by 50' long with about 40 piers required for the four new walls. Despite eliminating the need for boundary element reinforcing and jacketing of basement transverse walls, the addition of new shear walls still requires the provision of column collars throughout the building (except the roof level) and column jackets (fibrewrap)at the first floor. Furtherfnore, these walls greatly alter the architectural appearance of the buildings and require new pile foundations.

Retrofit for the Damage Control Objective

Reinforcement of Shear Critical Columns. See Section 5.3.1. Floor Beam Supports. See Section 5.3.1.

lTS.

ails. ouver fthe

s

"~'''I''~ .,

,. . . .

-~--'~~'--.ji··:··

··ir'

. .. .. "

.111..

I 18''''"

l.q.pllce '

I ...

. ~ .. (i),' ... @,' . . . liJ ...,: ~I..I"IJ" ~ 11---.' . . . '. . . '!''''!''!' ~oIl., '!P: . :. . . . . . . . _______1 - - _. . . . . , . . .

I

lent of to \fter

.: ....... ~.~: G EJj)

, .' .... , ... a-........."",'. CV

nt of nce ing It and

...

~-_--";'r--rt---I

J--1'-_L~L---.oI

.....

(Q)

10'-7"

... ..0 ...•·.3·. G

10--__- -.....:----.01..... :.... :....: .. ~'~'. CAl 12'

12'

• 12'-7"

12'

12'

12'

,

12'

7"

12'

'2'

Figure 5.:1,1. TYpical Floor Plan Showing strengthening for Life Safety

Mldrlse


A-IS


6.!

/ "·3· ~

to I pre ace dir giv fau ree spe wo

~

onl

... , coli.

typo

II ..... "'" 1~

,..

'N·

,~

,..

'--I-,.. ,~.~

fo"-- -' ,..

,..

Figure 5.6·2. TypIcal Floor Plan ShowIng StrengthenIng For Damage Control

Figure 5.3-2 shows a typical floor plan for this Damage Control retrofit scheme. The additional work does not seem warranted for the limited reduction of damage.

6.

Assessment of the product '.2 Methodology

Damage Prediction The Escondido Village Midrise buildings were damaged during the October 17, 1989 Loma Prieta Earthquake. This included moderate but widespread cracking of the cast-in-place concrete walls, including both shear cracking in classic diagonal "x" patterns, flexural cracking consisting of cracks that were approximately horizontal near the bases of the walls, and horizontal cracking along the construction joints present at floor levels. The walls around the stair towers experienced the heaviest damage. Most damage to the walls was repaired shortly after the earthquake with the injection of epoxy grout. Pushover analysis predicted flexural yielding of concrete shear walls. The most significant

6.1

A·36

damage was accurately predicted to be in the stair towers. However, damage due to shear cracking was not predicted. Perhaps this is due to the fact that shear capacities in the model are based on the combination of concrete shear capaci ty and steel shear capacity. Diagonal cracking of concrete shear walls does not indicate that concrete walls have reached their calculated capacities. The Lorna Prieta event was only a moderate short duration earthquake for the Escondido Village Midrise buildings; consequently, damage to the interior columns and beams did not occur.

comparlsori with preliminary Evaluation Findings FEMA-178 Evaluation Statements accurately indicated the problems associated with the vertical discontinuity in the transverse shear walls and the inadequate boundary reinforcing in shear walls. The Evaluation Statements, however, failed to discover the inadequacy of the beams and columns. Therefore, for this example building study, nonlinear pushover analysis prescribed by the Methodology proved to be a very useful tool in predicting damage and focusing retrofit efforts.

6.2

APpendix A, Escondido Village Mldrlse

usi ele Ra ela

tin

pre

hy:

pre

bel wi

as~

bel is : di!

T

D

Ti A fII

o

-

P

N

ill.

~

----

---------------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINCS --------------------------------------------------------------6.3

~ stair king fact )n the :teel

:e

alls rate nage cur.

'Y

Comparison with Inelastic Time·History Analysis

Limited time-history analyses were performed to determine the accuracy of the performance point predicted by the Methodology. Five near-field acceleration records, each with components in two directions, were selected. For each record, the given components were transformed to fault-parallel and fault-normal components. These records were then scaled so that the average spectral acceleration of the ten time-histories would be O.64g for a structure with a period of one second. The time-history analyses were performed using DRAIN-2DX. These models did not include element degradation or pinched hysteretic curves. Rather, the element constituitive models presumed elastic-purely plastic behavior. Table 6.3-1 compares the average maximum time-history displacements with the displacements predicted by the Methodology assuming similar hysteretic behavior and with the displacements predicted assuming pinched and degrading behavior. Comparison of the time-history results with the Methodology are good, when the same assumptions are made with regard to hysteretic behavior. However, when more realistic behavior is assumed, the Methodology predicts larger displacements, as would be expected. Table 6.S·1. comparison of Maximum Roof Displacement

ately ~rtical

td the lls. to

Time-History Average

7.5"

8.3"

MethOdology wlo Degredatlon and Pi

7.5"

10.2"

Iy,

e in

'Is.

For shears and overturning moments in the main longitudinal walls, the time-history average

Ildrlse


maximum compares well with the Methodology, as shown in Table 6.3-2. However, shears and overturning moments in the main transverse walls are substantially different. The shears obtained in the time-history evaluation are more than 5 times larger than those obtained through analysis per the Methodology. The overturning moments from the time-history analysis are only 30 percent larger than that from the Methodology. The large difference in shear and smaller difference in overturning moment indicates that higher mode effects are significant in the transverse direction of the building. The discontinuity in the transverse walls may be a significant contributor to the need for evaluating the structure for higher mode effects. In fact, the Methodology does suggest that irregular buildings (vertical discontinuity makes this an irregular building in the transverse direction) should be evaluated based on a Level 4 or Level 5 approach that does include the contribution of higher modes. Also note that the shear and overturning moment demands for the transverse walls are in the elastic range. Although forces in the time-history average and in the Methodology vary significantly, the transverse walls do not yield in either case. In other words, the Methodology did not miss any significant yield event. Consequently, the overall building behavior determined by the Methodology remains consistent with that determined by time-history analysis.

6.4

Conclusions

The following broad are reached conclusions regarding the use of the Methodology: • The Methodology adequately predicted the shear wall damage observed after the Lorna Prieta Earthquake. •

The Methodology also determined failure mechanisms that were not readily apparent in a conventional elastic evaluation, such as the hinging of beams and shearing of columns. Also note that the shear and overturning moment demands for the transverse walls are in the elastic

A-57


Table 6.6'2. Comparison Of Shea' and overtumlnll Moment

I.

2.

Longitudinal wall forces are taken from first floor elements of the main longitudinal walls (W13 and W23). Transverse wall forces are taken from second floor elements of the main transverse walls (W3l and W41).

range. Although forces in the time·history average and in the Methodology vary significantly, the transverse walls do not yield in either case. In other words, the Methodology did not miss any significant yield event. Consequently, the overall building behavior determined by the Methodology remains consistent with that determined by time-history analysis.

7_

Foundation Analysis

7.1

Introduction

As an additional part of our study, effects of various foundation parameters on the expected building behavior were evaluated. The large number of shear walls in the basements of the Escondido Village Midrise buildings essentially provide a fixed base foundation at the first floor level. Consequently, varying soil properties had negligible effect on structural behavior. Because the Escondido Village Midrise buildings were not very sensitive to varying foundation effects, the DRAIN·2DX models were modified by removing the spring at the first floor level representing the additional basement walls. In this manner, we produced a model that could be affected by varying soil parameters. It should also be noted that element degradation was not considered in the foundation evaluations presented here.

7.2

Varying Soli Parameters

Eight independent DRAIN-2DX models (four in each direction) were constructed to evaluate the effects of varying soil parameters. These models included:

A-sa

• •

Fixed base model Stiff soil model (with soil stiffnesses and yields that were 100 percent greater than the average values)

• •

Average soil model Soft soil model (with soil stiffnesses and yields . that were 50 percent less than the average values) Figures 7.2-1 and 7.2-2 show the pushover curves for each of these models in the longitudinal and transverse directions, respectively. As shown in these figures, it appears that soil stiffness and yield only effect the initial portion of the pushover curve. Regardless of the soil parameters, all curves seem to converge after yielding of members. This indicates that soil stiffness can effect the response of the building prior to yielding of structural elements. Once structural elements start yield, soil stiffness has negligible effect on building behavior. A more detailed evaluation of soil effects was performed for models in the longitudinal direction. Table 7.2-1 presents the roof displacement and base shear data points along the pushover curves where wall hinge formation initially occurs. In general, as soils become less stiff, hinge formation at basement walls occurs at larger displacements and shears. Figures 7.2-3 and 7.2-4 show the point of hinge formation of basement longitudinal walls and stair #1 walls, respectively, with various soil parameters. In addition, as shown in Table 7.2-2, the estimated performance point also moves further down the pushover curve for softer soils, resulting in larger expected displacements and shears for the same design earthquake.

-


-

IPU8C1C1y


eUUOj~p UIJOjl~

~!A\ 'l~qul1J UllOjl~d

llalfl Aq )ds~lJOJ

'!lUA

2000

~lfI

,

!

1800

ljj!d

.....-:;:.,

1600 .J>.

(!.,. ,.'.. ,. .., /~'

"00

,-

200

J Ii

, base

I!\,~ ,tiff, 400

!

-'"

,.

I.v'/-fu

600

~-t ,,40'" I

II

i': .."'~,avclll e soil

:: ,

§/ ,

",--,

soil

,

I

o 0.00

'.00

u.oo

10.00

Root Dilplacemenl [lnchell

Figure 7.2,1. Longitudinal Pushover curves wIth Various Sail Parameters

-

1600

1400

\" fucd base 1200

~ ..

j '"~

.

JI~S JI~S O~eAa13

800

400

200

/

'i,-~

600

JaUJOJ n~!6uO'

~'/

.i~~ , .'

,i i I !

,,

!

average s il

,I

I

soft soil

. o 0.00

! I!

""'- stiff s il

1/,..",,-

i ".

....J,: I,

(~l i'- .'

i

a

...... -

.....

',"

1000

~

-~

'.00

10.00

I, 15.00

RoorOllplacemeDt [inebes]

Figure 7.2,2. Transverse Pushover Curves with various Sail parameters

IAr:'Delndli]( A, Escondido Village Mldrlse

A-39 •

----------------------'-----------------_

eslAPI


'sJ' 2000

!

i

1800

' \ 1.86". 1337 k 1600

IZ

1200

•

11.

J !'"

., .'

I

1000

!

1..

"-5.67". 1319k 3.25" 1316k

I

~

..

I !

fuo dbase

800 600

400 200

nO!

sIle, lu!od

Sl1

j

'f'f < liL' ;:I// Y----

1400

i

,

'\2.59". 1~37k.

\

..,... - .. - .. - .. I

, ..- .

'z-,

U0!lll

S;l,

'UOp

seN

i

h(,~ stiffs il i/'..~~,avera e soil :: J,;:',I II • ~s soil

1

S

!

.,1

I !j

I

i !

o 0.00

5.00

IS.00

10.00

JO

t nO!

RoofDllplacemeat {Inebel]

Figure 7.2-S. Longitudinal pushover curves showing Hinge Formation at Main Basement Longitudinal Walls fW1S and W2S) 2000

I

v- 3.28".1455k 1800 ~ 1.65".I225k

1600

/ Ir

\ y:, ..L .

1400

I

•

I

'iLi I

•

•

/,i/ Y----

,. ..

., . ".-

.

400

:,.'

".

'. I

!

' - - 10.33".1628 k

!

,..

,, I

1

fuo d base

~s

I I

I

soil

l ~

o 5.00

10.00

IS.00

Figure 7.2-4. Longitudinal pushover curves showing Hinge Formation at Basement stair #1 Walls

A-40

;II

;

Ii

0.00

SP!
_.. ' !I

L.., - .- .. -".

,1~.z.,"-. stiffs n i/ ,..~~,avcm c soil ::

600

200

4.45 ".1494k

i

11lU!


-

laaay

---------------------------------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILOINGS

----------------------------------------------------------rallle 7.2,1. comparison of wall Flexural Yield poInts for Varying Soli Stlffnesses SOft Soil

;

drool.·.· .\~.." " . > . " .1" (iff.} '1' (kips} .

.....,

'..•.•.. '

~;'.V_.,. ..' (In.,..' I·· (kips) . .

LOngitudinal wailS

1.86

1337

2.59

corner wails

2.33

1407

2.76

Elevator Core

1.70

1263

2.67

stair #1

1.65

1225

3.28

stair 62

2.10

1375

3.03

·.drDDf .. ','··V.... .••.. .. flnJ. (kips)

1337

I

drDDI ">11'..... . (In.) .. (kips)

3.25

1316

5.67

1319

1388

3.16

1293

3.84

992

1362

3.63

1404

6.82

1439

1455

4.45

1494

10.33

1628

1449

4.04

1462

7.98

1534

2000

r 11.0".I650k

9.7".I600k \ 1800

9.3". 1620k

1400

!

•

~ 1200

j !'"

•

I

I

"

,..

;

'~'-"-"-L 14.4". 1720 k ..J

;....

/

,.

:

Il/~M d base

800

/!/,,~ stiffs il /!.;i~~.vera esoil

600

iN.' 1/ •

200

I

/{/

1000

400

-

,.

(

~\

, ..

1600

'. I

~sc ft soil

ji

o 0.00

'.00

10.00

IS.CO

RoofDllplacement [Inchel]

Figure 7.2·S. longItudinal pushover Curves ShowIng performance Points

Figure 7.2-5 shows the pushover curves for the various longitudinal models with their corresponding performance points as determined by the Methodology. As shown, it is apparent that performance points for models with softer soils are further out on the pushover curves With larger expected roof displacements at the performance point for softer soils, larger deformation demands will be placed on structural

and nonstructural elements. For the modified Escondido Village Midrise buildings, only the comer basement walls at the basement experience compression failure within the expected performance points for the cases with average and soft soil stiffnesses. The compression failure at toes of comer walls occurs at approximately 10.3" and 11.5" for the average soil and soft soil models, respectively. No walls experience

-, Appenalx A. Esconalao Village Mldrlse

A-41


compression failures for the fixed base and stiff soil models. Additional lateral deformations throughout the building due to softer soils will also result in the building's inability to satisfy required performance objectives. In the fixed base case, the building satisfies the Methodology's Immediate Occupancy Objective when based on the maximum drift. (The building satisfies the Life Safety Objective when based on the inelastic drift.) On the other hand, the soft soil case can only meet the maximum drift requirements for the Life Safety Objective. It can be inferred that perhaps for other buildings, stiffer soil assumptions may result in satisfaction of higher performance objectives as compared to the same building with softer soils. (Note that the example models evaluated here meet the Methodology's Life Safety Objective only because of the inelastic drift requirements.) It was previously shown that the shear critical columns at the first floor and basement levels req~ir~d strengthening to satisfy the Life Safety ObJective. Because of larger interstory drifts at all levels with the soft soil assumption, additional columns in floors above the first floor would require .strengthening. The beam supports would be required regardless.

7.3

Comparisons with Inelastic Tlme·Hlstory Analysis

A series of time-histories were ran for the fixed base (no basement wall spring) model as well as the average soil stiffness (no basement wall spring) model. For average soil stiffness, the Methodology predicts a maximum roof displacement of 17.7" which is substantially larger than the 9.72" calculated as the average for the time-history analysis. A comparison of time-history average demands and Methodology demands at the basement level of main longitudinal walls (WJ3 and W23) is shown in Table 7.3-1. The overturning moment at the longitudinal walls calculated by the Methodology compares well with that calculated from the time-history analysis. However, the shear at these walls as calculated by

A-42

Table 7.~1. Comparison Of Time-History Average Demands and Methodology Demands (with Average So/I stiffness) at Basement Level of Main LongitudInal Walls fW1S and W2SJ

7 st CI

Time·Hlstory Average

• 1280

364,000

• Table 7.S·2. Comparison Of Time-History Average Demands for Fixed Base and Average soli Models at Basement Level Of Main Longitudinal walls fW1S and W2SJ

•

I I.

the Methodology is approximately 30 percent of the time-history average value. As previously noted in Section 6.3 of this report, the longitudinal walls do not yield in flexure or shear at this level of demand. Consequently, the apparent discrepancy in the Methodology is not considered critical in determining the overall building behavior. Table 7.3-2 presents the shear and overturning moment demands for the time-history averages of the fixed base model and the average soil stiffness model. As shown, the shear demand in the fixed base model is larger than that in the average soil model. This is consistent with the fact that a more rigid structure would attract more lateral loads. It is interesting to note that the fixed base overturning moment is slightly lower than the average soil overturning moment. This indicates that higher mode effects are more dominant in a structure with a flexible base. Also note that the maximum roof displacements are 7.26" and 9.72" for the fixed base model and the average soil model, respectively.

APpendix A, Escondido Village Mldrlse

2.

3

4

5

verage ~

alMaln

'i!menf ~

,

7. 4

conclusions

Based on limited analysis of varying soil stiffness and yield strength, the following conclusions were made: • Soil stiffness controls a building's initial stiffness until major structural components start to yield. •

Walls tend to yield at larger roof displacements as foundation assumptions became more flexible.

•

Roof displacement and base shear at the expected performance point for a design earthquake typically increases as soils became less stiff.

'verage Idinal

8. I.

ent of Isly gitudinai is level


2.

lsidered

murning 3. rages of

stiffness ~

fixed ge soil t a more 4. loads. It

1 the

dicates 5. nt in a hat the nd 9.72" soil

•

Additional structural and nonstructural elements will fail with softer soils because of larger deformation demands for the same design earthquake.

•

Higher mode effects were increased with a more flexible foundation.

•

Desired performance levels may not be reached as foundation assumptions became more flexible.

References 6.

Gilbert, Forsberg, Diekmann, Schmidt; Structural Drawings. "Stanford University Married Student Housing" December 15, 1961

7.

Abbot A. Hanks, Inc., Report "Foundation Investigation for Married Student Housing Increment 2 Buildings 134, 135, & 136 Stanford University", March 16, 1962

Gilbert, Forsberg, Diekmann, Schmidt; Structural Drawings. "Stanford University Escondido Village Increment 2" April 1, 1964

8.

Gribaldo, Jones and Associates, Report "Foundation Investigation for Escondido Village, Increment III; Stanford, California", October, 1970

Campbell & Wong & Associates; Architectural Drawings; "Stanford University Escondido Village Increment 2" April 1, 1964

9.

Gribaldo, Jones and Associates, Report "Soil Investigation for Escondido Village, Increment IV; Phases I & II Stanford, California", July, 1970

Meserve Engineering; Letter Report "Barnes Hall Slab Cracking"; October 27, 1989

10.

Gribaldo, Jones and Associates, Report "Soil Investigation for Escondido Village, Increment V; Stanford, California", February, 1969

Woodward-Clyde Consultants. "Evaluation of Site Response and Design Earthquake Ground Motions, Stanford Urtiversity, Palo Alto, California" December 2, 1991.

11.

Building Seismic Safety Council; FEMA-178, "NEHRP Handbook for the Seismic Evaluation of Existing Buildings," June 1992.

Woodward-Clyde-Sherard & Associates, Report: "Soil Investigation for the Married Student's Housing Project Stanford University, Stanford, California", January 27, 1958

Ie Mldrlse Appendix A. Escondido Village Mldrlse

A·45


12.

Department of the Army, The Navy, and The Air Force. "Techriical Manual for Seismic Design Guidelines for Essential Buildings TM-5-809-10, NAVFAC P-355.2, AFM 88-3" 1986.

International Conference of Building Officials. "Uniform Building Code" Whittier, CA. 1991.

20.

Iwan, W.O. "Estimating Inelastic Response Spectra from Elastic Spectra". Journal of Earthquake Engineering and Structural Dynamics. Vol. 8. 1980.

E L

21.

Computers and Structures Inc. "ETABSThree Dimensional Elastic Analysis of Building Structures". Berkeley, CA.

PI

13. American Concrete Institute. "Building Code Requirements for Reinforced Concrete - ACI 318-92" Chicago, IL. 1992. 14.

15.

Prakash, V.; Powell, G.H.; and Fillipou, F.C., "DRAIN-2DX" Report No. UCB/SEMM-92/30, Department of Civil Engineering, University of California, Berkeley, December 1992. Woodward-Clyde Consultants. Report: "Soil Spring Evaluations - Escondido Midrise Buildings" February, 1994.

16.

Newmark, N.M. "Dynamic Response of Structures to Earthquakes" EERI, 1980.

17.

Moehle J.P. and Wallace J. W. "BIAX: A Computer program for the Analysis of Reinforced Concrete Sections" Department of Civil Engineering, University of California, Berkeley, Report No. UCB/SEMM-89/12, July 1989.

18.

Merovich A. and Zsutty T.C.: "Boundary Element Behavior in Shear Walls". Proceedings of the Structural Engineers Association of California 1993 Annual Meeting. Scottsdale, AZ.

A-44

,.

19.

22.

Aboutaha, S.M., Engelhardt, M., Jirsa, J.~. and Kreger, M.E. "Seismic Retrofit of RIC Columns with Inadequate Lap Splices" 1994 ASCE Structures Congress, Atlanta Georgia.

23.

Aboutaha, S.M., Engelhardt, M., Jirsa, J.~. and Kreger, M.E. "Seismic Shear Strengthening of RIC Columns Using Rectangular Steel Jackets" 1994 University of Texas, Austin.

24.

Aboutaha, S.M., Engelhardt, M., Jirsa, J.~. and Kreger, M.E. "Seismic Retrofit of RIC Columns Using Steel Jackets". Proceedings of the 1994 American Concrete Institute spi-ing Convention. San Francisco, Ca.

25.

Priestly, M.J.N., Seible, F. "Seismic Assessment and Retrofit of Bridges" , Department of Applied Mechanics and Engineering. University of California, San Diego. July 1991. .


I

, ~

~

-------------------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS -------------------------------------------------------------IlPpendlx B

;ponse Iof II

Example Building Study Barrington Medical Center LOS Angeles, California

3S f ;a, J.O.

fRIC " 1994 eorgia.

prepared by Rutherford & Chekene 303 Second Street, Suite BOON San Francisco, California 94107

a, J.O.

!rsity

a, J.O. fR/C dings te

San

~Idrlse:

Appendix B. Barrington Medical center

a-'

-------------------------------------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

---------------------------------------------------------------------------

Table of Contents I. Introduction .................................................................................................. B-5 1.1 Intent of Example Building Study ........................................................ B-5 1.2 Scope of Example Building Study ........................................................ B-5 1.3 Summary of Findings ...................................................................... B-5 I. 4 Update ........................................................................................ B-6 2. Building and Site Description ............................................................................. B-9 2.1 General ....................................................................................... B-9 2.2 Structural Systems and Members ...................................................... B-lO 2.3 Soil and Seismicity ....................................................................... B-lO 3. Preliminary Evaluation .................................................................................. B-lO 3.1 Summary .................................................................................... B-lO 3.2 FEMA-178 Evaluation Statements ..................................................... B-ll 3.3 Elastic Analysis ........................................................................... B-12 4. Evaluation by Product 1.2 Methodology .............................................................. B-12 4.1 Summary ................................................................................... B-12 4.2 Scope ....................................................................................... B-13 4.3 Structure Modeling ....................................................................... B-13 4.4 Pushover Analysis ........................................................................ B-18 4.5 Identifying Limit States on the Capacity Curve ...................................... B-21 4.6 Determination of Demand and Performance Point. ................................. B-23 4.7 Performance Assessment ................................................................ B-26 5. Conceptual Retrofit Designs ............................................................................ B-28 5.1 Performance Objectives ................................................................. B-28 5.2 Selection of Retrofit Elements .......................................................... B-28 5.3 Comparative Evaluation by Product 1.2 Methodology ............................. B-31 6. Assessment of the Product 1.2 Methodology ......................................................... B-32 6.1 Damage Prediction ....................................................................... B-32 6.2 Comparison with Preliminary Evaluation Findings ................................. B-34 6.3 Comparison with Inelastic Time-History Analysis .................................. B-34 7. References ................................................................................................. B-36


a·J

,...-----------------------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS ~---------------------------------------------------------------

APpendix B

Example Building Study Barrington Medical Center LOS Angeles, California 1.

Introduction

1.1

Intent of Example Building study Product 1.2 of the Proposition 122 Seismic Retrofit Practices Improvement Program, entitled Seismic Evaluation and Retrofit of Existing Concrete Buildings, is referred to here as the Methodology. Example Building Studies were executed primarily to "test" draft versions of the Methodology and to provide feedback to its developers. This study also illustrates techniques described in the Methodology and attempts to convey the scope of work and the level of engineering judgment involved in evaluating an actual building. However, since the study ignores some Methodology requirements for brevity, it does not represent a complete evaluation or retrofit design. This report assumes that readers are familiar with the basic Methodology scope and terminology. This study was completed by Rutherford & Chekene in several phases, in parallel with development of the Methodology. This report describes work done in the last phase, in March-April 1996. Some references to the latest Methodology requirements, equations, or scope may be out of date. Section 1.4 updates principal results to the latest Methodology requirements. 1.2

scope of EXample Building study This study presents the evaluation and conceptual retrofit design of an actual concrete building in Los Angeles, following the recommendations of the Methodology. Topics include:


•

Preliminary evaluation (Section 3 of this report)

•

Modeling, analysis, and assessment by nonlinear pushover analysis (Section 4)

•

Conceptual retrofit (Section 5) In addition, Section 6 of this report provides a limited assessment of the Methodology. Practical implementation of pushover analysis is also discussed. Issues include: •

Idealizations and simplifications (Section 4.3 of this report)

•

Modeling shear-critical components (Section 4.4.3)

•

Modeling stiffness and strength degradation (Section 4.4.4)

•

"Effective" yield point for performance point calculation (Section 4.6.4)

1.3

summary of Findings This study confirms the anticipated value of the Methodology as an analytical tool. Compared with conventional elastic analysis, the Methodology provides a more complete description of expected structural performance, allowing and encouraging better understanding by the engineer. Except in rare cases where elastic analysis clearly reveals exceptionally good or bad behavior, such an understanding is essential, and a nonlinear Methodology is worthwhile. The Methodology is valuable even where it relies on judgment or approximation, as it demands explicit consideration of expected inelastic response. Though software limitations and

8-5


bookkeeping requirements may make the Methodology more difficult to implement than elastic analysis, these hurdles will certainly be lowered with time and widespread usage. Specifically, the Methodology can be assessed by comparing its results with other perfonnance estimates (see also Section 6): • The Methodology's modeling rules adequately predicted the exterior column damage observed after the Northridge earthquake. (Interior walls were not inspected after the earthquake.) •

Evaluation Statements, supported by engineering experience and to a lesser extent by elastic static analysis, probably would have led to a retrofit scope similar to that indicated by the Methodology's inelastic pushover analysis. However, the pushover analysis provides a more detailed understanding of expected building and component performance. Additionally, the Methodology addresses the relative significance of potential damage (by defining multiple performance levels) and discusses retrofit approaches.

•

Limited time-history analysis suggests that pushover analysis can overestimate displacement and underestimate base shear. See Section 6.3 for further discussion. Key findings regarding implementation of the Methodology are also worth noting here. Each is discussed more fully in the text below. These findings are not intended as criticism of the Methodology. Rather, they should remind the prospective engineer that better analytical tools do not remove the need for engineering judgment and careful work. •

•

•

Substantial engineering judgment may be required in applying modeling rules and deformation limits. Engineering judgment and approximation may be required where available software can not directly model anticipated behavior. Some of the Methodology's demand and capacity estimates may be sensitive to

B-G

assumptions or approximations. For this example building, acceptability was heavily influenced by assumed soil type, assumed foundation properties, and deformation limits sensitive to assumed building conditions. Judgment and "envelope" techniques should be applied where equally reasonable assumptions yield very different conclusions.

1.4

Update Some Methodology requirements have been significantly revised since this study was first executed. This Section briefly discusses the most critical changes.

Modeling The Methodology now includes an explicit discussion of techniques for modeling degradation in its Section 8.2.1. The recommended technique, resulting in a "sawtooth" composite curve, is very similar to the approach taken for this study (described in Section 4.4.4), although there may be some minor differences. The previously determined capacity curves are assumed to comply with the latest Methodology requirements.

1.4.1

Del'lvlng Pel'fol'mance Points The Methodology offers several different procedures for deriving bilinear capacity curves, performance points, and/or target displacements. Among these are the displacement coefficient method and the capacity spectrum method (CSM). This study originally used a hybrid of what are now called Procedures A and B of the CSM. Updated performance points tabulated below were derived with Procedure A. Also, the revised Methodology now specifies use of the most appropriate yield point for the bilinear representation. This corresponds to the "subsequent yield" discussed in Section 4.6.4. The most significant revision to the CSM involves assignment of Structural Behavior Types and corresponding Damping Modification Factors (lC) and Spectral Reduction Factors (SRA and SRv). With reference to Methodology Table 8-4,

1.4.2

APpendix B, Barrington Medical Center

Bal

an bel dUl COl

bel

is

~

spe dis em Fi~

del soi pel is : 4.S spe dis typ Co chi pel ori de! on inc 75' the rel! 2iJ poi del am Ag gO(

hig bel dis pre Wit

Api

r ---------------------------------------------------------------------

~

SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS ~

---------------------------------------------------------------------------

s

Barrington Medical Center would be classified as an "Average Existing Building" and assigned lvily behavior type B or C depending on expected ed duration of shaking. If retrofit, it might be limits ,l. considered an "Essentially New Building" with behavior type A or B. (Note that the behavior type lould be is assigned on the basis of primary elements only.) lptions With corresponding K values and limits on spectral reduction, performance point displacements increase significantly, in many cases enough to affect acceptability. For example, ,een Figure 4.6-4 below shows the performance point st derivation for the fixed-base original building on most soil type D in a Design Earthquake. The , performance point spectral displacement, dp or Sd, is 3.45 inches, and the actual roof displacement is cit 4.9 to 5.4 inches (see Table 4.6-2). With limited dation spectral reduction, however, this expected roof lique, displacement increases to 8.1 inches with behavior s very type B or about 12 inches with behavior type C. Comparison with Figure 4.4-2 suggests that such a may change can have significant impacts. Table 1.4-1 summarizes the increases in omply performance point roof displacements for both the original building and the retrofit building described in Section 5. For the original building on soil type B, Methodology revisions have increased displacements by 1 to 3 inches (25 to yes, 75%) depending on behavior type. On soil type D, !nts. the increase is 3 to 7 inches (50 to 120%). For the t retrofit building on soil type D, the increase is 1 to :SM). 2 inches (30 to 100%). Ie Table 1.4-2 compares the updated performance point displacements (Table 1.4-1) with those were derived using the displacement coefficient method and the equal displacement approximation. Agreement between the three methods is very good for soil type B, but the CSM appears to give higher values for soil type D. Also, structure behavior types B and C lead to CSM displacements significantly higher than those ypes predicted by displacement coefficients, especially ctors with soil type D (12" vs. 5.5", 14" vs. 9.5"). SRv).

:enter


Table 1.4·1. Changes In Design Earthquake performance point displacements due to Methodology revisions finches} . .••..••• COndition····

'Behavlor' From

TO

original building Fixed Base Soil Type B

B

C

2.7

4.7

Soil Type 0

B

5.4

8.1

C

5.4

12

Soil Type B

B

3.6

5.0

C

3.6

6.4

Soil Type 0

B

7.2

10

C

7.2

14

A

2.0

3.0

B

2.0

4.8

A

4.3

5.4

B

4.3

6.7

2.7

3.4

Soft Foundation

RetrOfit Building Fixed Base, Soil Type 0

soft Foundation Soil Type 0

Table 1.4·2, comparison Of Design Earthquake performance pOint displacements calculated by various methods (Drlglnal building only) finches/

a-7


Note that Framing Type 2 was used for the displacement coefficient calculations; if Type I had been assumed, the predicted displacements would be 10% higher, still significantly less than those predicted by the CSM.

1.4.4 Retrofit Requirements With the updated performance assessments come updated retrofit requirements. These are summarized in Table 1.4-4, with references to report Section 5 below.

1.4.8 performance Assessment Given revised performance point displacements in Table 1.4-1, the assessments and retrofit requirements in Section 4 must be reviewed. Many conclusions have changed. Table 1.4-3 summarizes the revised conclusions with reference to report sections below. Clearly, the revised Methodology is more conservative than the earlier versions with which this study was originally conducted. Note that the effects of Methodology revisions have been checked for the Design Earthquake only.

1.4.5 Methodology Assessment Section 6.3 and Figure 6.3-1 below note that for the soft-foundation model on soil type D, the reported Design Earthquake performance point was higher than those predicted by inelastic time-history analysis but was still near the mean-plus-one-standard-deviation of 14 time-history results. By contrast, the updated performance point displacement (10 to 14 inches per Table 1.4-1) exceeds the mean time-history result by two or three standard deviations. This

Table 1.4-$. Changes In performance assessments due to MethodoltIIIY revisions (check olllle saFety performance In Design Earthquake onlyl

4.7.2 columns: ok with soli type B, not ok with soli type D.

4.7.2 Walls: limit of 5 Inches Ok for all cases: 4.7.2 Pile slip: soil type B: .25-.50', soli type D: 1.0-1.25'. 4.7.3 5011 type B deficiencies: unreliable exterior frame columns only. 4.7.3 5011 type D defiCiencies: Walls marginally acceptable.

a-8

Column performance is not ok with either soli type. Exception: barely ok for fixed-base mOdel on soli type BIf behavior B Is assumed. Barely ok with soli type B 14.7' vs. 5'); not ok with soli type D IS' or 12' vs. 5'). soli type B: .70-1.0', soli type D: 2-4', depending on assumed behavior type. With 4' Slip, pile mOdel Is suspect. unreliable and unacceptable ext. frame columns. walls marginally acceptable. Walls unacceptable; other deficiencies same.

Table 1.4-4. Changes In retrofit requirements due to MethodolOflY revIsions (desilln lor soli 0 ..'",,, onlyl

5.1 Original building meets Economic objective. only soli type D considered. Table 5.1-1, Flxed·base model wi either Table 5.1-1, Flxed·base model WI Economic Objective Table 5.1-1, 50ft-Fdn model WI Economic objective 5.3 Check retrofit on soli type Bagainst column L5 deformation limit: fixed base easily ok, soft foundation I I 5.3 Beams adjacent to retrofit Inflll panels ok.

wails not ok. Columns not ok. same, but tighter requirement on frame column deformation. Add: limit deformation In frame columns; Strengthen or add wailS. confirm pile model.

dis val rec bel me his

(i.€

bel

am

bel bui

Fixed base Ok assuming behavior A13.0" vs. 3.S"), not ok assuming behavior B14.S' VS. 3.8"). Soft foundation not Ok 15.4-6.7" vs. 4.2"). Not checked.

Appendix B, Barrington Medical Center

Api

-----

r ---------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS ~----------------------------------------------------

Its

.,- ----------[-----------i

'e :0

(ElINFlll PANEL.

EDGE OF 2ND I FLOOR ROOf ~

TYP, AT J PIfCES

,

,,

l--+--+--t--------@----@----. .. . , for nt

,, ,,

,

20'-5"

104'-0"

20'-5"

ckl' ·-t·-·-·--~·· --~--.-.-

i i i

}-+-+-...... ----.-.,-.-.

:hes 'ry lis

! ! !

20'-5"

~s

•

~

•

! !

! !

TYP. FRAME C'olU~N: 12"x42" I

!

!

! ;

([)8" WAll G~OUP, TYP.

8}-+--+-.. ·-·-·-·-·~N~·~~:;t~:·-·-~·-·-·-·-i·-·-·-·-~·-·-· -.-. 21'-4\1.

,1/

ELEVATOR

!

i

! 22'-4"

TYP. FRAME BEAM (NOr SHOWN): 8"x84"

21'-6"

21'-6"

-

21'-6"

!

!

21'-6"

22'-4"

130'-8"

1~tfl,~

ns

me n.

tion

Figure 2.1·1. Plan of OrigInal BuIlding

discrepancy arises from pushover analysis with K values significantly less than 1.0 and from spectral reduction limits imposed on buildings with behavior type B or C. Even if a structure is modeled with degraded strength and stiffness, time history analysis with DRAIN-2DX assumes full (Le. not pinched) hysteresis loops associated with behavior type A. Therefore, such time history analyses may not account adequately for the poor behavior expected from some existing concrete buildings.

ling 3.8"), 3.8"),

tok

:enter


2.

Building and Site Description

General Prior to its demolition in 1994, Barrington Medical Center was a six-story reinforced concrete office building in West Los Angeles. Although the building footprint was rectangular, a substantial setback of the northeast comer above the second floor resulted in the L-shaped plan shown in Figure 2.1-1. The building had a 17-ft first story (plus 1'-10" to top-of-pile cap), five 12·ft stories, a small (about 2000 sf), light penthouse with steel diagonal braces, and no basement. Barrington Medical Center was designed in 1963. Damage sustained during the Northridge Earthquake of January, 1994 is described in Section 6.1. No records of damage from previous

2.1

B·9


earthquakes were available for this study, but the building is assumed to have been in good condition at the time of the Northridge earthquake. Documents available to this study are listed under References. Original calculations and drawings do not cite a specific building code.

exterior spandrel beams were deep, they shortened and stiffened the exterior columns. Further, the strength and stiffness of piles relative to the walls they supported is unclear. Finally, the L-shaped typical floor may have given rise to torsional force distributions.

2.2

Materials • Specified 28-day concrete strength: 3000 psi for slabs, beams, columns, walls, and grade beams; 2500 psi for pile caps and piles

structural Systems and Members

Barrington Medical Center was constructed of cast-in-place concrete. Typical elements included:

Gravity load-Resisting System (see Figure 2.1·1) •

7 1/4" two-way flat slabs carrying floor loads to walls, frames, and columns·

•

8-inch wall groups at stairs and elevators

•

Perimeter frames of deep rectangular spandrel beams and columns

•

Round interior columns with 4-foot diameter capitals

•

Under interior columns and walls, pile caps and friction piles 26 to 43 feet long

•

Under perimeter frames, grade beams and 36-foot long piles

lateral load-ResIsting system (see Figure 2.1·1) •

Load-path: relatively rigid slabs, through shear walls and frames, to foundation

•

East-west direction: interior shear walls and perimeter frames

•

North-south direction: interior walls, perimeter frames, and three perimeter frame bays infilled to form de facto exterior walls Structural calculations from 1963 show that the shear walls and frames were both expected to resist earthquake forces. However, the distribution of forces and ultimate behavior of this building are unclear from a visual review of drawings. The interior shear walls were configured as stiff boxes, but were lightly reinforced in some areas and were softened by numerous openings. Because the

B·1D

•

2.3

dist Bec los! as-t Stat defi con

3.2

exi~

Stat all' Fall apl FEI

Reinforcing was called out as "intermediate grade deformed bars. "

Soli and Seismicity

The original soil report shows "moderately firm" sands and silts to a depth of about twenty feet, with somewhat firmer sands below and no water encountered for fifty feet. The description is consistent with Uniform Building Code Soil Type S2: "dense or stiff soil." The 1963 report does not address site seismicity. Located at the intersection of Olympic and Barrington in West Los Angeles, the building was about 5 km from the trace of the Newport-Inglewood fault in the COMO Beverly Hills quadrangle. .

Stat accl anal defi rest wer FEI avai full dra1

:I.

preliminary Evaluatlan

indi

3.1

summary

Gel

•

Methodology Chapter 5 recommends a preliminary evaluation to determine if nonlinear analysis is warranted. Such a preliminary evaluation identifies potential deficiencies using FEMA-178 Evaluation Statements and static elastic analysis. For Barrington Medical Center, Evaluation Statements identify a number of deficiencies in the original design. Some of these-particularly poorly-confined, shear-critical columns-require retrofit if structural damage in a Los Angeles Design Earthquake is to be avoided. Without substantial analysis, however, Evaluation Statements are unable to predict aspects of actual behavior such as elastic and post-yield force

•

•


L

~

--------------------------------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

~

Jrtened es If.

: given

) psi 'ade iate

:ly nty no lion is Type es not !clion !les, fthe 'rly

on

------------------------------------------------------------------------------------

distributions, hinge patterns, or ductility demands. Because it does not account for post-yield stiffness loss in perimeter frames, elastic analysis of the as-built structure adds little to Evaluation Statement findings and is still unable to quantify deficiencies. The preliminary evaluation is thus conclusive in a qualitative sense only.

FEMA·178 EValuation Statements FEMA-178 is a methodology for evaluating existing structures by responding to Evaluation Statements. When the appropriate Statements are all True, no analysis is required. When some are False, elastic static analysis is used to determine if a particular condition is a real seismic deficiency. FEMA-178 force levels check life-safety only. For this study, FEMA-178 Evaluation Statements are used for preliminary evaluation in accordance with Methodology Chapter 5. No analysis was performed to verify potential deficiencies. No testing or field investigation results were available. Non-structural elements were not considered. It is important to note that FEMA-178 evaluation relies heavily on the availability of design drawings; for this study, a full set of original structural and architectural drawings was available. False FEMA-178 Evaluation Statements indicating potential seismic deficiencies include:

ng

•

•

er,

tical >in a led. ltion :ual

!nter

Frames •

No Shear Failures. Exterior frame columns with clear heights reduced to 5' -0" by deep beams are probably shear-critical, but only if the Methodology's shear capacity is used. That is, the columns' nominal shear capacity, as defined in FEMA-178, may be enough to develop their flexural strength. The degrading shear capacity defined in the Methodology, however, indicates shear-critical behavior at expected deformation levels.

•

Strong Column/Weak Beam. Columns can not develop the strength of deep frame beams.

•

Detailing. Tie spacing, stirrup spacing, joint reinforcing, and bar splices are all potential deficiencies. Typical ties in columns and joints are #2@12" and are drawn (but not specified) with 135-degree hooks. No special ties are provided at hinge regions, joints, or splice zones.

•

Flat Slab Frames. Slabs have #5 bottom bars spliced over columns. With typical column capital, however, splice is two feet from critical section at edge of capital.

3.2

Ceneral !ar

may be significantly higher than 60 psi in frame columns and 50 psi in shear walls.

•

Vertical Discontinuities. Elevator and stair wall cores have significant setbacks, discontinuities, and/or openings. Torsion. If the exterior frames are weak or shear-critical (as suggested by False Evaluation Statements below), the center of rigidity after one frame yields is unclear. Also, the L-shaped plan and full second floor complicate predictions of two-dimensional response. Shearing Stress Check. Depending on the distribution of story shear between walls and frame columns, shears under FEMA-178 loads

Appendix B. Barrington Medical Center

Shear Walls •

Wall Overturning. Interior core wall hll ratios exceed 4.

•

Coupling Beams. Effective coupling beams over wall openings are not specially reinforced.

•

Confmement Reinforcing. Boundary elements have only typical ties: #2@12".

•

Reinforcing Steel. Typical wall steel ratio (#4@12"e.w.) is less than .0025.

•

Reinforcing at Openings. Typical detail calls for only 2#5 trim bars around door openings. Evaluation Statements requiring analysis or field investigation for a complete response are

B-11

SEISMIC I!VALUATION AND RI!TROFIT OF CONCRI!TI! BUILDINGS

conSidered Unknown. Among the Unknown conditions are: • •

General pre-earthquake condition of concrete Walls, columns, or frame components.

D~ft check. Drift is usually not an issue in bUIldings with concrete shear walls, but with shear-critical frame elements, small drifts may be Sufficient to cause damage.

•

Stirrups and tie hooks. Bar details are drawn with 135-degree hooks but do not specify hook angles or extensions.

•

FOUndation settlement or deterioration.

•

Lateral force on deep foundations. Pile bending and shear transfer to soil not checked.

•

LiqUefaction potential.

3.3

avaluatlon by Product 1.2 Methodology

4.1

summary

Specific technical findings are given in Tables 4.5-1 and 4.6-2. Deficiencies are summarized in Section 4.7.3. General assumptions, conclusions, and lessons regarding the Methodology include: Consideration of element inelasticity in the structure model allows a more complete and useful understanding of expected performance through a full range of lateral movement than is generally available from linear elastic analysis.

• •

Evaluation may be unable to satisfy some Methodology requirements because of software limitations. In particular, three-dimensional inelasticity and multi-linear load-deformation relations are difficult to model.

•

Engineering judgment is required to determine an appropriate model scope. A full inelastic model accounting for all potential failure modes is often unfeasible, but reduction to oversimplified "equivalent" models may miss key points of the Methodology.

Elastic Analysis

. Elastic analysis of existing buildings has been wlde!y accepted because it is procedurally consIstent with building code requirements for new construCtion. Elastic analysis can complement Evaluation Statement responses by predicting stresse.s and small deformations. Computerized :malysls can also account for peculiar structural Irregularities. However, elastic analysis does not acco~~t for post-yield force redistribution or ductIlIty demands. For this study, a three-dimensional elastic computer model, built and analyzed with ETABS soft~are, was used only to assess potential torsIOnal effects in the L-shaped structure. Member stresses were not checked. When the elastic model is subjected to east-w~St loads, torsional effects have no ap~reclable effect on response and can reasonably be Ignored. Loading in the north-south direction, however, indicates significant torsion.

a·12

4.

•

•

Engineering judgment may be required where building conditions are not directly addressed by available modeling rules. In particular, post-yield capacity of shear-critical and degrading elements (likely to be found in existing concrete buildings) is not fully addressed. Over-simplified or careless modeling can cause misleading analysis results. Engineering judgment may be required when categorizing elements as "primary" or "force-controlled, " when setting deformation limits, and when comparing limits to expected performance point displacements. Relative sensitivities, sources of uncertainty, and error magnitudes must be considered.


• • sub bey eva pro eng 4.~

jus1 her pre invi EVI in ( ana mal sei! rep ana pot bas

• • • •

• 4.! 4.. ~ nor gen

Api

~

----------------------------------------------------------------------------

--- -----------------------------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

c:t

•

Assumed soil type can control the most basic evaluation conclusions.

•

abies in

j

ons, de: le md lance than

Soil-structure interaction, represented by a reasonable range of foundation models; can significantly affect the global nature of building performance. While the Methodology still requires substantial engineering judgment, it goes well beyond the practical limits of linear elastic evaluation procedures. Also, the Methodology provides for a peer review process as a check on engineering judgment.

4.2 ftware ' lal :tion rmine tic :0

miss 'here ssed

scope

The complete Methodology involves more than just the nonlinear pushover analysis described here. Methodology Chapter 5 recommends a preliminary evaluation including field investigation, material testing, execution of Evaluation Statements, and limited elastic analysis in order to determine the need for nonlinear analysis. For this study, field investigation and material testing were not possible, and site-specific seismology data was not available. Preliminary evaluation (see Section 3 of this report) suggested that a nonlinear pushover analysis to quantify wall-frame interaction and potential column failure may be warranted. The basic steps of a pushover analysis are: • Structure modeling (Section 4.3 of this report) •

Application of pushover forces (Section 4.4)

•

Determination of limit states (Section 4.5)

•

Determination of inelastic demand and expected performance point (Section 4.6)

'hen

•

Assessment of building elements and systems at the performance point (Section 4.7)

Ition ected

4.3

'e

!rror

:enter '

these are often the same conditions ignored by commercial analysis programs. The limitations of currently available software can significantly affect the level of effort required to model and analyze the building, the nature of analysis results, and if not understood by the engineer, the correctness of analysis results as well. However, this does not necessarily reflect on the value of the Methodology; in time, more comprehensive software will be available. For this study, DRAIN-2DX was used for the nonlinear pushover analysis. Among its limitations relative to requirements of Methodology Chapter 9 are: • No Inelastic Panel Elements. Walls subject to potential flexural or shear yielding were modeled as columns. See Section 4.3.4. •

No Shear-Critical Elements. Only flexural yield can be simulated with a simple model, but the Methodology requires consideration of all relevant failure modes. See Section 4.4.3.

•

No Degrading Elements. All yielding elements maintain their strength, but the Methodology requires degrading for some elements with high ductility demands. See Section 4.4.4.

•

Limited Post-Yield Behavior. Upon yielding, the program modifies flexural stiffness only, not flexural strength on subsequent cycles (degrading), shear strength or stiffness, or axial strength or stiffness.

•

Two-Dimensional Framing Only. As noted above, the east-west direction is reasonably approximated by two-dimensional models, but a two-dimensional study of north-south loading would require special modifications to account for torsion. Methodology Commentary 9.3 questions the utility of three-dimensional inelastic analyses.

•

Limited Model Size. The Methodology recommends modeling joints, stairs, gravity framing, and other elements of uncertain rigidity, but this quickly increases model size

structure Modeling

4.s'1 SoFtware Considerations For an existing building, the point of a nonlinear pushover analysis is to assess conditions generally avoided in new designs. Unfortunately,


a-13


and analysis time. Also, premature failure modes such as shear can be modeled with multiple elements in series, but this can double the model size. No Graphics OR Post-Processing. This affects the efficiency of the analysis. Other Methodology requirements that are handled by DRAIN-2DX but might not be in other software packages include: • P-delta effects or geometric stiffness modifications.

•

Reinforcing rupture strain: .05. Refer to Methodology Figure 9-8.

•

Initial stiffnesses: Per Methodology table 9-3 with two exceptions per Section 9.5.3 Commentary. First, no reduction from gross stiffness for shear-critical frame columns. Second, reduction by .75 factor, not .50, for frame beams expected to stay below yield.

•

•

Unlimited range of rigid end offsets.

•

Explicit modeling of shear stiffness (but not shear yield) in primarily flexural elements.

•

Eigen solution at any point of the pushover

4.s'2 Materials Material test results were not available for this study, so strength data was taken from available design documents. Structural materials were modeled with the following assumptions: • Existing concrete strength: 3600 psi = 3000 psi design strength factored by I. 2 to reflect higher in-situ values (See Methodology Commentary 5.4.4.1.) The 1.2 value is a matter of engineering judgment only. •

Concrete Young's modulus, E = 3400 ksi. The value reflects the increased strength.

•

Ultimate compression strain (per Methodology Section 9.5.2.2 and Commentary 9.5.4.2): . 005 for most elements, .003 for poorly confined boundary zones subject to high compression.

•

Existing reinforcing strength: 40 ksi. (Methodology table 9-2 indicates a strength of 60 klii for "intermediate grade" bars. The use of 40 klii was an error left uncorrected.)

•

Reinforcing strength increases: 25 percent per Methodology Section 9.5.4.1, ignoring the 40 percent increase recommended in Section 9.5.4.2.

B-14

structural Systems 4.8.8 Three-Dimensional Effects. Only the east-west lateral load-resisting system was modeled for this study, in part due to software limitations. Elastic analysis confirmed that torsion is reasonably ignored for east-west loads. However, torsion may be more significant than elastic analyses indicate because the perimeter frames are brittle; if frames on opposite sides of the building yield at different times, the system could become subject to torsion. Even without torsion, bi-directional effects (ignored here) should be considered in a real building evaluation. These could affect corner columns, corners of wall assemblies, and the perimeter frame columns which have little strength out of plane. Idealized Fixity. Rigid diaphragms were assumed. This simplifies the model by reducing the number of elements and degrees of freedom. For a long narrow floor plate, this assumption might be inadequate. The small piece of slab between the reentrant corner at grid D-4 and the nearby stair shaft is an area of potential concern whose strength should be confirmed by hand . Other idealized fixity assumptions included: • Horizontal translation fixed for all ground floor nodes, assuming infinite soil stiffness. •

•

For fixed-base models, all walls and columns fixed at ground floor against rotation and vertical translation. For foundation models, bottom end nodes of piles fixed, otherwise ground floor nodes free in rotation and vertical translation.


• grie eaSi

Mel as t "eq con gra' mOl

pari Sec bay thrf mOl

4.3

loae stue rela

Lin In ~ wit!

f --------------------------------------------------------------------

~


--------------------------------------------)

Ie 9-3

1

gross

"C

IS.

1 I I

ca

I

...J

I

0

l, for Id.

--. .. --

"C

ca

..9

I

. - -I

0

-

I I

Deformation are

IS

'e

cing iom. on b i the cern :I. nd less. lumns d es of 's free

center

• • ...

•

!L-""

--

-~ I

Deformation

Figure 4.S·1. toad'Deformatlon Relations for Deformatlon'Controlled and DegradIng Elements Methodology RecommendatIon {•• •J and Model SImplificatIon {-J

torsion han ter ~s of tern )ut should These I

0

...

•

Beam-column joints rigid and infinitely strong over full beam depth and column width. (For shorter beams, Methodology Section 9.4.3.2 might have applied.) Secondary Elements. Gravity framing along grid lines B, C, D, and E was modeled as east-west slab-column frames in accordance with Methodology Section 9.4.2.2. Slabs were modeled as beams framing directly into columns; that is, "equivalent frame" stiffnesses from ACI were not computed. Some later models eliminated the gravity framing on line E in order to reduce the model to an executable size. Nonstructural elements such as stairs and partitions were not modeled, despite Methodology Section 9.3.1. Had there been solid infills of frame bays in the east-west direction, as there were in three north-south bays, these would have been modeled as shear walls.

4.$.4

structural Elements and components

The Methodology cites a generalized load-deformation relation in its Figure 9-15. This study used a simplified elastic-perfectly plastic relation; the "Lateral Resistance Deformation Limit" was monitored by hand. See Figure 4.3-1. In general, elements were modeled in accordance with Methodology Sections 9.5.5.2 through


9.5.5.7, using the material idealizations listed above. Most elements were modeled with a yield overshoot tolerance of 5-10 percent in accordance with Methodology Commentary 8.2.1. Unique modeling aspects included the following. Beams. No provision was made for hinges within the beam span, but this was adequate for this building where column hinging limited beam forces to less than yield. Because the beams were fairly deep, shear stiffness was included. Shear strength of each beam type was computed per Methodology Section 9.5.4.3; a value of k = 1 was used because beams were expected to remain essentially elastic. In each case, the shear strength was found sufficient to develop the full yield strength of the beam in reverse curvature, so careful monitoring of beam shears during pushover was unnecessary. Although design drawings show longitudinal bars spliced within the column depth, splice and development length were judged to not control beam strength. Columns. Since columns were not loaded within their clear height, hinging was anticipated at member ends only. The column clear span-to-depth ratio was small, so inclusion of shear stiffness was essential. Flexural yield strengths were computed as for beams, and a yield curve accounting for P-M interaction was specified for each column type.

a-15


For each column type, shear capacity was checked against the shear associated with flexural yield. For typical frame columns, Methodology equations 9-3 to 9-5 with k = 1 give a shear capacity sufficient to develop the nominal moment capacity. However, the columns are the first components to yield and may be subject to high ductility demands. If so, this would require the use of k = 0 in the shear strength equations, reducing the shear capacity and making the columns shear-critical. In a large enough earthquake, then, the columns may yield in flexure on one or two cycles, then yield at a low shear value on subsequent cycles as cracks open and shear strength degrades. Refer to Section 4.4.4 of this report. . . Joints. Beam-column joints were modeled as ngld zones, without separate model elements. Had the columns been stronger in shear, it might have been necessary to check the shear in the joint, es.pe~ially considering that tie spacing is large Within the panel zone. Note that separate modeling of the joints would have increased the model size substantially. . Slabs. Two-way slab behavior was modeled With one-way beam elements. Potential hinge locations were limited to the face of column capita\. Despite the recommendations of Methodology Section 9.5.5.6, slab column C?nn~ctions were not explicitly modeled; all slab Yleldmg was limited to simple flexural yielding of the Slab. This was considered appropriate because the slab-column "frames" were secondary elements that did not control overall performance. Walls. Because DRAIN-2DX does not offer an inelastic panel element, wall groups were modeled as eqUivalent columns (see Figure 4.3-2). The three wall groups at grid C-3, 0-3, and C-S.S were modeled. For each variation in a wall ~roup's reinforcing and geometry, a P-M mteraction diagram was developed with PCACOL software. For some wall groups, moment-curvature relations were also developed to

confirm that yield moments did not violate assumptions of ultimate concrete or steel strain. Lintel sections over stair and elevator doors are expected to behave as (unintended) coupling beams but are not properly reinforced. To account for these conditions, the concrete and steel area . input to PCACOL was limited to that which could be developed by the lintel in shear. This resulted in reduced model wall capacities that represented potentially non-ductile failure modes. In general, inelastic deformations in idealized model walls must be checked for local effects at critical locations such as openings. This building's walls were found to be shear-critical, so it was not necessary to consider local stresses associated with flexural yielding. Wall shear capacities were calculated by Methodology equations 9-6 to 9-8, using a beta value of 0.58. Capacities were such that, depending on the interaction of walls and perimeter frames at different stories, the wall might have been controlled by shear, so careful monitoring of wall forces during pushover was necessary. Where shear con trolled, wall flexural strength was modi fied to simulate a shear yield at the proper stage of the pushover. Potential sliding failure along construction joints was not modeled; this deformation mode is expected to occur before wall shear capacity is reached, but is further expected to stiffen or strain-harden, allowing development of wall nominal strength. The wall properties described above were assigned to single equivalent columns at wall centroids. To maintain deformation compatibility with adjacent components, artificially rigid and strong "outrigger" beams were provided (see Figure 4.3-2). As noted in the commentary to Methodology Section 9.4.3.1, this equivalent column approach can be kinematically incorrect. For this building, however, yielding was either shear controlled or concentrated in the piles. Also, walls are not rotationally linked to each other or to other primary lateral load-resisting elements, so the Methodology's concern was not prohibitive.

will

moe COUI

moc 9.3, pre) to fI suff sucl is IT perl mee inte pro' soft COlT

prel Wer and elas

moc on ~ Pile

AppendIx B, BarrIngton MedIcal tenter

f -------------------------------------------------------------------------------

~


--------------------------------------------------------------------------------

ain. )ors ling ;COunt .rea could Ilted mted .eral. lis

14TH

~

TRIBUTARY FLOOR LOAD + WALL WEIGHT. TYP.

3RD

~ 2ND

4-~9

~

~~ 9'

:1.

C1..l.

~4TH

RIGID OUTRIGGER, TYP.

..l.

EA H CORNER EQUIVALENT COL. I, ETC

CI~TYPE

1ST

8-~9

EA H CORNER ~alls

x~

:d with,

<~~~

~,~

R

'-'~>

1ST

eta Figure 4.$·2. Idealization of Wall croup as Equivalent column

II :ful las ~ural

eld at liding ieled; Jefore

,, 'e I

bility md

e :0'

It

rect. ler Also, r or to , so ive.

center

Foundations. This study considered models with and without foundation elements. In some models, wall and column bases were fixed. In counterpart models, foundation elements were modeled in accordance with Methodology Sections 9.3,9.3.2. and 9.4.6, and recommendations , prepared specifically for this study. It is important to remember that the foundation model is not sufficient to assess specific foundation components such as piles; rather. foundation and soil stiffness is modeled primarily to assess potential impacts on performance of the superstructure and its ability to meet given Performance Objectives. A range of model parameters for friction piles, intended to envelope actual properties, was provided as shown in Figure 4.3-3. Only the soft-weak relation was fully analyzed for comparison with fixed-base models. Results from preliminary analyses with the stiff-strong relation were about half way between the soft-foundation and the fixed base-models. Piles were modeled as elasto-plastic axial elements. Grade beams were modeled as spanning between piles; direct bearing on soil was not modeled. Rotational stiffness of pile groups under frame columns was modeled

Appendix B, Barrington Medical center

with additional spring elements. Rotational stiffness under shear walls did not require modeling because pile groups at each wall end provided rotational resistance as a couple.

0

f

K = 2EA

V

ci.

E 0

!d.

~

(ij

!§ 0

~

L/

/iI

/' K =.5EAlL

Tenston capac tty may be limned by bar strength.

Displacement

Figure 4.$·$. Force-Displacement RelatIons for Friction Plies (stlff·strong and SOft-Weak). ((hit provided In soli report. see References.)

B-17


,Shape: :hape:

0.00; 0.16 l.3 0.47 O.~ 1 i o.n 1.45 ,~a~__~4.f~~'. 8~~0~.OO~'~ 11.2~3.~ 1~.~~2.67~1~.6II,~(M~~~~~1~8l.~5~~1~2~ . 79 __~4-~~~

IO!~

~

: = 01a

0;

..

l.OI:"

;

o.oo,!

4;

i.()

4.5~

:

4~'·4.5E

i~::i .~ ~~"~'~::'Hr i~:: !:Ei~ Pl .. ,028 ,"6f~106 0' ;Shape: '.' 'O~~.' ; =~, 6: 15ha1le:03b • EFf;;.7 ~ ~ !, 6,1'=.7 fShaP8: lOab ?iT=.9 03b '·(f;;.9

i

0.00; 1.'~: 1.66! 2.s8 00: 00': ' 00, 0.09. O. . 0.38 "Ii:OO~ I 2.14 O. 0.39 0.00' 0.461 1.21 2.21

o:oo:-o.oe

0.13

, ..... "'1:4S:

-,

1.00

'

1-"· 2'·'!"~'0·e· "-·;····0:74 i-:- ...:-~,~'i-:--:-. ~. ~j-~

17:60;;;;"'''''"'-·.-: .

;

3.584.61,4.5; 18.13:12.43 ;.461 0.76. : '1.01 .' .. . 0.1: 1.00 4.51 4.56. 4.56: 0.56 0:79. --,:Qi . , '{4'7! 0.69. 3.2E4.45: -4':51;.-'_1:.:61.=;..24·_..:.111;.::1.0:::.77:....1._"::::::~---':.::!!.~ 0.59 0.79, "1.iii . i 3.3~· 4.4 4.51"---= 16I.'.:d 0.611; 24 . -~-~II'"IC'"'.'2""3;"-"-.:·=+-·-·---~=J 1.45'

,

!

,

SShhaa~~':: ~,03b~~_~8 ~~•.I~.~~.OO~' ~~)~.4 0.70. 1.451 , a . T=I. .00:. ~_~~~~' 0.62~ !.3~... ~0~.61L~~+,~I~.0~I,·;==~~~t: 3.41 4] i! 4.56'16.n =J[!4==J~~t==Ig~i_ 11.57: --. l

~. ~., 'ShaP&: i03br

.... .. .... ..' O:ibr ..

'!j[T=2 ..

..

9.IT=2.11 3'T=7. 3t~7.4

.00.' 0.23i

::[OOi 1.74,

).54 0.69 o.SSi --,:oc, -, ;:42. 13.C! 3.91 4. it5l"--....::= 201 . 2~4:......:,1:!.:41. 2::::;....:_:::: O~" __~:L_ ["'"OM 0.70 0.15' TI' i .,.. . " ···1

o:Dii: 0.24! 0.00 . . i:8oi2.2~, . 3.07 i3:9:l·4.794.5i:

.

"'20.39

.... 0.84

1.42:

Figure 4.4-1. spreadsheet calCulation of Changing Pushover Forces, alpha, and PF from clven Masses and calculated First Mode shape, Flxed·/lase Model

4.8.5

Masses and cravlty Loads

For pushover analysis, masses are needed only to derive mode shapes and pushover forces. Because floor nodes were slaved together as assumed rigid diaphragms, only lumped floor masses were needed for the eigensolution. Model masses are shown in Figure 4.4-1. Conservative dead loads were applied to appropriate model elements as gravity loads. The conservatism (estimated as 10-15 percent) is expected to make up for exclusion of "Typical Service Live Loads" recommended by Methodology Section 9.2.

4.4

pushover Analysis

Capacity curves resulting from the pushover analysis are shown in Figures 4.4-2 and 4.4-3. Each curve is actually the composite of curves

a-'8

from separate analyses needed to capture degradation effects in some elements. The following sections describe derivation of these curves.

4.4.1

SoFtware Considerations

Solution strategies of different software packages can affect analysis efficiency and, if not understood by the engineer, correctness of analysis results. The first issue, as discussed in Methodology Section 8.2.1, involves a choice between direct nonlinear analysis and a series of linear analyses. This study used DRAIN-2DX for direct nonlinear analysis. The Methodology requires software capable of P-delta analysis and eigensolution. Other useful features for pushover analysis, some of which are offered by DRAIN-2DX, include:

Appendix B, Barrington Medical center.

1

App

r

----..,.'

.------------------------------------------

-


3000

2'

1.43

8

1~45:

'" Ii

--;-.."

~

---:----. Tool

~

~

1500

~

--r---i ·····.. ···--f

1.46

.t: (J)

Gis

m

---:----., 1".001

~

.,"

'"

pasO

~

03br

J/\)~,

Max[

{3b W II M chanls

Wan

I~ Fn me

1000

Oe 6 Ma

sr

=

~,d:'!-

c-

-7

Wall

I olSS

oiLS

I'

_
"

't 2000

------y

I

Shear Canae Iv

2500

1:36

W.II~

WaY 10

/'waIiO 3

"

Yield In

"" 'W"'W

Oe amatl n Limit 10 =Irnme ~lal. 0 cupan y LS = Life afety SS = Stru tural S obllRy Pe ~ormar ~e Pol s. • Oe ~ = O. Ign EQ Max = Maxim mEQ

500

--1.

1".47'

N. e: Col mno.

10

Model

0

eluatac as Se ondary

EI manta

4

3

2

0

5

B

7

6

9

10

Roof Displacement [in1

Figure 4.4'2. Annotated Capacity curve for Fixed·Base Model (see Update section 1.4.2 regarding performance point displacements)

1.42:

1.42: 'and

3000

I

2500

:;y

ColA

Wall

PII. , 1:ld /

Pile' 1.1

!se

'" Ii

~

I 1500

malysis;,

ce !S of IX for

" " m

.t: (J)

.'"

." I 2,.113

"~.~--'

1000

.'

500

l S

Ico

I~

"D';"4'0

Max

0.s6

Da10 "",lion ImH. 10 = mmedla Ie Occ pancy LS = en. sa .'Y, .... = ".ru

Fn me HI glng

-

MaxD

Perla rmance Points

ItU18' :

•

,~

6=S 011 Ty," 6,0= SOIlT 'Pe 0

I

110

Mode

Note:

0 0

Mach "ism

14a

I J

lii

if not ,

,"

./

Pu.

$v-

'icol L ,

"'.2000

~

lie Vie

,;

2' 0 0

!/all C·

1

land lover

2

3

Colu"" seval ated a S.con dalY Eh menta

4

5

6

7

B

9

10

Roof Displacement [in1

Figure 4.4·S. Annotated capacity curve for Soft·FOundatlon Model (see Update Section 1.4.2 regarding performance point displacements)


a-,.


• • • •

Displacement controlled solution or other stability controls Event-to-event strategies User-controlled yield tolerances or overshoot factors Graphics and data extraction tools for post-processing

Deriving and ApplyIng Pushover Forces Pushover forces were derived in accordance with Methodology Section 8.2.1. First, the initial, elastic mode shape was computed by DRAIN-2DX and the value at each floor was multiplied by the corresponding floor mass to derive an initial set of floor forces (see Figure 4.4-1). Only the relative values of forces from one floor to another, not the absolute values, are important. The initial force pattern is used throughout the pushover even as the structure model yields and the first mode shape changes (refer to Methodology Section 8.2.1). For this study, a series of analyses was required to capture degrading effects. This provided opportunities to update the mode shape and pushover forces between analyses (see Figure 4.4-1). Despite an effective mass coefficient at times as low as 70 percent (see Figure 4.4-1), only the first mode was used for this study. (Refer to Methodology Section 8.2.4.) 4.4.2

4.4.8

Monitoring Shear-Critical Elements In DRAIN·2DX Barrington Medical Center had potentially shear-critical frame columns and walls. Because DRAIN-2DX does not offer a shear-yielding element, it was necessary to monitor column and wall forces at each stage of the pushover and to adjust the model where necessary. The process of monitoring degrading frame columns is described in Section 4.4.4. For the fixed-base walls, modeled as equivalent columns, initial pushover results showed that first story walls would probably reach their shear capacity before their flexural capacity (depending in part on the level of

a-20

column degradation and resulting wall-frame interaction). The wall flexural properties were then revised so that "yield" would occur at the proper stage of the pushover.

4.4.4

I If

I

Modeling strength Degradation In DRAIN·2DX !, Barrington Medical Center had potentially shear-critical frame columns and walls. In accordance with Methodology Sections 9.5.4.1 and 9.5.4.3 Commentary, shear-critical columns (but not walls) are subject to degrading strength at moderate to high ductility demands. Degrading behavior is reflected in Methodology equation 9-4 and Figure 9-11. As discussed in Methodology Section 8.2.1, degrading components must be modeled different from ductile yielding components. When ductile components yield, they maintain their internal forces through additional displacements. When degrading components yield and reach a critical ductility demand, they release their internal forces to adjacent components. If degrading elements are modeled as ductile, the model will overestimate base shear by the amount held in all such elements. For this building, ductile modeling of degrading frame columns would have overestimated the pushover base shear by 30 to 50 percent. DRAIN-2DX does not offer an element with degrading strength. In order to model degradation, critical elements had to be removed from the model and replaced with softer, weaker versions (see Table 4.4-1 and Figure 4.3-1). The pushover analysis would then start again with the revised model and continue until more frame columns became critical (i.e. ready to degrade). A series of models was thus used to represent the degrading nature of the building. For example, consider Figure 4.4-2. The first portion, starting at the origin, is valid until a group of frame columns reaches a ductility demand of 2. Analysis results indicated that frame columns in stories 3 through 6 yield in flexure when roof displacements are between 0.20 and 0.35 inches and reach twice the yield drift


Ta,

co.

Iien Axl

MO lin' ShE Yle

-

lin·

(rep disp first secc floo disp anOI

toge Figl degl

moe dem initi recc may typi, degl Met wit!: strel func Tab onl)

4.5

4.5. Figl

App

~f~---------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

~'

~------------------------------------------------------------------------

,e

Ta,,'e 4.4,1. properties Used to Model Frame Columns With Degrading strength and stiffness

~re then -roper

Model property ,

,.'

.',

Initial "

..

3400

3400

250 (50%)

250150%)

74088

7400 110%)

3700 (5%)

snear Area Iin'J

504

300 (60%)

100120%)

YleJd Moment

2016

1200 160%)

400120%)

lly

Mom of Inertia lin')

with dation,

;ions hover sed ns cries of ding e first I

lin·kl

(representing a ductility demand of 2) at roof displacements of 0.40 to 0.60 inches. Thus, the first partial curve is valid until D" = 0.60". A second model, with partially degraded columns at floors 3 through 6, does not apply until the roof displacement is about 0.60" and is valid until another set of columns becomes critical. Taken together, the series of partial curves shown in Figure 4.4-2 approximate the behavior of a single degrading model. While it is impractical to change models whenever individual columns reach demands of exactly 2 or 4, jumping from the initial model to a final fully-degraded model is not recommended either because intermediate models may represent key points in the analysis. Table 4.4-1 gives model properties of the typical frame column in its initial, partially degraded, and fully degraded state. The Methodology provides no quantitative guidance with respect to post-yield shear stiffness, axial strength, axial stiffness, or degradation rate as a ' function of ductility demand. Thus, values in Table 4.4-1 are based on engineering judgment only.

4.5 frame

Ire

nd

FullY Degraded

504

Axial area Iin'l

i, they anal ; yield elease .If the mount ductile d have ) to

"

3400

-E Iksil

ftion

4,1 amns ngth at ling on 9-4 Jgy Je

partiallY Degraded

Identifying limit states on the capacity Curve

4.5.1 EVents Yielding. As shown in Figure 4.4-2 and Figure 4.4-3, initial frame column hinging does

center. Appendix B, Barrington Medical center

not significantly reduce the structure's lateral stiffness. A noticeable turn in the capacity curve occurs only when either walls (Figure 4.4-2) or piles (Figure 4.4-3) yield. This supports the decision to treat exterior frame columns as "secondary" elements. Yielding of slab-beams at gravity columns does not control and is not shown. Other premature failure modes, such as bar buckling or splice failure, were not explicitly modeled and were assumed not to control based on visual review of design details. Bar buckling in this building should be carefully checked, however, since ties in wall boundary elements are small and widely spaced. This concern may be partly mitigated by limiting allowable compression strains to low values in deriving element properties. Sliding along wall construction joints also was not modeled (refer to Section 4.3.4). Mechanism. For the fixed-base model (Figure 4.4-2), a story mechanism is created when all three of the first story wall groups reach their shear or flexural capacities. In the soft-foundation model (Figure 4.4-3), a pile mechanism develops instead. Although a mechanism indicates essentially no remaining lateral stiffness, it does not necessarily represent the end of the pushover. Deformation controlled elements may sustain displacements beyond mechanism as long as they remain stable with respect to gravity loads, that is, as long as they do not collapse from P-delta effects. For this reason, me capacity curve can extend beyond mechanism.

Building Limits 4.5.2 Methodology Section 11.3 sets limits for overall building response. Section II. 3.2 limits degradation to "20 percent of the maximum resistance of the structure." As noted above, frame columns would account for one fourth to one third of the structure's base shear capacity if they did not degrade. Because the columns are judged to be secondary elements (see below), this apparent violation may be overlooked. Drift limits are given in Methodology Table 11-2. Barrington Medical Center has a

B·21


typical story height of 12 feet and a roof height of about 79 feet. The Life Safety drift limits (at 0.02 in/in) are thus 2.9 inches in each story and 19 inches at the roof. This is well beyond yield, mechanism, and the expected displacement demand (derived below), so the drift limit does not control the evaluation. For Immediate Occupancy, the roof drift limits are 9.5 inches total or 4.7 inches beyond the effective yield displacement. While this limit might be reached, it would not control the evaluation. A rough check at the Structural Stability performance level gives a drift limit of .05 or a roof displacement of 48 inches.

4.5.$ Element and CompDnent Limits Categorization. Methodology Section 9.5.4.1 defines force- and deformation-controlled actions. For this building, the walls are deformation-controlled primary elements. Slabs and gravity columns are secondary elements. The piles, not specifically addressed by the Methodology, are primary elements where they support the walls and are considered deformation-controlled because slipping of friction piles is expected to be repeatable over many cycles. However, as noted above, the foundation was modeled to assess overall softening effects, not to evaluate specific foundation components. The exterior frame columns, because they degrade and become shear-critical, are considered force-controlled. They would normally be considered primary elements due to their significant initial stiffness. As force-controlled primary elements, however, their early yielding would stop the pushover before the walls could develop full strength. This was considered an unreasonable representation oCthe building's capacity, so the exterior frame columns were allowed to deform as secondary elements as they could support gravity loads. The Methodology provides no guidance regarding post-yield axial capacity of shear-critical columns, so engineering judgment on this point is critical. For this study, degrading columns were kept in the model as long

a-22

Table 4.5'1. RDDf DIsplacements CO"espondlntl to Deformation Umlts

ineio wou Stat pusl Life (Me fran acce 60 i colu

Primary WailS, EQulv. Flexure

no yield

no yield

no yield

secondary elements but were earmarked for gravity load retrofit regardless of expected deformations. Limits. The Basic Safety Objective (see Methodology Section 3.4.1) requires checks of both Life Safety and Structural Stability deformation limits. Roof displacements corresponding to stages in the pushover analysis where deformation limits are reached are summarized in Table 4.5-1 and noted on Figures 4.4-2 and 4.4-3. Columns. Frame columns are checked as secondary elements controlled by shear. Methodology table 11-4 allows inelastic deformations depending on column details and axial load. Although tie spacing is less than d/2, it is still large by current standards. Also, the ties are small, lack substantial cross ties, and have unknown hooks and extensions. Table 11-4 also allows inelastic rotation in columns with small axial loads; the critical third story columns have P/A.f, values of .07 to .11, at or near the table limit of .10. Because the frame columns barely comply with either table condition, engineering judgment must be applied here. For this study, the columns are treated as eligible for inelastic rotation, but are earmarked for gravity load retrofit regardless. Were the columns not eligible for inelastic rotation, or were they needed as primary elements, then by Methodology table 11-4 no


Sh01

stor: inch the: are:

"eql tabl. tang com equi inel: plas (Me wall is re the I Stab are I reml

(poc rota' chee was an a cap, rupt Met rota: high Why than

~


~

-----------------------------------------------------------------------

---

inelastic, rotation would be allowed and their yield would mark both the Life Safety and Structural Stability limit states. ~lIlty Table 4.5-1 is constructed from detailed ,8" pushover analysis results. For example, with a Life Safety inelastic rotation limit of .01 radians 3" (Methodology table 11-4), and considering that frame rigid zones will force all drift to be accommodated within the column clear height of 60 inches, the inelastic drift limit for typical frame 2" columns is 0.6 inches per story. Analysis results show that this inelastic drift is reached for most field stories when the roof displacement is about 3.8 inches in the fixed-base model, about 4.2 inches in the soft-foundation model. Walls. Walls modeled as equivalent columns are most conveniently checked against "equivalent" hinge rotation limits in Methodology table 11-7. Table 11-8, which is based on of tangential drift with an inelastic shear drift component, can not be used effectively because the equivalent column model is unable to reflect ysis inelastic shear drift. Table 11-9 was not checked. For the fixed-base walls, the Life Safety ures plastic rotation limit is .0033 to .0040 radians (Methodology table 11-7), depending on each s wall's axial load and shear when flexural capacity is reached. All three walls reach their limits when the roof displacement is about 5 inches. Structural ld Stability rotation limits of .0060 to .0080 radians i/2, it are reached at about 8 inches. It is important to :ies are remember that the wall details in this building (poorly confined boundaries, etc.) may limit Iiso rotation and ultimate compressive strength. To ~I check this, a separate moment-curvature relation ave was derived for the critical wall at grid 0-3 with ble an axial load of 850 kips. It showed a curvature !Iy capacity of 2.4E-4 radianslinch Iim:ited by bar rupture (with concrete strain at only .002). ng y, the With a plastic hinge length of Ill" (per Methodology Section 9.5.5.7), the ultimate retrofit' rotation capacity is .027 radians, substantially higher than the Methodology's limits. It is unclear mary why the Methodology limits are so much smaller than a moment-curvature analysis suggests.

1ng

_raJ

-

-' -'

-

:enter


Piles. The Methodology provides no limits for slip of friction piles. Instead, engineering judgment is applied to pile slip at the performance point (see Section 4.7.2). The emphasis is not on the piles themselves but on the impact of pile yield on overall behavior. Slab Beams. Slab plastic hinge rotations did not approach the limits of Methodology table 11-6 for secondary elements with Continuity Reinforcement. However, these elements were not carefully modeled.

4.6

Determination of Demand and Performance point

4.6.1 Performance Objectives This study evaluates Barrington Medical Center relative to the two-part Basic Safety Objective defined in Methodology Section 3.4.1. Expected performance in both the Design Earthquake (DE) and the Maximum Earthquake (ME) must be determined.

Elastic Response spectrum 4.6.2 The 5 percent damped demand, called the elastic response spectrum, is derived from parameters described in Methodology Section 4.4.2.4. For this building, the Seismic Zone Factor, Z, is 0.4, and the Near-Source Factor, N, is taken as 1.0, although a factor of 1.2 or 1.5 may be warranted. Two soil types, B and 0, are considered for this study. Table 4.6-1 gives coefficients; Figure 4.6-1 shows elastic response spectra. Table 4.6-1. Seismic Coefficients for Elastic Response spectra (Z .4, N 1.0, 596 dampedl

= =

Design Earthquake E = 1.0

CA Cv

= .40 = .40

CA = .44 Cv = .64

Maximum Earthquake E = 1.25

CA = .50 Cv = .50

CA = ,55 Cv = .80

a-25


1.4

:§i

c: 1.0

~ GJ

0.8

)

0.6

DOBa

"

\ "'\

"-

"-...

"'- " I'-..

iii

~

"

MaxB

12

0.4

DeaD

~

-

..............

B= Sol Type B D;; Sol TypeD Del= t 0019" ee ,,"quake E=1.0 Max = M axlmum anhqua ., E=1. 5

/Ji 02

MaxD

""

~

............

r---.....

--I-

r--....

r-......

0.0

o

2

3 4 5 Spectral Displacement [in]

6

7

8

FIgure 4.6·1. Elastic RespDnse spectra (Z=.4, N= 1.0,59(, damped)

4.E

cap per is c sho par

coe for

sho cal! Fig con

PF poi:

FIgure 4.6,2. PartIal spreadsheet calculatIon for ConversIon of capacIty curve to capacIty spectrum, Flxed·Base Model

B-24

Appendix B. Barrington Medical center.

l

ApI

r

~\

~

~~

------------------------------------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

-------------------------------------------------------------------------------

0.30 0.25

:§ <:

~

I

0.20 IV

~ 0.15

i

//,,,

"V

~

,re 4.6-1.' 'Spectra' rtampel/J'

~I,,-S1

,, ,

~

0.10

en

Fixed-Base

/ ~~ ~

,.

BV

,.

.J

-'

.,.---

I 0.00 0

f-- - -1 - - - Soli Foundation

Blllne r Represen ~tlon Contra Points IV = If.cllve Inltl I Vleld SV = ~lfecllve Su sequent Vie d U=E fectlve URI, ",te, at Sd > 6" FI ed-Base: SI = .22g S II-Foundallo ~: Sa = .18g

I

0.05

---

1

2

3

4

5

6

Spectral Displacement [in]

Figure 4_6-S_ Capacity spectra, Fixed Base (-) and Soft Foundation (- -) Models, With control PoInts for Bilinear RepresentatIon

4.6.S

J

Conversion of Capacity Curve to capacity spectl'Um

For comparison with demand spectra, the capacity curve is converted to spectral coordinates per Methodology Section 8.2. The converted curve is called the capacity spectrum. Figure 4.4-1 shows the first mode shape with calculated modal participation factor (PF) and effective mass coefficient (alpha) at various stages of pushover for the series of fixed-base models. Figure 4.6-2 shows some of the corresponding conversion calculations. Converted curves are shown in Figure 4.6-3. The somewhat jagged nature of the converted curves is due to the fact that values of PF and alpha were computed at a few discrete points only.


4.6.4

Bilinear Representation of capacity Spectl'Um

One Methodology procedure for determining effective damping and corresponding demand spectra is based on a bilinear representation of the capacity spectrum. The bilinear representation requires three points: the origin, an effective yield point, and an ultimate point that represents either collapse or any stage of pushover beyond expected demand. The area under the bilinear representation should approximate the area under the capacity spectrum. For buildings with degrading stiffness, the "yield" point at the "corner" of the capacity curve (or capacity spectrum) may be different on each cycle. This effect is represented by the partial curves in Figure 4.6-3. Each model has an "initial yield," which may best represent behavior in

B-25

-


Ca =:0.44 cv =:0.64

:

I

Ji"'.,

I

0.50 JIS.,

0.40 .!!! c

,

I !

I

i! !

~

I i'"

0.30

Ell jotl.

/'

'"

/< fo.'ors

0.10

~:;~I

I.-,i ,., 1.00

2.00

., -- r-...

'.:=~~, r:'~.h

~

ii,

~.~.'

3.45

cf' ;.

~= · ..•.

apla{;;

..

'

0.10

DeSI

' PoI~t du au

,nee. Point!

(ip';!

.,= ....... o.~ .....

-

5.00

k

Vidy =, ay =:

...........-

..........

4.00

...

...,... (SeE

. ".p'n OJ

~

l.--"'"

3.00

/'

\

V

II not .hown 1....1 V

0.20

0.00 0.00

!

,FI.4: 1,1 for 5

,

TalJ

'fFla iii:,) "-

Ma~

. 1.42.

coel= 3.99 SRa=; 0.23 SAY=: .. 0.41 SAd=; 0.53

Des

E=

... 1'09 . .... •.

-

slODe a =,' 0.60] de =: 2.85 sloDe b =; 0.005 ............ "e.= ! . -0.. 1.6 " .. •.

E=

. . ····,A"',,;;,

6.00

21%

Spectral Displacement rll1)

;

resp disc roof (Tal limi are i so tI Objl

Figure 4.6-4. partial Spreadsheet GraphIcal CIIlculation 01 perFormance pOint, Flxed·Base Model (Z = .4, E= 1.0, N= 1.0, Soli Type D, InItIal YIeld} (see upaate section 1.4.2.1

events with one or two dominant pulses, and a "subsequent yield," which may be better for long events with many inelastic cycles and potential degradation. (See Methodology Sections 4.5.2 and 9.5.4.1.) It is generally conservative to use the subsequent yiel(,l for computing demands. (See Update Section 1.4.2.)

4.6.5

Derivation of Demand spectrum and Performance Point (See Update Section 1.4.2.)

Iterative procedures are needed to find the unique "performance point." Figure 4.6-4 is a sample performance point calculation from a spreadsheet written to aid the necessary iteration. Performance point displacements are summarized in Table 4.6·2 and noted on Figure 4.4-2 and Figure 4.4·3. Table 4.6-2 illustrates a few trends for performance point roof displacements in this building: • Displacements are about 113 higher with the soft-foundation than with a fixed-base.

•

With Z = 0.4, ME displacements are about 50 percent higher than DE displacements.

Soil type D doubles the displacements relative to type B. Using.the subsequent yield instead of the initial yield increases the performance point spectral displacement by little more than the difference between the two yield values: about 0.4 inches with a fixed-base, about 0.6 inches with the soft-foundation. •

4.7

4.7.

(the of3 Earl dem inch softcom B 01 perf coni

performance Assessment

(See Update Section 1.4.3.) Performance point displacements (Table 4.6-2) must be checked against limits (Table 4.5-1). Global Building Performance 4.7.1 The building as a whole must be checked for stability, strength degradation, and excessive deformation (see Methodology Section 11.3). Pushover analyses showed no instabilities with

agai simi

the, type

B-26

Appendix B, Barrington Medical center,

l

ADD

,

-------l--


I ~, --------------------------------------------------------------------------

Table 4.6,2. Performance point spectral Displacements and corresponding Roof Displacements

(see update section 1.4.2.1 Flxed·Base Model

.'

-Design Earthquake

E ~ 1.0 Maximum Earthquake

E ~ 1.25

Soft;FoundlltIDnMOde/ Design Earthquake

E = 1.0 Maximum Earthquake

E = 1.25

Yield point '.'

SOilt"/Pe,S . ~

1.5"

Drf = 2.2"

5,

~

Subsequent

5,

~

1.9"

Off = 2.7"

5,

~

Initial

5,

~

2.2"

D"

3.1"

5,

~

5.2"

D"

~

7.4"

Subsequent

5,

~

2.6"

3.7"

5,

~

5.6"

0"

~

7.9"

.,<.","" '.

5,

~

2.0"

D"

~

2.9"

Subsequent

5,

~

2.6"

D"

~

3.6"

5, - 4.5" 5, ~ 5.1"

Initial

5,

~

2.9"

D"

~

4.2"

5,

6.6"

0"

5, = 3.5"

0"

~

5.0"

5, = 7.3"

0"

Subsequent

component Force and Deformation Checks

(See Update Section 1.4.3.)

~e

,d for e ). 'ith

center

~

Initial

4.7.2

: 4.6-2) ).

D"

~

Columns. The fixed-base Life Safety limit (the capacity) was reached at a roof displacement of 3.8 inches (Table 4.5-1). The Design Earthquake performance point displacement (the demand) is 2.2-2.7 inches on soil type B, 4.9-5.4 inches on soil type D (Table 4.6-2). For the soft-foundation model, a capacity of 4.2 inches compares with a demand of 2.9-3.6 inches on type B or 6.3-7.2 inches on type D. Expected performance therefore depends on site soil conditions. Checking the Structural Stability capacity against the Maximum Earthquake demand yields similar conclusions: with either foundation model, the capacity falls between the soil type B and soil type D demands.


= 4.9" ~

5.4"

:''i.e'·'"'' 1"'\'.';"»""'." ,I."',""""""",•• , I":,"; ,

Le

out 0.4 vith the

,

5,

'elative

lbout ts.

..

Initial

respect to gravity loads. Degradation was discussed in Section 4.5.2. All performance point roof displacements in the Design Earthquake (Table 4.6-2) are less than the 19 inch Life Safety limit, and the Maximum Earthquake displacements are less than the 48 inch Structural Stability limit, so the building as a whole meets the Basic Safety Objective.

E=1.0,

,

soi/TYpeD 3.5" Drf 3.8" D"

.

~

<

0"

~

6.3"

0"

~

7.2"

=

9.4"

~

10.4"

Walls. For the fixed-base model, the Life Safety roof displacement limit of 5 inches exceeds the Design Earthquake demand. With soil type B, the building has some margin, but on soil type D, the capacity just matches the demand. Similarly, the Structural Stability capacity of 8 inches is sufficient on soil type B but barely acceptable on soil type D. With a soft-foundation modeled, the walls do not reach their shear or flexural capacity, so Methodology deformation limits do not apply. Instead, pile actions should be checked. Piles. Piles are modeled to assess potential softening of building response, not to evaluate specific foundation components in the way that columns or walls are evaluated. The important questions are whether the assumed bilinear pile model applies at the performance point and whether modeled pile behavior affects the building's ability to meet a given Performance Objective. From the pushover analyses, maximum soft-foundation pile slips corresponding to different performance points are: • Design Earthquake, Soil Type B: .25 to .50"; Type D: 1.0 to 1.25" •

Maximum Earthquake, Soil Type B: .50 to .75"; Type D: about 2"

a-27


I These are judged to be within the bounds of the simple bilinear pile model (Figure 4.3-3), although a 2 inch slip probably approaches the limit of plastic behavior. Had slip values been much higher (on the order of a foot or more), the bilinear model would not apply, and analysis results might have been invalid. Most importantly, development of a pile mechanism in the soft-foundation model protects the walls from demands requiring their full shear or flexural capacity. This conclusion can significantly impact the scope and cost of retrofit and even the decision to continue occupying the building. Confirmation of the pile model is therefore warranted.

4.7.S

DeFiciency summary

(See Update Section 1.4.3.) With respect to a Basic Safety Performance Objective, deficiencies are summarized as follows:

With Soil Type B •

Because of expected degradation and unknown tie details, the post-yield gravity load carrying capacity of exterior frame columns is unreliable.

With Soil Type 0 • •

•

•

Expected deformations in exterior frame columns are unacceptable. Because of expected degradation and unknown tie details, the post-yield gravity load carrying capacity of exterior frame columns is unreliable. Expected deformations in interior wall groups are only marginally acceptable with a fixed-base model. (A soft-foundation may protect the walls.) The bilinear pile model requires confirmation if its beneficial effects are to be accepted.

a·28

5. 5.1

conceptual Retrofit Designs performance Objectives (See Update Section 1.4.4.)

I

!I i,\l4 >--'Z'<§':"'"

Retrofit designs for Barrington Medical Center ! were developed for two Performance Objectives: ! the Basic Safety objective already described and used for evaluation, and a lower, "economically driven" objective as described in Methodology table 3-4c. For brevity, only the Design Earthquake with soil type D is considered here. The "Economic" objective requires only Structural Stability in the Design Earthquake. Tables 4.5-1 and 4.6-2 show that the building as designed might meet this standard already. The walls are acceptable. With a fixed-base model, the secondary frame columns are barely acceptable as well. With the soft-foundation model, however, the Structural Stability roof displacement limit of 6.2 inches (the capacity) is exceeded by the expected Design Earthquake roof displacement of 6.3-7.2 inches (the demand). Column detailing remains a likely deficiency as well. Required structural work for the two objectives is summarized in Table 5.1-1.

5.2

Selection of Retrofit Elements

FIXE

SOf ion

-• •

5.2

5.2.1 Structural Considerations Retrofit requirements can be met with a variety of structural schemes. Considerations include: • Because the building is already stiff, additional frames or coupled piers will not be as effective as additional walls deep enough to match existing stair and elevator cores. As an alternative, diagonal steel braces forming an "exoskeleton" may be sufficiently stiff. •

.,Tal

(SeE

shol infil CODI

adv:

• •

If the soft-foundation is an accurate model, then ultimate behavior is controlled by pile yielding. Therefore, where walls are stiffened or added to limit frame deformation, they may require foundation strengthening.

•

APpendix B. Barrington Medical Center ,

l


t Table 5.1,1. Required Retrofit work for Different performance Objectives (Soli Type 0) (see update section 1.4.4.1

....

· . .···",odei

I Center' :tives: I and .cally ogy

lere.

Fixed·Base

ditional ffective :h

Enhance gravity resistance of frame columns.

Enhance gravity resistance of frame columns.

Limit deformation in frame cols.

Assume foundation ok.

sofHoundat Enhance gravity resistance of frame columns. ion Limit deformation in frame eols. confirm pile model with tests.

•

i

mit of

1S

"EConomiC" Objectlve- ;,. (structura/stablllty In· DesIgn Earthquake)·

Assume foundation ok.

~ver,

lent of ling

.• llaslc Safety Objective .. '(l.kesaFtJty in Design Earthquake) .•.•......

strengthen or add walls .

y

:e. ng as The jel, the able as

2

•

If walls are added, column deformations in the stiffened building may be acceptable. If not, supplemental gravity load-carrying elements, such as steel columns, can be designed to pick up gravity loads as the concrete columns crack and lose integrity. Alternatively, frame columns can be wrapped or plated to maintain integrity after cracking. Or, part of each beam-column connection can be cut, effectiv\!ly lengthening the columns and making them less shear-critical. These approaches are beyond the scope of this study.

5.2.2 Practical Considerations Wall strength and stiffness can be added by shotcreting against existing core walls or by infilling exterior frame bays with cast-in-place concrete. The latter approach has practical advantages: • Effectiveness of shotcrete on core walls is limited by openings and setbacks. •

del, pile ffened ey may

Interior walls are probably limited by pile capacity already, so strengthening these walls may require foundation work for which access will be difficult.

•

At the interior walls, coordination with mechanical and elevator systems may be difficult.

I center


Enhance gravity resistance of frame columns. Limit deformation in frame cols. Accept pile model.

•

Perimeter infill walls provide an opportunity to deal with torsion and bi-directional issues ignored for this two-dimensional study.

•

Although some strengthening may be required, existing grade beams at perimeter frames provide convenient footings for new infill panels.

5.2.3

preliminary Sizing with the Capacity spectrum

For this study, new walls are proposed; they will add capacity and limit column deformations as required. A preliminary approach to sizing these walls uses hypothetical capacity spectra to find performance points within required deformation limits. Spectral values at the hypothetical performance point are then converted back to absolute values, and the required strength of additional shear walls can be determined. Further assumptions include: • The new mode shape matches the existing mode shape. Assume modal PF = 1.4; choose appropriate alpha for hypothetical new performance point from evaluation data. • Initial and post-yield stiffness of the hypothetical capacity spectrum match those of the original building.

B-29


-

0.40 Hypo hetical Perfo manca Pc int at Sd < 2.7" ......

~

.9 0.30

I'.:.....

c: 0

:;::

r

I!!

~ CD

l:l «

0.20

"" ""

"

\ Hypothet cal billne ar

-

............. r-

capacity spectrum Slopes match or ginal bUil ~ing (Fig

I!! u

CD

Co en 0.10

... u-.. "

I

.,

Y'''''U ..

w

ultimate apacltiet are raised as necessa

IV'

0.00 0.00

1.00

2.00

3.00

4.00

Spectral Displacement [in) Figure 5.2-1. Hypothetical capacity spectrum for preliminary sizing Of Retrofit Elements

•

For the hypothetical strengthened building, the initial yield point is appropriate.

•

New walls have approximately the same yield displacement as existing walls.

•

Effects of new wall weight and material properties can be ignored. For the fixed-base model, the Life Safety roof displacement of 3.8 inches corresponds to a spectral displacement limit of about 2.7 inches. Figure 5.2-1 shows a hypothetical capacity spectrum that satisfies this requirement. The hypothetical performance point acceleration is .26g. With alpha = .83 at the hypothetical point, this corresponds to a base shear of about 2900 kips, about 600 kips more than the pre-retrofit capacity. Similarly for the soft-foundation model, the limiting spectral displacement is about 3.0 inches, the hypothetical performance point requires

a-so

.27g, and the new shear capacity required (with alpha .83) is 3000 kips, about 1000 kips higher than the original capacity. By comparison, achieving the "Economic" Performance Objective requires even less additional capacity. Figure 5.2-1 suggests that a minimum scheme might still rely on substantial inelastic behavior. Because new infill panels will engage poorly confined and lightly reinforced existing columns as de facto boundary elements, high compressive strains should be avoided. Therefore, while two panels may be sufficient for the "Economic" Objective, four panels are assumed for a conceptual Basic Safety retrofit scheme. Four panels also provide reserve capacity against potential torsion and should be able to accommodate exits, windows, or other architectural requirements. New piles and pilecaps will also be required. Figure 5.2-2 shows the

=


COl

as! eat

S.l

Sol

wa bu cal ev:

lip

T-------------------------------------------------------------------------------

~;


-------------------------------------------------------------------------------

F l---t------r~.....,..~-I--~~-i_

® TYP. I 21'-4\:1.

E

®TYP.

@-.@-

- - - - - ® (N) 'NFILL WALL ®

(N) SHOTCRETE WALL

© ®

(N) PILE & PILE CAP

-

--I

(N) TS COLUMN

~~ }-J-----1.--l- ~ ¢t»."t[J~· NOTE SOME (E) ELEMENTS NOT SHOWN

20'-5"

0

}-J-----1.--l104'-0"

20'-5"

C

20'-5"

8 J-+---+---E!!!l'

'~OL

,

-

@- . -

@-

@- - . @-

~)

8" WALL, TYP.

@-

~E)Ol.,EXTTYP. I'

©

21'-4\;.

.. TYP.,

A

22'-4"

21 '-6"

21'-6"

® 'ith gher !ctive ;cheme vior, y

umns as ive : two

." ,ur

22'-4"

21'-6"

5

6

Figure 5.2·2. conceptual Plan of Basic safety RetrOfit Scope Of work

conceptual Basic Safety retrofit scope of work, assuming that similar work will be required in each direction.

5.3

comparative EValuation by Product 1.2 MethodolOgy (See Update Section 1.4.4.) Figure 5.3-1 shows the fixed-base and soft-foundation capacity curves with four infill wall panels together with the curves of the original building. It is important to note that the retrofit capacity curves represent "comparative evaluations" more than they do retrofit designs.

Actual design must comply with Building Code force levels, strength reduction factors, and allowable deflections that may differ from those in the Methodology. The analyses presented here merely apply the same Methodology procedure to a hypothetical upgrade. Relative to the original building, the retrofit models include: • New wall elements, as described in Figure 5.2-2, modeled as equivalent columns. •

Existing columns engaged by new infill panels and acting as boundary elements: checked for ultimate concrete compression strain of .003.

lilecaps he

II center


a-S1


4000 /

3500

~

~

1500

500

-

I / --, --r ," -'/1. f

2000

~ 1000

,,

/

~ 2500

~ .,

DeaD

V

~ 3000

~ g

L _ ;.. Ret pill Fix. d-Basa JlIfIIIT

1--

.8iIIiI!rDeaD

Retrol Solt-, oundat on Orlllin I Fixe. Ba ••

"'DeaD

---

..2!ISlnal S It-Fou dallon DeaD

-

I

, J

,,

Pe orman e Poln De D = De Ign EC , Soli , ype D

,

o

o

1

2

3

4 5 6 Roof Displacement [In)

7

8

9

10

Figure 5.11-1. capacity Curves and perFormance points lor Original and Retrofit structures (see Update Section 1.4.2 regardIng performance point dIsplacements.!

• •

Separate material properties for new elements: f'c = 4 ksi, fy = 60 ksi.

Additional pile stiffness and strength at ends of new walls, for the soft-foundation model. Figure 5.3-1 also shows performance points for the two retrofit models. With four panels, the fixed-base model's expected performance requires a roof displacement of 2.0 inches, less than half the expected displacement for the original building and well below the column Life Safety deformation limit. The soft-foundation roof displacement is 4.3", about two-thirds of the original building value but not yet below the column LS limit of about 4.2". With a soft, yielding foundation that controls overall behavior, additional walls alone are of marginal value unless foundations are also strengthened. With exterior frame bays stiffened by infill, beams in adjacent bays become subject to high local rotations and

i'

must be checked against limits in Methodology table 11-3. At the calculated performance points, . I, bearns actmg as secondary elements were found to be acceptable.

I

G.

6_1

Assessment of the product t.2 MethodologV Damage Prediction

I ! .

Barrington Medical Center was inspected twice : after the 1994 Northridge earthquake, classified as "unsafe," closed, and later demolished. Figure 6.1-1 shows typical severe column damage along the south facade. Exterior frame columns in the second, third, and fourth stories are visibly cracked within their clear height in classical X-patterns indicative of shear failure. North and west side columns show similar damage to a lesser


i

deg: sho' SOUl

darr buc: buci X-c colt and the elen the for Stat Wer haVI

loss

r --------------------------------------------------------------------------------

~,


---J --------------------------------------------------------------------------------

FIgure 6.1,1. EXterIor Column Damage on LIne A 'rom January 1994 Northridge Earthquake

ogy oints, lUnd to

lamage mns in bly I 1 and 1 lesser

i

degree. Beams and connection panel zones do not show damage. As shown in the photographs, the southwest corner colurrm (grid A-I) is severely damaged, with extensive spalling, shortening, and buckled reinforcing. Shortening of colurrms and buckled window mullions indicate that typical X-cracks extend through the full width of the colurrms and that the colurrms have lost both axial and shear capacity. No information is available on the performance of interior walls or foundation elements. In terms of Methodology performance levels, the observed damage is beyond the level acceptable for Life Safety and is at or near the Structural Stability limit. Although the interior shear walls were not observed, some of the exterior colurrms have clearly shortened, indicating at least a partial loss of gravity load resistance. However, they did

I Center

I


:d twice ified as

not collapse, and they may still be capable of holding their loads through additional lateral deformations. Note that assessment relative to Performance Objectives requires some knowledge of ground motion at the site. Pushover analysis predicted flexural yielding in third and fourth story frame columns under relatively small displacements, followed by strength degradation and shear-critical behavior. However, no significant flexural cracking is visible in available photographs. Second story colurrms, with greater flexural strength, were not expected to be as critical as photographs suggest they were. Actual yield strength of 60 ksi, as opposed to the erroneous model value of 40 ksi, may explain why frame colurrms appear to have been even more shear-critical than the model assumed.

a-JJ


The pushover analyses showed that exterior frame columns, while the first to yield, were not as important to overall building response as the interior walls. With no inspection of these walls, a full assessment of the Methodology's damage prediction capabilities is not possible.

6.2

Comparison with preliminary EValuation Findings

FEMA-178 Evaluation Statements (if coupled with engineering judgment regarding concrete column shear capacity) uncovered all of the deficiencies found by inelastic pushover analysis, except for effects of a potentially soft, weak foundation. However, the Evaluation Statements and elastic analysis use a broad brush, pointing to high shears and low reinforcing ratios, but nearly missing the shear-critical degrading columns and failing to distinguish their impact on building performance. By contrast, the Methodology's pushover analysis showed that the columns yield early and degrade, controlling performance even as secondary strength elements. The inelastic analysis also quantified potentially significant soil-structure effects. Although software limitations and other practical considerations preclude assessment of some complex behaviors (e.g. potential torsion or higher mode effects), the Methodology's nonlinear static procedure are still expected to provide a more complete and more useful picture of expected performance than is linear elastic analysis.

6.3

Comparison with Inelastic Tlme·History Analysis

(See Update Section 1.4.5.) Limited inelastic time-history analyses were executed as a rough check of performance point displacements predicted by the Methodology. Four near-field acceleration records, each with components in two directions, were selected. For each record, the given components were transformed to fault-parallel and fault-normal components. Scale factors were computed so that

a-:54

-

the average spectral acceleration of the eight histories would be O.64g for a structure with a I-second period. That is, the records used as time history input were scaled to match a single representative point on the elastic response spectrum for the Design Earthquake on soil type D. Time-history analyses were executed with DRAIN-2DX. Five percent damping was assigned to the first two modes. Only the soft-foundation model with fully-degraded frame columns was analyzed due to two software limitations: • Potentially shear-critical walls modeled as equivalent columns. In a pushover analysis, critical conditions are apparent, and flexural capacities are easily adjusted to simulate shear yield at the proper stage of loading. In time-history analysis, this can not be done, so results for fixed-base models (which require adjustment) are not reliable. •

Capacity curves were constructed with a series of models in order to capture degradation effects. In time-history analysis, this can not be done. Only the fully degraded model was analyzed, meaning that the model began each time-history run with already reduced stiffness. Figure 6.3-1 shows the soft-foundation capacity curve with combinations of maximum roof displacement and base shear taken from the time-history analyses. The Design Earthquake performance point, also shown, appears to overestimate displacement and underestimate base shear. Possible explanations for this discrepancy involve both the relative and absolute precision of pushover and time-history analysis. For this study, whether the pushover or the time-history results are closer to the "truth" is unclear. The most likely explanation for the discrepancy is that higher mode effects are considered in the dynamic time-history analysis but are ignored for this static pushover.

dis psc rei:

COl

sca or am ste for

PUi

in 1 ele

dUI

dru

Sir del

Unl

del


,

l

--f, . '--------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

---i,

a 3500

time

3000

-- ......

2500

:. In

:;;0 0

h igned ion

"

~

2000

• ~:

g

IS

. rn . ~

1500

Q)

.r=

IS

;is, lral shear

Q)

III

1000

In

500

Ie, so

.'1

: ,

;

.. '"

..

-'

.s

m the nt, ment h the nd r the :r to

'sis

I.

.m

. ..... . .. - .. .~-'

I

.

·h

I I

. . . . ..

~m :.h

--~

~Parfor lnanee EQ,

~:I~I~

'n..

~ Istory F eak Re ponse g,h Cppltola I m,n ! ylmar S,I R naldl

Time

I o

1

2

3

oint,

0

•

4

5

6

7

B

9

10

Roof Displacement [in]

series not was each

..... -

I~ .g

0

lire

1

...":'

Mea n+1 sla~dard dey tion a 14 pol Is

"'

ri

" ~

V

g. h

!

FIgure

6.~·1.

capacIty Curve wIth peak Responses from Nonlinear TIme HIstory Analyses (Scaled and weIghted to O.Ug at T= 1.0 sec), soft·Foundation Mode/(see Update Sections 1.4.2 and 1.4.5.1

Other factors contributing to the observed discrepancy may include: inapplicability of pseudo-acceleration, velocity, and displacement relations, implicit in the pushover analysis, to conditions of high damping; use of time-histories scaled to the design spectrum at one period only; or numerical inaccuracies in the time-history analysis due to rough overshoot tolerances or time steps. In the end, if the Methodology underestimates force levels, it is only unconservative when the pushover analysis predicts nearly-elastic behavior in buildings governed by force-controlled elements. When the analysis indicates high ductility demands in force-controlled elements, damage will be indicated regardless of force level. Similarly, where performance is governed by deformation-controlled elements, an underestimated force level is not important since deformation levels determine acceptability.


Three observations about the reliability of inelastic analysis for this example building: • The simply modeled piles would typically yield in compression but not in tension. This led to "ratcheting" inelastic displacements that are probably not realistic. If displacements ratchet in one direction, it raises a question about the meaning of "maximum displacement" and which peak values should be used to gauge the pushover results. •

Even with the soft-foundation model (whose walls do not yield in pushover analysis), time-history results gave wall forces much higher than the expected wall capacities, leading to peak base shear values well above the pushover curve. This indicates that the time-history analyses were probably invalid for this building.

a-Is


•

However, even if this building had had stronger walls, its capacity curve (controlled by pile yielding) would have been the same, and valid time-history results would have been off the curve. This suggests that pushover analysis can underestimate component forces and/or overestimate roof displacements.

,_

References

General references (e.g. FEMA-!78) are given in the Methodology. Documentation for DRAIN-2DX, ETABS, and PCACOL is widely available. Only available documents specific to Barrington Medical Center and this study are listed here. I. "Barrington Medical Center." Cover sheet, ten architectural drawings, and seven structural drawings by Charles Wormhoudt AlA Architect & Associates and Eugene D. Birnbaum and Associates, Structural Engineers. February 17, 1964. 2.

"Olympic Barrington Medical Building." One volume of structural calculations by Eugene D. Birnbaum & Associates. Variously dated from 8/63 to 2/64 with additions 9/64 and 10/65.

B-3&

AI

E: At

C. a1 3.

"Report of Foundation Investigation, Proposed Building, Olympic Boulevard and Barrington Avenue, Los Angeles, California." Cover letter and 5 page report with Appendix, by LeRoy Crandall & Associates. September 20, 1963.

4.

"City of Los Angeles Department of Building and Safety Rapid Screening Inspection Form." Two forms for 11665 W. Olympic Boulevard. Dated 1-17-1994 and 1-19-94.

5.

Letter from Craig Comartin regarding Foundation Effects for Case Study Buildings. January 16, 1996.


pre

__...r_-----------------------------------------------------------------,


--...:. ~

APpendix C

Example Building Study: Administration Building california State University at Northridge i and

prepared by 'port

Nabih Youssef & Associates 800 Wilshire Boulevard, Suite 510 Los Angeles, California 90017

55 W. ' md

ldings.

center:

Appendix C, Administration Building, CSUN

c·,

f----------~

~'

--


Table of Contents I. Introduction .................................................................................................. C-5 1.1 Intent of Example Building Study ........................................................ C-5 1.2 Scope of Example Building Study ........................................................ C-5 1.3 Summary of Findings ...................................................................... C-6 1.4 Update ........................................................................................ C-7 2. Building Description ....................................................................................... C-8 2.1 General ....................................................................................... C-8 2.2 Structural System ........................................................................... C-8 3. Observed Earthquake Damage ........................................................................... C-9 3.1 1994 Nortbridge Eartbquake .............................................................. C-9 4. Preliminary Evaluation .................................................................................... C-9 4.1 General ....................................................................................... C-9 4.2 Dynamic Characteristics of Building .................................................. C-lO 4.3 Elastic Analysis ........................................................................... C-lO 5. Evaluation Of Existing Building By Product 1.2 Metbodology ................................... C-l1 5.1 General ..................................................................................... C-II 5.2 Structure Modeling ....................................................................... C-l1 5.3 Pushover Analysis ........................................................................ C-13 5.4 Seismic Demand ......................................................................... C-16 5.5 Response Limits .......................................................................... C-17 5.6 Performance Objectives ................................................................. C-17 5.7 Performance Evaluation ................................................................. C-18 5.8 Evaluation of Foundation Effects ...................................................... C-19 5.9 "Limited" Nonlinear Time History Analysis ......................................... C-20 6. Evaluation Of Strengthened Building By Product 1.2 Metbodology ............................. C-20 6.1 Retrofit Scheme ........................................................................... C-20 6.2 Dynamic Characteristics of Strengtbened Building .................................. C-21 6.3 Evaluation of Strengthened Building .................................................. C-21 7. Concluding Remarks ..................................................................................... C-21

AppendIx C, Administration Building, CSUN

C·I

,, r~-------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

~--------------------------------------

APpendix C

Example Building Study: Administration Building California state university at Northridge 1.

Introduction

1.1

Intent of Example Building Study This example building study is an application of the analytical procedures incorporated in Volume 1 of the document Seismic Evaluation and Retrofit of Existing Concrete Buildings to a real concrete building, the Administration Building of the California State University at Northridge. The primary intent of this study is to validate, or "test" the (draft versions) Methodology, and to provide , feedback to its developers, Secondarily, the study also demonstrates the use of the Methodology and provides an evaluation of its applicability. 1.2

I

scope of Example Building study This example building study report presents an illustration of the use of the Methodology for the seismic evaluation and retrofit of an existing concrete building. This report describes work performed in earlier phases of the development of the Methodology up to March-April 1996. Accordingly, some references to Methodology requirements, equations and scope may be out of date. Section 1.4 provides a limited update of principal results to the latest (final draft, August 1996) Methodology requirements. A seismic evaluation of the Administration BUilding (pre-Northridge Earthquake) is performed Using the Methodology. The methodology requires a nonlinear analysis of the building to determine the force-displacement characteristics of the lateral

"ppendlx C, Administration Building, CSUN

force resisting system. The seismic hazard of the building site, determined from site soil conditions and the proximity of the site to seismic sources determines the seismic demand. The force' displacement characteristics and the seismic demand are used to predict the performance of the building. The predicted building performance is compared with the desired performance objective to determine if seismic retrofit is required. , In this study, the structural behavior of the building predicted by the Methodology is compared to the observed earthquake damage. An elastic response spectrum analysis is performed and the results compared to that of the Methodology to determine whether it produces more useful results and greater insight into the behavior of the building. A conceptual retrofit strategy is developed using the Methodology to satisfy the selected Performance Objective. The deformation and movement of foundations can significantly affect the seismic response and performance of structures. The effects of foundations on the building response is investigated using a simplified two-dimensional model of a portion of the building, where the stiffness and capacity of the foundation and soil materials are considered. A pushover analysis of the "fixed" base and "flexible" base models are performed to determine the effects of the foundation on the building performance.

c·s


Limited nonlinear time history analyses are also performed to determine the reliability of the results produced by the Methodology. The "average" response of a set of time history . analyses is compared to the response predicted by the methodology.

1.3

summary of Findings

The subject building is highly susceptible to torsional response which complicates the application of the Methodology. In order to account for the torsional effects using the proposed methodology, a three-dimensional pushover analysis is required. At present there is no readily available (and reliable) computer code that has the capability of performing automated 3D nonlinear pushover (nonlinear static) analysis similar to the currently available 2D codes (such as Drain-2DX). Therefore, for this case study, a three-dimensional piecewise linear analysis of the building was performed. In a piecewise linear pushover analysis the demand and capacity of every critical element needs to be updated and checked at each step of the analysis. Members that have yielded need to be identified and "removed", and the model updated at each step. This process involves a tremendous bookkeeping effort. Based on the level of effort experienced in this study, the absence of analytical tools capable of performing three-dimensional nonlinear pushover analysis makes the direct application of the Methodology to three dimensional analysis of torsionally susceptible buildings impractical to implement at this time. Alternative approaches, using two dimensional nonlinear analyses, perhaps combined with three dimensional linear analyses, will be more practical. The proposed Methodology relies on the results of a pushover analysis to approximate the post-elastic capacity of the structure. The analytical tools most commonly used to perform this analysis (DRAIN-2DX) implicitly assumes that all components have elasto-plastic behavior i.e., they are assumed to perform in a ductile fashion with no strength and stiffness degradation.

e-G

Thi sim app Thf mo: pIa: she. mo. efff

Therefore, simplifications in the component behavior are required in developing the analytical models. The results of analyses based on such simplified modeling assumptions requires careful review and significant engineering judgment . The Methodology addresses the issue of global strength degradation (Section 8.2.1) and cautions that the modeling of this behavior requires considerable judgment, as strength and stiffness degradation depends on the magnitude of the response, and the number of loading and unloading cycles an element experiences. Strength and stiffness degradation can have a significant effect on the nonlinear seismic response and the level of damage that older concrete buildings experience. Limited nonlinear time history analyses were performed on a 2D model of a portion of the building. Comparison of results obtained from nonlinear time history analysis and the Methodology indicates that the methodology predicts higher roof displacements and lower base shears than the time history analyses do. This discrepancy may be attributed, in part, to the lateral load distribution used in the pushover analysis (Level 2: code distribution without the concentrated force at the roof level, described in Section 8.2.1 of the Methodology). The actual load distribution is probably significantly affected by higher mode effects, related to the torsional mode of response and the irregular mass distribution found in this building, which is accounted for in the time history analysis. The results of a pushover analysis are sensitive to the lateral load distribution used. Therefore, modifying the load distribution to include higher mode effects (see Levels 4 and 5 in Section 8.2.1 or the recommendations of Section 8.2.4 of the Methodology) may result in better correlation between the results from nonlinear time history analysis and the Methodology. The Methodology does predict the damage to the shear walls and coupling beams that was observed after the 1994 Northridge Earthquake. The extent of damage, however, appears to be underestimated by the proposed methodology.


ana insi ide] the yiel pos 1.11

earl the stU( brif Ihei dis] dell AUI add bui. COl

ApI the ana dra "fil

i

L

~~---------------------'1--SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

~I r

__------------------------------------------------------------

It

This discrepancy is the cumulative effect of the simplifying assumptions that had to be made to apply the proposed ~ethodology. to this bu.ilding. The specific assumptIOns we belIeve contrIbuted most to this "discrepancy" are the bilinear (elastoplastic) component behavior model that ignores shear strength degradation and the fundamental mode load distribution that ignores higher mode effects The Methodology, using nonlinear static analysis procedures, does provide important insight into the building's seismic performance by identifying failure mechanisms and accounting for the redistribution of forces during progressive yielding. This level of understanding is not possible using traditional linear, elastic analysis.

alytical mch :areful 1t.

If global •

utions fness

Ie 110ading: td ! effect :vel of i

!

!ence. were .e :om

1.4

Update

This case study was performed based on earlier versions of the Methodology. Revisions to ,r base: the Methodology have been made since the case study has been completed. This Section provides a his brief update of the most significant revisions and I their effect on the results of the study. r The performance point and the associated roof the displacement of the existing building has been :d in determined based on the revisions contained in the ual August 1996 Final Draft of the Methodology. In fected addition, the target displacement of the existing nal building was calculated using the Displacement Coefficient Method and Equal Displacement Approximation. The following table summarizes rhe the roof displacement results from the various ) the analyses; including results from both the "earlier draft" version (as reported in this study below) and ~her "final" draft version of the CSM. U.l the n >ry

I

~e

to

ke.

Ie

, CSU,.,


A summary of the results is tabulated below. Method' ,........

'..

Roof Displacement (In.)

CSM (Earlier draft)

1.05

CSM (Finall; Type B, K-0.67

0.9

CSM (Final>; Type C, K-0.33

1.3

Displacement coefficient Method

4.1

Equal Displacement Approximation

0.9

As can be seen in the table, the roof displacements computed using the final version of the CSM differs from that computed using the earlier draft version. These changes are due to the introduction of a correction factor, 1(, to specifically account for the type of structural behavior expected of the building's primary lateral force resisting system. For this building, structural behavior of type B or C is expected, requiring use of 1(=0.67 or 1(=0.33 respectively (compared to an implied value of 1(= 1.0 in the earlier draft version). This factor reduces the amount of effective damping which can be assumed, limiting the reduction of the seismic demand spectrum, and increasing the expected roof displacement. For the poorer, type C behavior, the predicted roof displacement shows the expected increase. However, for type B, "average," behavior a decrease in the computed displacement occurred. We believe that this discrepancy can be attributed to our use of a graphical approach in our earlier studies to determine the performance point (and roof displacement) compared to our use of the more rigorous "Procedure A" for the final version. The equal displacement assumption yields target displacements which correspond very well with those calculated using the final version of the CSM. It should also be noted that the actual differences in predicted displacements (between the various CSM procedures) is quite small, and does not result in revision of any significant conclusions regarding evaluation or retrofit of the building.

C·7


The displacement coefficient method predicts a roof displacement which is significantly higher than those calculated using the other methods. We believe this discrepancy may be due to the unique features of this building, i.e., significant torsional (higher mode) response, very short period, and very irregular mass distribution (40% of total building mass occurs at the second floor level), which are not adequately accounted for by these generalized, and simplified, nonlinear static procedures.

2.

Building Description

2.1

General

The subject building is located in Northridge, California and was constructed in 1963. The building has five floors above grade in the rectangular main tower segment and one floor above grade in the attached north and south wings. The east wing has one floor above grade and is a separate structure. Plate I shows the northwest building elevation. The combined first floor has an area of approximately 70,000 square feet and measures 400 feet by 260 feet. Figure 1 shows the floor plan of the first floor level. The upper floors measure 74 feet by 227 feet and provides an area of approximately 16,500 square feet per floor. The floor plan of the typical tower floor level is shown in Figure 2. There is a partial basement floor level below the north wing.

2.2

Structural System

The building's foundation system consists of drilled, cast in place, concrete piles, grade beams, and pile caps. All piles are specified as 3000 psi concrete. The piles are either 21 inches or 24 inches in diameter. The first floor slab is cast on grade, and is 6" thick in the main tower area and 5" thick for the one story wings. The construction drawings specify that the concrete used for construction of the floor slabs, beams, columns, girders, and walls were typically 2500 psi. However, 3000 psi concrete was

c·a

specified for the first floor columns and walls in the main tower area of the building. All concrete reinforcement in the building was specified as "intermediate grade" reinforcement that has a nominal yield strength of 40,000 psi. The typical floor framing consists of a 4" thick reinforced concrete one-way slab spanning between 7-112" wide by 15" deep concrete pan joists. The concrete joists are supported by . reinforced concrete beams varying in size from 10" to 24" wide and 24" to 80" deep. Typical columns are rectangular in cross-section (l8"x24") with 16-#11 at the 2nd floor and 4#9 at the 5"' floor. All columns above the first floor of the main building are laterally tied columns (#3 at 10" or #3 at 12 "). A few columns at the first floor have spiral reinforcement (112"$ 3"). The lateral resistance in both directions for the building. is provided by concrete shear walls. A few brick walls at the first floor also act as shear walls. The concrete floor acts as a rigid diaphragm to collect and transfer the lateral forces to the walls. All shear walls in the buildings are lightly reinforced (#4 at 12" centers). The shear walls range in thickness from 8" to 14". In the east.west direction of the building the shear walls are typically 8" thick concrete, while those in the north-south direction are typically 10" thick concrete. A number of discontinuous shear walls are located in the tower. The most prominent of these are the east-west running corridor walls in the tower and the west facing wall at the main entrance of the building. The discontinuous walls are supported by concrete columns at the first floor. The columns supporting the corridor walls have spiral reinforcement while the other columns are tied.

I. 3.1 the

EaJ spe bee fun

In I stn tOll

con 10\\

eral

obs loc: seci

obs thn bui bui dee and app reir

The at d cral aw~

PIal

con suff buil bea huil

con floc the

eXte

Appendix C, Administration Bulldlnll, CSUN

App

---r ---' f

Is in Icrete as

a

------S-E-IS-M-IC-E-V-A-L-U-A-T-I-O-N-A-N-D-R-E-T-R-O--FIT-O-F--C-O-N-C-R-E-T-E-a-U-IL-D--IN-G-S---

-------------------------~~---

J.

Observed Earthquake Damage

1994 Northridge Earthquake The building site is located within two miles of the epicenter of the January 17, 1994 Northridge Earthquake, moment magnitude (Mw) 6.7. The spectral acceleration at the site is estimated to have been anywhere between 1.3 to 1.7g in the fundamental building period range of the structure. In general, the building suffered substantial structural damage but was not at risk of partial or total collapse. The damage was primarily concentrated in the concrete shear walls of the tower and included moderate to severe diagonal cracking, crushing and spalling. Cracks were observed in most of the concrete shear walls located in the interior of the building from the second to fifth floor. The most severe damage was observed at the east end of the tower. Plates 2 through 4 show the damage at the east end of the building. The concrete shear walls at the east end of the building are linked over the exit door opening by deep (lO"x60"), lightly reinforced (2-#8 bottom and 4-#8 top) coupling beams. These beams do not appear to have ties but instead had vertical reinforcing which terminated at the bottom steel. These coupling beams suffered significant damage at the second through fourth floor levels. The cracking was so severe that the concrete spalled away in many areas exposing the reinforcing steel. Plate 4 shows a damaged coupling beam. The concrete walls adjacent to these beams also suffered some of the most severe cracking in the bUilding. No damage was observed in any other beam element. The exterior walls at the east end of the building displayed signs of lateral sliding at the construction joint above the beam at the second floor line. There was evidence of lateral sliding at the construction joint at several locations of the exterior stair tower wall. The reinforcing steel 3.1

I

~" thick.'

Jan

'om al ;nx24"} t !, :st

~/;:. '. Y

for the

;.A shear Jhragm' Ie ghtly ills st-west

walls lt of Is in lin walls 'st walls )Iurnns.

,

IIl1l1endlx C, Administration Building, CSUN

dowels across these joints were exposed in some locations. Similar damage was noted in the stair tower between Grid 10 and 11. No damage was noted to any of the columns in the building, including the columns supporting the discontinuous shear walls. The elevated floor slabs adjacent to the heavily damaged shear walls on the second and third floors contain moderate to severe cracking. The slab cracking appears to extend through the full depth of the slab. This damage can be attributed to the introduction of a 10" thick concrete wall along Grid line 11.5 below the third floor. The introduction of this wall increases the shear demand on the slab, as the slab has to transfer the shear load from the wall at grid line 14 to the wall at grid line 11.5.

4.

preliminary Evaluation

4.1

General The preliminary evaluation of the building was based on a comparison of the demand placed on the structure by earthquake ground motion and the ultimate capacity of the structural system. The comparison of strength versus demand was made using the concept of ductility demand. Generally, most structural elements have sufficient ductility to allow demands greater than their calculated capacity. The measure of ductility demand is known as the Inelastic Demand Ratio (lOR) or as sometimes referred to as the Demand Capacity ratios (OCR). Chapter 8 of the methodology describes this approach. The lOR allows the direct examination of the amount of ductility needed to meet force demands for various structural elements, and provides a direct measure of probable building performance. A three-dimensional linear elastic computer model of the fixed base building was developed using the computer program ETABS. This model was developed to study the overall distribution of

c·g


4.2

Table 2. story Mass Distribution

the lateral force and drift response of the building. The structural model included all elements believed to contribute to the lateral resistance of the building. This consists of all shear walls in the building, columns under discontinuous walls and coupling beams. Figure 3 shows the threedimensional computer model of the building. The seismic base of the building was assumed to be the ground/first floor level. Tables 1 and 2 show member stiffness properties and mass distribution used in the analysis, respectively. Note that the wall stiffness properties used here differs somewhat from the default stiffness values included in Chapter 9 of the methodology. The values selected here are believed to be more representative for this building. The transverse walls have been assigned lower stiffness values because they are expected to have significantly higher stress levels than the walls in the longitudinal direction.

C·1D

Dynamic Characteristics of Building

cou exc,

dan late enti obs, pro' def; eart tow foUl anal wha

The building periods of vibration were determined from an eigenvalue analysis using ETABS and are presented in Table 3. The fundamental period in the transverse direction (north-south) is 0.36 sec. and in the longitudinal direction (east-west) 0.19 sec. Figures 4 through 6 show the fundamental mode shapes of the building. It is evident from the eigenvalue analysis that the building exhibits highly torsional behavior. This behavior can be attributed to the high concentration of shear walls in the north-south direction near the west end of the building. This disproportionate distribution of stiffness places a higher displacement demand on the walls at the east end of the building. In the post-elastic range, the torsional behavior becomes more pronounced once the coupled shear walls yield. Elastic analysis does not provide any insight into the probable behavior of individual elements, and thus the building as a whole, in the post-elastic range.

valr disc suff onl) !)let

-•

Tal

V0......0.:..:.. Stt

4.3

-• --•

Elastic Analysis

A linear elastic response spectrum analysis of the building was performed using the 5 percent damped default site spectrum (presented in Section 5.4.2 of this report). The maximum roof displacement measured at the center of mass and at grid line 14 was 2.4 inches and 3.6 inches, respectively. This result reflects the torsional behavior of the building. !DRs' (see chapter 8 of the methodology for definition) were calculated for critical elements of the lateral system. Table 4 gives the !DRs' for selective elements of the building. The !DRs' of the shear walls and

--

•

i

--

,

I Appendix C, Administration Building, CSUN

i

1

App,

~~---------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

---: ~-----------------------------------------------------

g

on linal )Ugh 6. lalysis ' lavior .. th :his es a :he mge, nced lalysis : e

coupling beams on the east end of the building exceed acceptable levels, suggesting severe damage and probable complete breakdown of their lateral resisting capability. This result is not entirely consistent with the actual damage observed during the Northridge Earthquake which produced ground motions comparable to that of the default spectrum. Although, the most severe earthquake damage occurred at the east end of the tower (line 14) where the structural elements were found to be most highly overstressed in the elastic analysis, the damage was significantly less than what one would expect from the high lORs' . No column had an lOR which exceeded a value of 1.0. The columns supporting the discontinuous shear walls were found to possess sufficient strength to meet the force demand, and only minor damage is anticipated to occur in these members.

Table 4. IDRS DF selected structural Elements

)k1Mi~j t

•

_:

--,

--

• is of nt

:,

oof 1 mdat I,

•

lof :d for:

•

,

Coupling Beam @ Line 14 Roof 5th 4th 3rd 2nd Wall L·P t2nd·3rd) @ Line 14 Shear [email protected]. Flexure Wall Q·K t2nd·3rd) @ Line 14 Shear [email protected]. Flexure COlumns Supporting Discontinuous Shear Walls

4.2 5.2 5.1 3.9 2.0 2.1 3.7 3.7 1.9 2.6 S.1 < 1.0

5.

Evaluation Of Existing Building By Product 1.2 Methodology

5.1

General

The seismic performance of the building was evaluated using the Capacity Spectrum Method (CSM) presented in Chapter 8 of the Methodology. This method of evaluation considers two aspects in the performance of a structure, the demand placed on the structure during a seismic event, and the strength/capacity of the structure. The performance of the building is measured by its ability to withstand the force demand imposed on it during a seismic event. This is accomplished qualitatively by comparing the anticipated performance of the building to a predetermined performance objective. The determination of the strength/capacity of a building requires a pushover analysis to be performed on the lateral force resisting system of the building. The pushover analysis determines the levels of building lateral forces and corresponding roof displacements that are associated with successive stages of the development of yielding in the major building members. The determination of the seismic demand on a structure requires a quantification of the seismic hazard at a site due to ground shaking for various earthquake hazard levels. The seismic hazard at a site considers the local geology and soil characteristics, and the seismicity chanicteristics of the site.

5.2

structure Modeling

5.2.1 computer Models A total of three different "base" models were developed for this study, a three·dimensional elastic model, a two·dimensional elastic model, and a two-dimensional inelastic model.

lents and

CSUN


c·n


Table s. 2·D wall Models for studY Of Foundation

'Hem

Table 6. Assumed Properties Of Clmstruction Materials ~~

All other strength calculations

40 ksl

To determine the strength/capacity of the building a pushover analysis was required. The preliminary evaluation of the building indicated that the building is highly susceptible to torsion. In order to account for the torsional effects in the building response, a three-dimensional piecewise linear pushover analysis was performed. The ETABS model of the building, which was used in the preliminary evaluation of the building, was used in the pushover analysis. A simplified two-dimensional model of a portion of the Administration Building was also developed to investigate the effects of foundation on the seismic response. The wall and coupling beams along grid line 14 were modeled for this purpose using the SAP90 computer program. The properties used in this model were analogous to those used in the 3-D ETABS model. The structural properties of the foundation and soil

were modeled with vertical spring elements. To account for the uncertainty of the soil properties, upper and lower bound values were used for the siiffness and capacity of the foundation. The stiffness and capacity of the foundation system was determined by a geotechnical engineer. A total of three models were used to study the foundation effects, a "fixed" base model and two "flexible" base models. Table 5 list the models used. A DRAIN-2DX model of the wall and coupling beams along grid line 14 was developed to perform nonlinear time history analyses. This model is analogous to the "fixed" base SAP90 model used in the evaluation of foundation effects. The shear walls were modeled using beam/column elements and joint elements in series. The beam/column elements were used to represent the flexural behavior of the walls, and were infinitely rigid in shear. The joint elements were used to represent the shear behavior of the walls, and were infinitely rigid in flexure. The DRAIN-2DX model was "validated" by performing two simple analyses. The capacities of all of the elements were arbitrarily increased to "linearize" the system, then an eigen-analysis was performed, and the results compared to that of the SAP90 model. A pushover analysis was also performed, and the results compared to that of the piecewise linear analysis. In both cases the results from the DRAIN-2DX model compared favorably with the equivalent SAP90 result.

5.2_2

"spe over gain strai In th "ex~

appr 25 p Ho.... rein1 and calc\ colU! valUl dicta joint thee earli and,

5.3 5.S. I

dete! the I, The dime thet capa push piecf

Modeling Assumptions

The values for the material properties, used in this study are given in Table 6. The methodology only gives default strength values for , reinforcement (see Section 9.5.2 of the methodology) which differs somewhat from those I assumed in this study. The values assumed here are I the expected strength of the material and are normally greater than the minimum "specified" ! values called out in the construction document. Specified values of material strength are used in design and reflect the minimum value. The "expected" values are always larger than the

state thet stol) proc, top. (Seci Alth, SUch stren the ~ it im

I

C·12

Appendix C, Administration Building, CSUIII

,j

lIpPI

~r

----S-E-.S-M-.-C-E-V-A-L-U-A-T-.-O-N-A-N-D-R-E-T-R-O-F-'T-O-F-C-O-N-C-R-E-T-E--B-U-'-LD-'N-G-S---

-----II __----------------------------------------------------------------------------------

To ties, the ,

'm was: tal of ' Ion

ble" ing is

!

~O

ffects. ; olumn ; nt the nitely to d were

"specified" values because of inherent overstrengths in the original material, strength gained over time, and increase in strength due to strain rates that are expected during earthquakes. In the absence of in-situ test results, the "expected" strength of concrete was assumed to be approximately 20 percent higher and reinforcement 25 percent higher than the "specified" values. However, the "expected" values of the reinforcement were used to calculate only the axial and flexural strength of the members. For all other calculations such as shear strength of beams, columns, shear walls, etc., the specified minimum values of the reinforcement were used. The limit state of most of the walls were dictated by the shear capacity at the construction joint. The effect of gravity loads were included in the calculation of the friction capacity. As noted earlier, "expected" values were used for the steel and concrete strength to calculate all capacities.

Pushover Analysis 5.3 " by ties of . Procedure 5.501 I to A pushover analysis was performed to is was determine the force-displacement characteristics of of the the lateral force resisting system of the building. The pushover analysis was performed in threeof the dimensions to account for the torsional behavior of 'esults Jrably I the building. In the absence of analytical tools capable of performing three-dimensional nonlinear pushover analyses, the analysis was performed in a piecewise linear fashion. The analysis proceeded in sequential stages. As [sed in stated earlier, the lateral forces were applied, in )Iogy the transverse direction of the building, to each story in proportion to the 1991 UBC code procedure without the concentrated FT force at the those lere are top. This is described in the methodology (Section 8.4) as Level 2 pushover analysis. Although theoretically other levels of sophistication ed" such as direct inclusion of higher modes and :nt. strength and stiffness degradation can be used for din the pushover analysis, lack of analytical tools make it impractical to implement at this time.

g, CSU"


Furthermore, the methodology is not clear as to how to include higher mode effects. The results presented herein based on force distribution representing fundamental mode response and no post-yield strength and/or stiffness degradation should, therefore, be viewed with some judgment. Member forces were calculated for the required combinations of vertical and lateral load. The lateral force level was adjusted so that an element was stressed to within 10 percent of its member strength. Once an element reached its member strength, the element was considered to be incapable of taking additional lateral load . The base shear and roof displacement were recorded. In the next stage, the yielding element was removed from the model. Increments of lateral load were applied to the revised model until another element yielded, and the increment of lateral load and the corresponding increment of roof displacement was added to the previous totals to give the accumulated values of base shear and roof displacement. This sequence continued until either a failure mode or mechanism was obtained. The procedure followed is consistent with the Methodology guidelines ..Table 7 is a sample of the spreadsheet used to track the demand and capacity of individual structural elements at each stage of the pushover analysis.

Results 5.502 The results of the pushover analysis indicate that the walls and coupling beams along grid line 14 are the first elements to yield. The walls between grid lines G and K yield in flexure, the walls between grid lines Land P yield in shear at the construction joint, and the coupling beams yield in flexure. This behavior is consistent with the results of the elastic response spectrum analysis. As the structural elements of the building yields the center of rigidity of the tower floors shift westward. At the last stage of the pushover analysis, the center of rigidity of the 3rd floor shifted 92 feet (40 percent of the tower length) from its original (undamaged) position. The shift

C-1S

-

SEISMIC EVALU~TION AND RETROFIT OF CONCRETE BUILDINGS

Table 1. samPle oF Spl"eadsheet used In Pushover Analysis

• Tal

J~I

Wal wal wal Wal

11] wal wal

~

--bITS

wal wal

\I) wal Wal

I.

2. 3. 4.

5. 6.

-

Taj

,

i,q,?~

--..

--

... C-14

APpendix C, Administration Building, CSU'"

Ap~

-r -1-

-----S-E-.S-M-.C-E-V-A-L-U-A-T-.-O-N-A-N-D-R-E-T-R-O-F-.T-O-F-C-O-N-C-R-E-T-E-B-U-.L-D-.-N-G-S---

,

TallIe 7. (continued) Sample Of Spreadsheet used in pushover Analysis

1. 2. 3. 4. 5. 6.

~,CSUN

Structural Element - Structural element and limit state being tracked. Element Capacity· Computed capacity of element. Updated at each stage of analysis. Element Demand· Force demand on element (from ETABS). Demand/Capacity - Demand-capacity ratio. Shaded cell indicates elements which have yielded. Scaling Factor - Scaling factor used for scaling forces and displacements. Remaining Capacity - Element capacity at end of stage. used to update element capacity at subsequent stage.


C·1!


of the center of rigidity accentuates the torsional behavior of the building. Eight stages were necessary before a mechanism was formed in the structure. The results of the pushover analysis are presented in Table 8. A capacity curve was developed from the results of the pushover analysis and is shown in Figure 7. The capacity curve represents the forcedisplacement characteristics of the lateral force resisting system of the entire structure (Chapter 8). The capacity curve is a plot of the base shear vs. roof displacement at the various stages of the analysis. A capacity curve was also developed for the wall along grid line 14. It was assumed that the wall supports 30 percent of the total tower weight. The assumed distribution of weight is in direct proportion to the effective stiffness of the wall. The normalized base shear was computed based upon the story shear at the 2nd floor level and the displacement at the roof level of grid line G. Figure 8 shows the capacity curves computed at the center of mass and at grid line 14. The figure illustrates the torsional nature of the building response and suggests that the displacement demand on the wall governs the performance of the building.

5.4 5.4.1

Seismic Demand Seismic Hazard Level

For this study, the earthquake hazard is represented by a Maximum Earthquake (ME), as defined in Chapter 4 of the Methodology. This level of seismic hazard is defined deterministically as the maximum level of earthquake ground shaking wh'ich may be expected at the building site within the known geological framework. The ME represents an upper bound level of ground shaking, which for this site may be taken as the level of earthquake ground motion that has a 10 percent probability of being exceeded in a 100 year time period, which corresponds to a return period of approximately 1000 years.

c-nl

Table 9. FaCtors IIsed tD Determine SeIsmIc Hazard

Ta ~

I

~

Wi Wi

we W, N - 1.2

Type B. <

Skm C. - 1.0·ZEN = 0.6 Cv - 1.6·ZEN - 0.96

Be;

Be, Bel Bel

Be;

NOI

5.4.2

Demand Spectrum

A default 5 percent damped site response spectrum was developed using the procedures of Section 4.4.3.3 of the Methodology. The building site has a stiff soil profile corresponding to a SD Soil Profile Type and is located in seismic zone 4, seismic zone factor (Z) of 0.4, from Table 4-4 of the Methodology. The seismic sources used for design are Type B and are located within 5 km of the site for a near-source factor (N) of 1.2, froni Table 4-5 of the Methodology. An E-factor of 1.25, corresponding to ME, was used to determine the shaking intensity, ZEN. used in Tables 4·7 and 4-8 of the Methodology. The default value of the effective peak ground acceleration (C.). from Table 4-7 of the Methodology, is 0.6g. The value of the seismic coefficient (Cv), given in Table 4-8 of the Methodology, is 0.96. The coefficients and factors used in the development of the site response spectrum are presented in Table 9. Figure 9 is a plot of the default 5 percent damped site response spectrum used in this study. The default site response spectrum is transformed into the default demand spectrum using spectral relationships. The demand spectrum used in this study was constructed from the default 5 percent damped site response spectrum using spectrum reduction factors (Chapter 4). These

APpendIx C, Administration Building, CSUN

fac bui Pu! stit bui res] bui

5.5 leVI rOll

at e rOll

perl of t COl

0.9 flex 2.5 limi

C.P rota alor Thil anal

r

---r -----


Table 10, Chord RotatIon of structural Elements Along Grid tine 14 /,Imlt

Parks' .

Stage 1

--

'Soil

-. _. l e4

:. EO

S, <

:m

-:

_i !

: of

ding So ne 4, 4 of Jr n of om

f :mine 7 and the

alue ; 4-8

and

ped

trum ;fault g

CSUN

Note: Additional walls are introduced at the lower level of the building. This resulted in the walls between Grd. and 2nd floor to remain essentially elastic.

factors account for the "effective" damping of the building above 5 percent of critical. The modal pushover curve is used to calculate the effective stiffness and the total energy dissipated by the building. These characteristics of the building response define the effective damping of the building,

5.5

RespOnse Limits

The response limits of the various performance levels were determined by considering the chord rotations of the concrete walls and coupling beams at each stage of the pushover analysis. These rotations were compared to the maximum permitted chord rotations presented in Chapter 11 of the Methodology. These limits are as follows: Coupling Beams: 1.0. = 0.5 percent, L.S. = 0.9 percent, C.P. = 1.4 percent; Walls with flexural limit states: 1.0. = 0.5 percent, L.S. = 2.5 percent, C.P. = 3.0 percent; Walls with shear limit states: 1.0. =0.5 percent, L.S. = 1.0 percent, C.P. = 1.5 percent. As can be seen from Figure 8, the chord rotations of the walls at the east end of the tower along grid line 14 govern the response limit states. This is consistent with the results of the elastic analysis and the observed earthquake damage, as


these walls experience high displacement demands. Table 10 contains the chord rotations of the structural elements along grid line 14 at every stage of the analysis. Table 11 summarizes the response limits as determined from the pushover analysis. Table 11. Response limits For BuildIng performance Levels

Immediate Occupancy

0.98

Life Safety

1.92

structural Stability

2.6

5.6

performance Objectives

Structural performance Level 5.6.1 A SP-l Structural Performance Level Immediate Occupancy (Section 3.2.1) was assumed for the ME. The immediate occupancy performance level assumes that the structure experiences very limited structural damage. The basic vertical and lateral force resisting systems of the building retain nearly all of their preearthquake characteristics and capacities. The risk

c-"


of structural failure is negligible and the building should be safe for unlimited entry.

5.6.2

Nonstructural PerFormance Level

An Immediate Occupancy (NP-B) Nonstructural Performance Level (Section 3.2.2) was assumed of the ME. At this performance level, nonstructural elements and systems are generally in place, minor disruption and clean-up should be expected.

5.6.8

Building Performance Level

Based upon the SP-l structural performance level and the NP-B nonstructural performance level that has been assumed, the overall performance level for this building is I-B, from Table 3-1 of the Methodology. Thus, the building is considered an essential facility (continuous occupancy) where the spaces and systems are presumed reasonably useable after a ME event, but continuity of lifeline service, either primary or backup, is not fully provided. This more restrictive performance objective was selected to demonstrate the applicability of the proposed methodology in selecting a retrofit scheme. A more representative performance objective for essential facilities would be an overall performance level of 1-B for a Design Earthquake and an overall performance level of 3C for a Maximum Earthquake. If this performance objective was used for this case study the methodology would have predicted that no retrofit was required.

5.7

Performance Evaluation

5.7.1

capacity Spectrum

A capacity spectrum was derived from the capacity curve. The capacity spectrum is a plot of spectral acceleration (S.) vs. spectral displacement (S.). It is constructed from the capacity curve of the pushover analysis by transforming the forcedisplacement points of the capacity curve to spectral acceleration and spectral displacement points using relationships developed from modal

C·18

wa me all the pre

analysis techniques (Chapter 8). The transformation was limited to the fundamental mode of vibration in this analysis. Table 8 provides the results of the transformation.

PerFormance Point 5.7.2 The performance point of the building was determined assuming a ME event for the site. The CSM diagram and the procedure outlined in the Methodology was used to establish the performance point. For the existing configuration of the building, the performance point corresponds to a spectral acceleration of 0.9g and spectral displacement of 1.3 inches. The performance point was transformed from values of spectral acceleration and spectral displacement to values of force-displacement. The transformed performance point has values of 7260 kips and 1.05 inches. 5.7.8

hal

are dis bili stn

5.1 5.1

car pel

the exi Th stn sti!

sulldlng Performance

The performance of the building is evaluated based upon where the performance point lies relative to the performance goal. Figure 10 shows the performance point along wi th the performance levels plotted on the capacity curve. As stated earlier, a Level 1 structural performance level, corresponding to immediate occupancy, was assumed for this study. As can be seen from the figure, the performance point falls outside of this performance range and as a result, the methodology requires the building to be strengthened in order to achieve the immediate occupancy performance goal. The CSM, as outlined in the Methodology, estimates the performance of the building to be very close to the immediate occupancy response limit. This implies that that the building would suffer "moderate" damage in a ME event. This is inconsistent with the level of damage sustained by the building during the Northridge Earthquake. The damage sustained during the Northridge Earthquake (which is believed to have generated less severe ground motion at the site than that assumed here in the analysis for the ME event)


rna 5.1

the pel

usi Me for the mc ace roc of poi bas pel

res 5.1

bui Var

I

AlII

--f-------------------------------------,;


- 'i I

was more severe than that predicted by the methodology. This discrepancy is probably due to all the assumptions that had to be made to apply the proposed methodology to this very complex problem. The assumptions that are believed to have contributed the most to this" discrepancy" are the assumed fundamental mode response, load distribution that ignores higher mode effects, and bilinear component behavior model which ignores strength and stiffness degradation.

11

as · The the ation 'ponds 1 from · The 7260

ated hows lance d

the · this

.le

.y, be nse ld lis is ld by

;e. lted .t It)

',CSUN

i

5.8

Evaluation of Foundation Effects

5.B.1

General

The deformation and movement of foundations can significantly affect the seismic response and performance of structures. Techniques to include the effects of foundations in the evaluation of existing buildings are presented in Chapter 10. These techniques require the expansion of the structural model to include representations of the stiffness and capacity of the foundation, and soil materials.

5.B.2

i it;

":.',

.

..

·2·DWlllt( ',<7,_-,,:;' :;>:_:: ,,, -." - -Moder;· "> ...•. "' Fixed I"Flexlble.1,~ Ftexll1le.2

Performance point

Sa s, ~Of Response Limit

Immediate Occupancy Life safety structural Stability

0.359 4.6 In. 4.7 In.

0.359 6.0 in. 6.2 In.

0.369 5.9 in. 6.1 in.

1.4 In.

1.9 in.

2.1 in.

2.2 In. 2.9 In.

2.7 in. 3.6 in.

2.9in. 3.8 in.

Table 1S. Chord Rotations at Roof Displacement of21nches

~~~

Analysis

A pushover analysis was performed for each of the models, and the performance points and performance level response limits determined using the procedures presented in the Methodology. Figure 11 shows the CSM diagram for each of the models. The performance point of the "fixed" base model has coordinates of 4.6 inches spectral displacement and 0.35g spectral acceleration. These spectral values correspond to a roof displacement of 4.69 inches and base shear of 1350 kips. Table 12 gives the performance point and response limits for the fixed and flexible base models. Figure 12 shows the plot of the performance point on the capacity curve with the response limits superimposed. 5.B.~

Table 12. performance points and Response Limits for 2·0 Wall Models

EValuation

To evaluate the effect of the foundation on the building performance, the chord rotations of various structural elements were computed for a

Appendix C, Administration BuUdlng, CSUN

roof displacement of 2". These values are given in Table 13. As can be seen from the table, for a given roof displacement the foundation has the effect of reducing the chord rotations of the coupling beams and walls. The results of this limited analysis suggest that the performance point of the fixed base model lies outside of the structural stability response limit. This result is consistent with the fact that only one wall of the building has been evaluated and does

C·,g


not reflect the effects of redistribution of loading between multiple bracing elements that typically exist in most buildings. The performance points of the flexible base models also lie outside of the structural stability response limit. Here, including the foundation in the analysis had the effect of shifting the performance point of the building. The roof displacements corresponding to the performance points of the flexible base models are significantly greater than that of the fixed base model. It may be mentioned that in the actual case where all the bracing elements of the building are included and load distribution is properly accounted for, the performance points calculated with and without the foundation effects are not expected to differ by so much. The increase in the roof displacement can be attributed to the "rigid body" rotation of the building due to the flexible foundation, the relative displacement of the structural members, or a combination of both.

5.9

"Limited" Nonlinear Time History Analysis

5.9.1 Ceneral Nonlinear time history analyses were performed, and the results used to evaluate the predicted performance of the building based on the Methodology. These analyses were performed using the computer program DRAIN-2DX. 5.9.2 Analysis A set of twenty ground motion time histories were used to excite the structure. These time histories were scaled to the 5 percent damped default site response spectrum used in the evaluation of the building performance. Figure 13 shows the scaled composite response spectrum for the set of time histories along with the default site response spectrum. Figure 14 shows the peak roof displacement response and maximum base shear for all of the time histories, and the average values of these parameters. The average maximum roof displacement is 2.35 inches and the average maximum base shear is 2188 kips.

C·2D

5.9.5 EValuation A comparison of the results from the "limited" time history analysis and the Methodology indicates that the methodology overestimates the roof displacement and underestimates the base shear (see Figures 14a and 14b). This discrepancy is primarily attributed to the lateral load distribution assumed in the pushover analysis. In the pushover analysis, a triangular load distribution is assumed. The actual load distribution is believed to be quite different and is significantly affected by higher mode effects. The inelastic analysis considers these effects, thus its results are thought to better approximate the actual response of the building.

S.

6.1

Evaluation Of strengthened Building By Product '.2 MethOdology Retrofit Scheme

The structural damage sustained during the Northridge Earthquake was concentrated at the east end of the building. This is consistent with the analytical results which indicate that torsion dominates the dynamic response of the building. The torsional response places an increased displacement demand on the walls at the east end of the building. To achieve the performance goal, the proposed retrofit scheme consist of strengthening of the wall along grid line 14 and the construction joints along grid line 11.5. The shear wall along grid line 14 will be strengthened by a full height six inch thick reinforced concrete wall constructed adjacent to the existing wall. The new wall has #6 at 18" reinforcing and will be connected to the existing concrete wall by means of #4 dowels at 30" o.c. horizontal and vertical. New boundary elements with 4-#9 were added to increase the flexural capacity of the wall. Additional diagonal reinforcing was also provided at the construction


NO

ttro RO

Ea! (10

join the "ba buil tors alor witl inse 6.2

mot

wall vibr anal fum (nol

dire sho, sIre) figu the

6.3

stre) Was mec coul elen disp alJe'

-


Table 14. Building period Of Vibration

...

1ited"

....... · c

...

....

Mode.·.• .

.

.......

perlridISeCJ $trengthen~

16xlstlng

the e 'ancy

North· South (transverse)

0.33

0.36

Rotation

0.22

0.23

. In

East· west (longitudinal)

0.19

0.19

nd is The ; its actual

ng

joints of the existing wall. The new wall increases the strength of the existing wall and also "balances" the stiffness distribution of the building, although the strengthened building is still I torsionally susceptible. The construction joints along the wall at grid line 11.5 are strengthened with #5 x 1'-5" dowels at 12" o.c. The dowels are inserted at an inclined angle into the existing wall.

6.2

he :he ·ith the .ing. tend goal, and the

The ETABS model of the building was modified to reflect the increased stiffness of the wall along grid line 14. The building periods of vibration were determined from an eigenvalue analysis and are presented in Table 14. The fundamental period in the transverse direction (north-south) is 0.33 sec. and in the longitudinal direction (east-west) 0.19 sec. Figures 15 and 16 show the first two fundamental mode shapes of the strengthened building. It is evident from the figures that the strengthened wall helps to reduce the torsional behavior of the building.

6.3

,. ,

1t to sting o.c. lents

al Iction

Dynamic Characteristics of strengthened Building

Evaluation of strengthened Building

A piecewise linear pushover analysis of the strengthened building was performed. The analysis was terminated prior to the formation of a failure mechanism. The results indicate that the walls and coupling beams along grid line 14 still are the first elements to yield. This is expected, as the higher displacement demand on this end of the building is alleviated, but not eliminated, by the proposed

Ig, CSUII Appendix C, Administration Building, CSUN

strengthening. Thus, this wall continues to experience a higher force demand. A capacity curve and capacity spectrum was developed for the strengthened building from the results of the pushover analysis. Figure 17 shows the capacity curve for the existing and strengthened buildings. As can be seen from the figure, the capacity of the strengthened building is greater than that of the existing building. The performance point of the strengthened building was determined to correspond to a spectral displacement of 0.73 inch and a spectral acceleration of 0.79g. The transformed performance point has a base shear of 7817 kips and a roof displacement of 0.86 inch. The response limits of the various performance levels were determined for the strengthened building. The response limit of the immediate occupancy performance level corresponds to a roof displacement of 1.18 inches. The performance point along with the response limits are plotted on the capacity curve, and shown in Figure 18. The performance point of the strengthened building lies within the immediate occupancy performance range and thus satisfies the performance objective.

7.

Concluding Remarks

This report presented an application of the document Seismic Evaluation and Retrofit of Existing Concrete Buildings to the Administration Building in the California State University at Northridge. The purpose of this example building study is to evaluate the applicability of this Methodology as well as to validate and demonstrate it's application. The following conclusions can be drawn from this study: 1. For highly torsionally susceptible buildings such as the one studied here, three-dimensional analysis is required. Accordingly, a direct application of the methodology would require a three-dimensional pushover analysis. At present there are no analytical tools available that has this capability. As demonstrated in this study, piecewise linear pushover analysis

C-21


can be done to approximate the behavior but that involves tremendous bookkeeping effort which essentially makes the application of the Methodology to such buildings impractical. 2. Most analytical tools e.g., DRAIN-2DX implicitly assumes that all components have elasto-plastic behavior i.e., they are assumed to perform in a ductile fashion with no strength and stiffness degradation. Requisite simplifications and modeling assumptions, therefore, are required to create building models for pushover analysis. Many of these assumptions can very often yield analytical results which are misleading and can provide false sense of security. 3. Comparison of results obtained using nonlinear time-history and Methodology indicates that the methodology overestimates the roof displacement and underestimates the base shear. This discrepancy is primarily attributed to the lateral load distribution (code distribution without the concentrated force at the roof level) assumed in the pushover

C-22

analysis. The actual load distribution is quite different as it is believed to be significantly affected by the higher modes effects. Therefore, realistic load distribution should be used for the analysis since the results of the pushover analysis can be very sensitive to the lateral load distribution. 4. The methodology appears to reasonably predict the earthquake damage to the shear waIls and coupling beams observed after the Northridge Earthquake. The extent of damage, however, appears to be significantly underestimated by the proposed methodology. This discrepancy is probably due to all the assumptions that had to be made to apply the proposed methodology to this very complex problem. The assumptions that are believed to have contributed the most to this "discrepancy" are the assumed fundamental mode response, load distribution that ignores higher mode effects, and bilinear (elastoplastic) component behavior model which ignores strength and stiffness degradation.


__ _________________________________ T~


Administration Building, CSUN

uite :ly lid be :he Jlhe

ar the mage, logy. Ie . the i lex I 'ed to

II

ltal ores

h 1.

I f

I

I

Plate 1. NDl'thwest BuildIng ElevatIon

,

I.

I

,f I

Plate 2. Flexural Damage at East Wall (GrId 14J

g, csuN


C-2J



Plate s. Slldlnll Damalle at EXterior wall (Crld 111'

Plate II. Damage CouplfnllBeam at East wall (Crld 111'

C-24

Appendix C, Administration Building, CSUIi

I\iIlle

-



Concrete Walls (Typ.)

•

Figure 1. Floor Plan DI FIrst Floor level.

g, CSUN


C·25.

~~~!'~';~r';/, '--;,_.- . .

'

-' .."

..

'

.

;'~~~~~l!~~1'> :!i:~~:MMi#f~~f";r:-'~:-' ,

>, -

•

"



1

o.

•

"",p

..

l

<=

~

:;..--'

"',

....

I

_-

-. 1- I ,.J!... ,

I

,

I

i

I

-0-

"

.......

b.

"'"

G

paa"....

. ;

II

G

I

- _ _ "" _ _ _

-

f~

-

\

~"",,,,

10=,

I

_ _ _..-..c=::===:o:t==m:::=.,=-..-==

=-----.--'''F''-

H •

i

!

~

Concrete Walls (Typ.)

~

......

iX' ,I I

-

~=

-

,

l

i,,

...lL-..Jl. _-1L .... JL-----'L-

iL_

~

l....·.··_ .. t

~It::r.

---'"

p

ITI

Il

.

I

r .J

-0

Figure 2. Floor Plan 01 TYpical Tower Floor Level. (Only Transverse Walls are ShOwn)

C-2&


---



Coaaete w.... Typ.

o

FIgure S. Three DImensIonal computer Model.

<2>

----'

,-

If-i-

--I·

- L'

FIgure 4. First Mode oF VibratIon (North-South TranslatIon) oF EXIstIng Building.

I, CSUN


C·2'



L __ ~_~ ___ JJ I r.-----!

~-

-.JI

-- - --.---I

----'i/11---

---:I

L1 - -

Fillure 5_ Second Mode DI Vibration (Rotation) DI EXlstlnll Bulldlnll_

----

--- --._. --~ I !llL

I.

I

I~

n

-----rio

1 _____

~J= ~I

I r~l,~---, III

~

li-

~ II

11" Fillure 6_ Third Mode 01 Vibration (East-west Translation) 01 EXlstlnll Bulldlnll_

C-28

APpendix C, Administration Building, CSUN

--


Administration Building, CSUN 0.7

VIW

0.6 0.5 . ...... .......

0.4

....-......' - - - - ' -_ _ ~bear Yle~g

. . . ... @

Lini4, 2nd & 3i-d Floors. G· P

0.3 Shear tlelding@CJ.

@

Lin~ 14, 2ndFI~r, L· P

.. ; ..

0.2 0.1

o~--~--~--~--~--~--~--------~--~~

o

0.5

1.5

I

2

2.5

3

3.5

4.5

4

5

Roof Displacement (in.) Figure 7. capacity Curve of EXisting Building at center of Mass.

0.7

VIW

0.6 0.5 0.4 0.3

0.2 0.1 o~--~--~--~----~--~--~--~----~--~--~

o

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

Roof Displacement (in.) Figure B. capacity curves of EXIsting Building at Center Of Mass and GrId LIne 14.

I,

CSUN. Appendix C, Administration Building, CSUN

C-2!J


Administration Building, CSUN Spectral Acceleration (g)

2

r----~.- -------------- ------ --------- ----------'- -~-~ ;~~- -----

1.5

I

-----------------------------------------~--~--~-----~----~----~-----~---

0.5

o

o

0.5

2

1.5

I

2.5

3

Period (sec.) Figure 9. Plot of 596 Dampetl Default site Response Spectra.

0.7

VIW ..... I.!) . (~.98) ....•..... I.,S. (j,lIl). .. 5.$. (2,6>...

0.6

................ ,..

0.5 0.4

............

0.3 1.0. ImmeIdate Occupancy

0.2

L.S. IJfe Safety

8.S. 8tructural 8tabWty

0.1 0~------~--~--~--~+-----------~--~--4

o

0.5

I

1.5

2

2.5

3

3.5

4

4.5

5

Roof Displacement (in.) Figure 10. Performance point anti Response Limits of EXisting Bulltllng Plottetl on capacity CUrve. (Roof Displacement at Center Of Mass)

c-:so

Appendix C, Administration Building, csuN

----


Administration Building, CSUN 0.7

Spectral Acceleration (g)

0.6 0.5 0.4 0.3 ----------------------------------'~---~----~----~----~----~----~---------

0.2

Reduud Demand Spectrum

0.1

o

o

2

4

6

12

10

8

14

Spectral D'isplacement (in)

Figure 11a. CSM Diagram of Fixed Based Two Dimensional wall Model.


0.7

----------------------------------------------~--------------------a.dac.
0.' 0.5

0.'

-------------------

0.3 __

0.2

_ ____________________

~

_"!, ~'1'! _51' !c_I,:"_"'_______________________________

_

0.1

o

o

•

2

•

8

10

Spectral Displaccm cnt (in)

FIgure 11b. CSM Diagram Of Flexlble·1 TwO DimensIonal wall Model. 0.7


0.' 0.5

-----------------------------------

0.' 0.3 0.2 0.1 0 0

2

•

•

8

10

Spectral D isplacem eDt (in)

70f

Figure 11c. elM Diagram Of Flexlble-2 TwO Dimensional wall Model.

I, CSUN


C-!1


Administration Building, CSUN N~onn~W=-ud~B=ase~=Sh~~~~~~~

035

i-

0.3

____________________________,

........... .

0_25

0.15

o

1.0. Immediate Occupancy . . ....... _...... --_. -,. -L.S.ur.,SBf'elf·

0_1

" 8.8. 8truc:tural Stability

~'

0.05

'I' P.P. Perfol'Dl8l\Ce Point . .. c ............. _.. _.. _.. ___ . . . . .

..... .

o~~~~~~~~~~~~~~~~~~~~-J

o

2

3

4

5

6

7

Roof Displacement (in.)

Figure 12. performance point and Response Limits of Fixed Base, Flexlble-1, and Flexlble-2 TWO Dimensional wall Model_

2

1.5


1rI--\-:"....,,-""'-""\-- - - - - - - - - - - -Scaled Composite Spectrum -5% Default Spectrum

0.5

Spectral Displacement (in)

Figure 1:5. scaled Composite Response Spectra.

C-S2

APpendix C, Administration Building, CSUN

-

ss-:)

Nn~

Nn~ 'IlUIPllna UOI:j8.l:PIUIWPY ':I XJPueddy

'III

Max. Base Shear (kips) 0

I

8

§

8-

§

§

§

§

Capitola - 38 Deg.

Capitola - 128 Deg. Corralitos - 38 Deg. Corralitos - 128 Deg.

;U

1

EI Centro, Station 6 -143 Deg.

ill


S


16


'" fZ-=

~ ~

t S'

•i is

i

~ :to

l ';i

ill',

I

I!

I ;;:

a'l

a ~

~

n <:>

a

C/l

III

'15'

~

CIl

.e

~ ~ a ~

Newhall - 80 Deg.

1:1 1:1

s::<:>... -<:> 1:1

~

.:Ii'"

6-

Newhall- 350 Deg.

. ~

51

e

Petrolia - 350 Deg. Rinaldi - 32 Deg.

~ ~

I~

iii

Rinaldi - 122 Deg.

/eUD/5

Ii ~ ~ ....

a Z '"'" N .8

Petrolia - 80 Deg.

I

'""

I

e

= 1:1 fIJ

~

g:

III

'"

.8

~

m :t

Sylmar Hospital - 32 Deg. Sylmar Hospital- 122 Deg.

-

S=»NlallDa iI.LilB:aNO:a ::10 .LhIOB.LilB aN" NOI.L"nl"AIl :aIMlSIIIS

-

IQIIII1I

,.ns:l 'llulPllna uOI~e.qslulwPV 'J xlPuelldv

0

.1:.:;'

-

Max Roof Displacment (in.)

'"

'"

...

'"

'"

....

co

J:;,;"~>'f~:S"':+~**,'4!1fl~;!t:,*.~,';~"l'i7i;~Wi~.,-t~~_"''!'fu..~r,~-:ti

Capitola - 38 Deg. Capitola - 128 Deg. Corralitos - 38 Deg. Corralitos - 128 Deg.

i

5


~ If

:2


~


::i


S

! 21 't

a'.l

f

1:1

J :..

2

i... •

!

I l

I

a c: Q,

Newhall - 80 Deg.

=:: 10

Newhall- 350 Deg.

Iil

Petrolia - 80 Deg.

::.

~ S

a.

I s:: ~

a

sa. 0 g" til

I

~

1'>

..~

.~

.!:!!

I

~

N

."

Rinaldi - 32 Deg.

-

= i;!. !it ~ ~ Q 51 ~ '"

Petrolia - 350 Deg.

~

or f:

t=

>Iil

~ f. = ::I

Rinaldi - 122 Deg.

"til

SCS -32 Deg.

~ ;!. -:;;; .::l

SCS - 122 Deg. SCSE - 32 Deg. SCSE -122 Deg. Sylmar Hospital- 32 Deg. Sylmar Hospital- 122 Deg.

SDNla'lna II.LIIII:INO:l 110 .LIIIOII.LIIII aNY NOI.Lyn'YAIl :I1 ..SI1I5

'"

r

••,••

,e EVALUAT'. . AND .......F'T DF eDNe . . . . BUILD'• •'

I, Administration Building, CSUN

L __ -==-_-=-__ . JJ I ~I ~-.--...,--i

---' i -

---'1Ii-I-

, - 1_ _

Figure 15. First Mode of Vibration (North-South Translation) OF strengthened Building.

L_--___-_-__ JJ ~I~ ___ ~I

1

I

---'III-I-

----:1

< - I- -

Figure 16. Second Mode oF Vibration (Rotation) of strengthened Building

I CSUN ,


c·ss


---------------------------------------------------------------Administration Building, CSUN 0.7

API

EJ

VIW Sh.... Yieldlnc:@CJ.· @:iJDe7,2nd&3rdFloOr,L·P

.:.:.:;.:.:..-:

........... ~ ...

0.6

~Ir;'~; ... ~,~""..."".. :11........ifj,:~~.~~-~sfr~@

0.5

G.K

Ha

...........~ ..

I.iDe

Va

45, 3n1 I1oOr , L:P ......... .

.

0.4

preJ

0.3

. . . . . . . . ;. . . .

EI

0.2

.... ; ....

1-<

21 .Stren thened Buildin 1-" .....".. Yielding@ l.iDe 14, 2nd &

3rd Fioor,L·r

0.1 o~--~--~----~--~--~--~--------~--~----+

o

0.5

I

1.5

2

2.5

3

3.5

4

4.5

5

Roof Displacement (in.) Figure 17. capacity curve for EXisting and strengthened Building (ROOf Displacement Is at the Center Of Mass)

0.7

VIW j 1.0. (1.18)

0.6

............................

~-----

0.5

-_ .......

,...-.~ Sbear Yield..g @ C.J.· Ol.iDe 7, 2nd & 3rd FJi.or, L· P

0.4

0.3 0.2 aI YIelding

@ ~e

14, 2nd ~ 3rd Fk»or, L· P

0.1 o~--~--~~--~--~--~----~--~--~----~--~

o

0.5

1

1.5

2

2.5

3.5

3

4

4.5

5

Roof Displacement (in.) Figure 1B. performance Point and Response Limits Of strengthened Building Plotted on capacity CUrve.

cos.


I\JIpen

-


APpendix D

Example Building Study Holiday Inn Van Nuys, California prepared by Englekirk and Sabol Consulting Engineers, Inc. 2116 Arlington Avenue Los Angeles, California 90018

fMass)

Jlrve.

D·'


Table of Contents 1. Introduction ................................................................................................. D-5 1.1 Intent of Example Building Study ....................................................... D-5 1.2 Scope of Example Building Study ....................................................... D-5 1. 3 Introduction to the Methodology ........................................................ D-5 1.4 Summary of Findings ..................................................................... D-6 2. Building and Site Description ............................................................................ D-9 2.1 General ...................................................................................... D-9 2.2 Structural System and Members ......................................................... D-9 2.3 Earthquake Damage ...................................................................... D-11 3. Preliminary Analysis ..................................................................................... D-12 3.1 Evaluation Statements for Basic Building System ................................... D-12 4. Detailed Analysis Using the Product 1.2 Methodology ............................................ D-13 4.1 Introduction ................................................................................ D-13 4.2 Elastic Analysis to Establish First Mode Response ................................. D-13 4.3 Static Nonlinear (Pushover) Analysis ................................................. D-14 4.4 Static Nonlinear Analysis Results ...................................................... D-17 4.5 Time History Comparison ............................................................... D-18 5. Rehabilitation Scheme ................................................................................... D-19 5.1 Introduction ................................................................................ D-19 5.2 Exterior Frames ........................................................................... D-19 6. Concluding Remarks ..................................................................................... D-20

Appendix D. Holiday Inn

D·S

-


APpendix D

Example Building Study Holiday Inn van NUYS, California ,.

Introduction

1.1

Intent of Example Building study

This report presents the an application of Products 1.2 and 1.3 of the Seismic Retrofit Practices Improvement Program, titled Seismic Evaluation and Retrofit of Existing Concrete Buildings to the Holiday Inn in Van Nuys, California. The purpose of this example building study is to illustrate the use of the Methodology document as an example for other engineers to follow.

1.2

scope of Example Building study

This study presents the evaluation and concept retrofit design of an actual concrete building in Los Angeles based on the recommendations from the Methodology. Topics include covered in the case study include: • Preliminary evaluation (Section 3) •

Modeling, analysis, and assessment by nonlinear static procedure (Section 4)

•

Conceptual Retrofit (Section 5) This study was undertaken by Englekirk & Sabol Consulting Engineers, Inc. based on the Second Draft of the Methodology (December 8, 1995) with updated references to the Third Draft (May, 1996).

1.3

Introduction to the Methodology

Chapter 2 of the Methodology outlines the recommend steps to undertake the evaluation and, if warranted, the seismic evaluation of the existing building. This section introduces the steps. Some


of these steps are procedural, not technical, and are not discussed in this case study. Therefore, case study will concentrate on Steps 3 through 5 and 7 through 10 of the Methodology. Step 1:

Initiate the Process: This step is not addressed in this example because it addresses owner actions, jurisdictional requirements, and the like.

Step 2:

Select Qualified Professionals: This step, presumably, has been taken already.

Step 3:

Establish Performance Objectives: From Section 3.4.1, the Basic Safety Objective is taken to be a Building Performance Level of "Life Safety" for the Design Earthquake, and the Building Performance Level of "Structural Stability" for the Maximum Earthquake. The implications of this choice are addressed in Section 4 and 5.

The "Life Safety" Building Performance Level is intended to achieve a damage state that presents an extremely low probability of threat to life safety. The Design Earthquake is taken to be a ground motion with a 10 percent chance of being exceeded in 50 years. The "Structural Stability" Building Performance Level is intended to achieve a damage state involving the main building frame or vertical load carrying system and requires only stability under vertical loads, and no margin of collapse may be available.

D·S


Step 4:

Review Building Conditions: A review of the existing building conditions is presented in Section 2 (Building Description) and Section 3 (Preliminary Evaluation) .

Step 5:

Formulate a Strategy: As will be seen in the case study, the preliminary evaluation explained in Section 3 suggests that a detailed analysis is required. The analytical procedure selected for this case study is the Capacity Spectrum Method (CSM). The application of the CSM is presented in Section 4 of the case study.

Step 6:

Begin the Approval Process: This step is not directly relevant to this case study and is not discussed further.

Step 7:

Conduct Detailed Investigation: A site analysis was not undertaken for the case study, but Section 2 of the case study outlines the assumed material properties and relevant construction details.

Step 8:

Characterize Seismic Capacity: The modeling rules discussed in Chapters 9 and 11 of the Methodology are applied the case study building in Section 4.

Step 9:

Determine the Seismic Demand: The case study uses the procedure outlined in Chapter 4 of the Methodology, and its application is presented in Section 4. ** of the case study.

Step 10: Verify Peiformance: Based on the Performance Objective selected in Step 3, the CSM was applied to the structure and, using the seismic capacities established in Step 8, the performance is evaluated in Section 4. It will be shown that the structure does not satisfy the required Performance Objective, and conceptual seismic rehabilitation schemes were developed. Only one

D·.

scheme, an exterior concrete frame, is presented in the case study. Step 11: Prepare Construction Documents: This is not within the scope of the case study. Step 12: Monitor Construction Quality: This is not within the scope of the case study.

1.4

summary of Findings

1.4.1 Intl'Dductlon This case study applies the Methodology to a real building that sustained damage during the 1994 Northridge Earthquake. Since the performance of the building was known beforehand, the predictions of the Methodology can be compared to the observed performance of building. Just as clearly, however, one had to guard against altering the model to match the known results. Within this context, the case study served as a valuable tool in outlining the strengths and weaknesses of the methodology. We believe that it is unreasonable to expect this, or any other, new approach to evaluating existing concrete buildings, to be immediately useable by all. or even a majority of all, licensed engineers, architects, or building officials. The exception to this statement might be a set of extremely conservative, prescriptive provisions that would quickly prove unacceptably expensive because all buildings undergoing evaluation would require extensive rehabilitation. The flexibility of the Methodology recognizes that concrete buildings utilize a complex building material, consist of infinite combinations of physical layouts, framing system variations, and member proportions, and are subjected to different and unpredictable site ground motion. The depth of our knowledge in addressing these critical issues is very limited. Our application of the Methodology to the building in this case study makes us even more convinced that the goal of a "cookbook" method that can be applied by all registered engineers to produce nearly identical results in similar situations is not currently achievable. The

Appendix D, Holiday Inn

engine whatl accur; availa in con and hi result: requir profes neede, the av many

1.4.2 TI evalu1 rehabi buildiJ infom (Chap (Chap model model seismi impor appro< Metho infom minim suppOJ simpli reach recom Tl consid dealin: that th not su·

other,

trainin a soun Worth, Dedic: and us are no unders AIIpen

--- :, is This ;tudy.

to a e


engineering profession must better understand what parameters control concrete behavior, how to accurately model concrete structures using available software, what are the critical limit states in concrete and the different structural members, and how ground motion demand, analytical results, and observed damage are related. This required level of knowledge in the engineering profession does not currently exist because much needed information is not available and much of the available information is not well understood by many engineers.

General Findings 1.4.2 The Methodology outlines an approach to evaluate and, where needed, propose a seismic ,gy rehabilitation scheme, for an existing concrete e of building. The Metbodology provides useful ) information to develop earthquake demand (Chapter 4), to identify potential deficiencies ,tudy (Chapter 5), to develop reasonable analytical ngtbs models (Chapter 9), to reasonably interpret tbese models (Chapter 11), and to consider different ~ct seismic rehabilitation concepts (Chapter 6). The g importance of a clear, philosophically consistent y approach cannot be overemphasized. The nsed Metbodology brings the above-referenced he information into a single package, witb only a minimal number of references to needed illS supporting documentation (e.g. FEMA 178) to [sive simplify tbe engineer's task by placing witbin easy Nould reach a set of relatively consistent tyof ildings recommendations. The Metbodology requires tbe use of ,f considerable engineering judgment because we are ming dealing witb very complex problems. We believe and tbat tbe general level of engineering expertise is .ite not sufficiently advanced to apply tbis, or any in other, available metbodology witbout additional ;I. training. Nevertbeless, tbe Metbodology provides Ie a sound philosophical approach which will be ,re worthwhile to use as a basis for training. hod Dedicated individuals can take tbis information s to and use it to great advantage. However, tbose who are not willing to put in tbe effort to better understand tbe issues tbat are critical to seismic

lay Inn


rehabilitation of concrete buildings may not produce adequate retrofit designs even with this Methodology at the ready. The Methodology clearly identifies the need to establish a performance objective, and tbe importance of involving the Owner in its selection. The use of different performance objectives, and tbe resulting consequences, are presented in very broad terms. We believe this is appropriate since tbe engineering parameters in the Methodology used to evaluate and rehabilitate buildings have not yet been well correlated with actual earthquake demands. Therefore, caution must be exercised to avoid the perception that use of the Metbodology to execute a retrofit design, intended to promote "damage control" for instance, cannot constitute a "guarantee" that such a level of performance will, in fact, be achieved when tbe building is subjected to an eartbquake. It is our opinion that such correlation cannot be expected for many years, and certainly not before designs based on the Metbodology are subjected to very large earthquakes. Given tbe desire to ensure that nearly all buildings requiring detailed evaluation and/or eventual rehabilitation are properly identified, tbe preliminary evaluation in Chapter 5 is conservatively written (as is tbe referenced FEMA 178 document). This conservatism can require substantial effort at tbis preliminary juncture and still conclude tbat a detailed analysis is needed. The value of tbe preliminary analysis is tbat it provides a sound starting point for tbe total process by helping tbe engineer to focus on potential deficiencies tbat might be overlooked while interpreting tbe output generated by the detailed analysis . As tbe title suggests, the Metbodology focuses on concrete elements. The designer should be aware, however, tbat additional guidance, not found in tbe Metbodology, is needed to properly implement non-concrete retrofit elements or specialized techniques such as energy dissipation. In addition, tbe Metbodology does not provide detailed guidance needed to comply with building

D·'


code requirements. It is our opinion that this additional information cannot be comprehensively treated in a document such as the Methodology, but appropriate references could be identified.

1.4.$

Specific Findings

An index would greatly simplify use of the document. The Preliminary Analysis, described in Section 3 of this report, did a good job of identifying the elements that were later found to be critically deficient (e.g. shear in columns). Nevertheless, a number of the "false" statements identified conditions that did not appear to be relevant to the ultimate evaluation (e.g. most of the diSCUssion regarding beam bars and splices). We believe this is an artifact of the general preliminary evaluation methodology and do not recommend that it be changed; however, additional discussion regarding the general nature of a preliminary evaluation could be provided. Using information obtained from prior studies of the SUbject building, we were able to provide a reasonable match between the measured building period and the analytical building period using an elastic mode\. It is not clear if such a good fit would have beim possible without the detailed information about the existing building. We did not examine explicitly how different elastic models would have varied the vertical distribution of seismic forces used in the nonlinear pushover analysis; however, we do not believe that the difference would have been significant. Perhaps the greatest challenge in conducting the nonlinear pushover analysis is that shear, as identified in the preliminary analysis, is a critical limit state for the building, yet this limit state is very difficult to model using currently available software. Very careful examination of the computer output is required, and multiple manual alterations of the nonlinear model are required. This is a shortcoming of most available software and not of the methodology; however, the MethodOlogy could provide additional guidance that warns the less experienced engineer that shear

D-.

performance must be evaluated manually and the model adjusted to reflect shear critical elements. As discussed in Section 4.5, a limited number of inelastic time-history analyses were executed as an approximate check of the Methodology. The results indicate a wide scatter in predicted maximum roof displacements which is to be expected because of the differences in the time-histories frequency content relative the dynamic response of the building. Two trends were identified: 1. The time histories predict higher shears, at lower displacements, than suggested by the pushover curve. 2. The time histories predict higher maximum displacements, a lower shears, than suggested by the pushover curve. It is our opinion that these results are not unexpected for the following reasons: • Adjustments made in the nonlinear pushover model to better account for shear behavior are not readily made in a time-history analysis. These would tend to soften the structure, thereby decreasing the demand that one might expect for a given level 'of displacement. •

•

The use of assumed, low levels of damping is consistent with the use of pseudo-accelerations, displacements, and velocities inherent in a spectral analysis, but such an analysis will tend to underestimate the demand compared to time-history analyses that permit the use of multiple damping levels. This appears to be the case in the time-history analyses, and the reported underestimation of base shears is probably not completely accurate. In general, a pushover analysis assumes, a priori, a vertical distribution of forces consistent with a single mode of vibration, In the case study, this corresponds to the fundamental mode. It is possible that a time history analysis, which is not limited by this assumption, will predict higher base shears because the effects of higher modes are


in te p' w

•

0 til th th re til pI c. fo m

w

pi In of the proba' defici, differ, nonlir signif evalUl sake ( into tl: donol arbitn memb effect! additi, metho

2. 2.1 Tl concn of the Theb feet 01 drawh buildh

Appen

-

-

the Its. nber ed as he

at Ie

m

:sted

ver Ir are is. night ng is ltions,

a II tend )

of be the


included. While multi-mode pushover analyses to better reflect these higher shears are possible, the level of complexity hardly warrants their use. One cannot conclude that because the time-history analysis predicts higher demands than does the non-linear pushover analysis that the non-linear pushover analysis is not reliable. This conclusion presupposes that the time-history analysis is superior to the pushover analysis. This is not necessarily the case because the influence of shear behavior for example, which can be reflected in the ' nonlinear pushover analysis, is not handled well by most currently aViliiable nonlinear programs. In conclusion, it is our opinion that the results of the Methodology reasonably describe the probable behavior and identify essential deficiencies in me building, and that the differences between me time history results on me nonlinear pushover analysis are, ultimately, not significant. We do not believe that limiting me evaluation to me use of an elastic model for the sake of simplicity would provide as much insight into me anticipated dynamic behavior. Furtl1er, we do not believe mat it is appropriate to impose arbitrary limits on acceptable forces in critical members to account for possible higher mode effects. This is but one of many areas mat require additional research before we can produce a simple memod for evaluating concrete structures.

1.

2. 2.1

In

me this Irs

lay Inn

.Date of Cons,trOCtlori Date Of Drawings

Building and Site Description Ceneral

The Holiday Inn is a seven-story reinforced concrete structure located in Van Nuys, just east of me San Diego Freeway at Roscoe Boulevard. The building consists of roughly 63,000 square feet of floor area. The original construction drawings are dated February 19, 1965 and me building is believed to have been constructed


1965·66 1965·66

L.A. City BuHdingCode , 1964 (assumed) , ,

•

s

a

Table 1. Building Summary

7

Ground Floor

13',6"

second through sixth Floor

8'·8V,"

Seventh Floor

S'·8"

during 1965-66. Table 1 presents a summary of me building's parameters. The building is essentially rectangular in plan with overall dimensions of approximately 62'-8" by 150'-0" in me norm-south and east-west directions, respectively.

2.2

structural System and Members

Foundations Foundations supporting me Holiday Inn consist of 38-inch deep pile caps, supported by groups of two to four poured-in-place 24-inch diameter reinforced concrete friction piles. All pile caps are connected by a grid of tie beams and grade beams. Each pile is approximately 40 feet long and has a design capacity of over 100 kips vertical load and up to 20 kips lateral load. Gravity Load System All structural weight and superimposed load on me building is carried by a system of reinforced concrete flat slab and perimeter concrete beams supported by concrete columns. The concrete slab is 10 inches mick at me second floor, 8 '12 inches mick at me mird to sevenm floors, and 8 inches mick at me roof. The typical framing consists of columns spaced at approximately 20' -0" centers in me transverse (N-S) direction and 18'-9"

D·9


centers in the longitudinal direction. Figure 1 shows the typical floor framing plan.

Lateral Load system Lateral forces in each direction are resisted by perimeter spandrel beam-column frames as well as interior slab-column frames. Typical interior columns are 20"x20" between the Ground and Second Floors and 18"xI8" above the Second Floor. The column reinforcing varies along the height of the column with 6-#7 above the Fifth Floor, 6-#8 between the Fourth and Fifth Floors, 8-#9 between the Third and Fourth Floors, and 10-#9 below the Third Floor. The ties are #3 at 12" centers between the Ground and Fifth Floors and #2 at 12" centers above the Fifth Floors. Slab reinforcing in the column strip in both directions typically consists of 16-#6 at the top near the columns and 8-#6 at the bottom. Figure 2 shows the elevation of the North perimeter frame; the South frame is similar. The columns are 14"x20" and have their weak axis oriented in the plane of the frames as shown in Figure 3. They are reinforced wi th 10-#9 between the Ground and Second Floors, 6-#9 between the Second and Fourth Floors, and 6-#7 abov.e the Fourth Floor. The reinforcement is spliced immediately above the floor level and has a lap length of approximately 38 bar diameter as shown in Figure 3. The ties are #3 at 12 in. on center between the Ground and Fourth Floors and #2 at 12 in. on center above the Fourth Floor. At typical floor levels, the beams are 16"x22.5" and are reinforced with 2-#6 at the bottom and anywhere from 2-#8 to 3-#9 at the top. Figure 2 also shows the location on the north perimeter frame of four bays of brick infill wall between the Ground and Second Floors. Nominal I-inch and Ih-inch expansion joints separate these walls from the exterior columns and the underside of the Second Floor spandrels, respectively as shown in Figure 3. Although, these elements were not designed as part of the lateral force-resisting system, they appear to have participated in resisting the imposed demand as evident from

D·10

around - 2nd Floor

5,000 psi

6,250 PSi

Columns, 2nd - 3rd Floor

4,000 psi'

5,000 PSI

Beams and slab,2nd Floor only

4,000 psi

5,000 psi

stre] exhi stiff load that the] test buil, the I

Eart 3,000 psi

3,750 psi

secc estir stan,

15-2

"exI appr Beams and Slabs

Column Bars

Note:

intermediate· grade ASTM A-15 and A-305)

401<51

Deformed Billet bars ASTM A-432)

60 ksl

50 Ksl

75 ksi

"Expected" values used in this study are approximately 25 percent higher than the "specified"

values giyen above.

the damage sustained by these walls during both the Northridge (1994) and the San Fernando Earthquake (1971). In the transverse (N-S) direction, the perimeter columns have their strong axis in the direction of the frames. These frames are hidden behind 1" thick cement plaster supported by metal studs.

Materials Table 2 presents typical material properties obtained from available record drawings, The structure is constructed of regular weight reinforced concrete. The values in Table 2 are "specified" values which can significantly underestimate the actual strength (referred to hereafter as the "expected" strength) of the in-situ material. The "expected" values are nearly always larger than the "specified" values because of the inherent over strengths in the original material and


"spe the; on tl of d used strer calcl colu rein!

2.3 dam

Eart seve wen eart! leve: dam was infOJ engi: eard relia to th to th flext

---

SEISMIC EVALUATION AND RETROIilIT 0111 CONCRETE BUILDINGS

-

,_~,

o psi

o PSi

I()

PSi

;0 psi

I

Ksl

; ksl

-

ified"

Jth

meter Ion of nd 1"

es

e I-situ ways the Ii and

ay Inn

strength gained over time. Furthermore, concrete exhibits a significant increase in both strength and stiffness and reinforcing steel in strength when loaded at increased strain rates, e.g. at strain rates that are expected during earthquakes. Compared to the normal rate of loading for standard cylinder test which is 35 psi/sec, the concrete in this building was strained at a rate estimated to be on the order of 8,000 psi/sec during the Northridge Earthquake assuming a building period of 1.5 second. This higher rate of loading alone is estimated to increase the strength obtained from standard cylinder tests by as much as 15-20 percent. In absence of in-situ test results, the "expected" values of all materials can be approximately 33 percent greater than the "specified" values, and this fact was reflected in the analysis as shown in Table 2. However, based on the Methodology guidelines, "expected" values of the reinforcement given in the document were used to calculate only the axial and flexural strength of the members. For all other calculations such as shear strength of beams and columns, the specified minimum value of the reinforcement was used.

2.3

Earthquake Damage

The building experienced extensive structural damage during the January 17, 1994 Northridge Earthquake. The building was red tagged and several bays along the perimeter of the building were temporarily shored immediately after the earthquake. Shoring was provided up to fifth floor level to bays where the adjoining columns were damaged and the vertical load carrying capacity was believed to have been compromised. This information, while typically not available to engineers analyzing a building prior to a damaging earthquake, must be considered in evaluating the reliability of the analytical results. The structural damage was primarily confined to the longitudinal perimeter frames with damage to the transverse direction frames limited to minor flexural cracks in the end bay beams. Figure 4

Appendix D, HOliday Inn

iIIustrates the elevation of the longitudinal perimeter frames indicating the location of major damage. The damage was most severe between the Fourth and Fifth Floors of the south perimeter frame (Line A). Figures 5 and 6 show photographs of the damage sustained by these frames. Damage primarily consisted of shear failure of the columns immediately below the Fifth Floor spandrel beam. At several locations, extensive shear cracking may also have promoted the buckling of vertical column reinforcement as shown in Figure 6. In addition to damage to the columns, many beam-column joints below the fifth floor level also sustained minor to moderate shear cracks. The damage was observed in beam-column joints of both longitudinal frames although the south perimeter frame appeared to have slightly more damage. Along with shear cracks, concrete spalIing was also observed in one of the joints. In addition, all four perimeter frames experienced structural distress in the form of concrete spalIing and hairline flexural cracks observed in several spandrel beams. Figure 5 shows a close-up view of infilled bay of the north perimeter frame located towards the east end of the building. Cracks in the Second Floor beam-column joint are clearly evident. Cracks occurred at the same location during the 1971 San Fernando Earthquake. The photograph also shows typical crack pattern in the "nonstructural" brick infill walls along the north face of the building. These cracks occurred at second floor beam soffit and near the comers of each panel. The observed damage clearly suggests that these brick infill walls participated in resisting the imposed seismic demand. Nonstructural damage was not very extensive and was mostly confined to the Fourth floor. Doors, windows, and drywall partitions in the east-west direction suffered severe damage between the Fourth and the Fifth Floors. This is attributed to the large deformation of this story during the earthquake. The response of the building during the Northridge earthquake was recorded by a total of

D-11


sixteen sensors and analyzed to obtain information about the dynamic response [Islam, 1995]. Based on the results of these analyses, the following conclusions were drawn: 1. The brick filler walls between the Ground and Second Floors at the northeast end of the structure interacted with the confining frame introducing asymmetry in the longitudinal direction. This is believed to have resulted in higher displacement demand and therefore, damage along the south end of the building. 2. The second mode contributed significantly to the building response. Consequently, the shear demand between the fourth and fifth floor levels appears to have been close to if not the maximum experienced by the building. This, along wi th the fact that the building shear capacity of this story is less than that of the stories below, contributed to the damage being concentrated between the Fourth and Fifth Floor levels. 3. The columns between the Fourth and Fifth Floor levels failed in shear. The lack of adequate ties in the column exacerbated the damage and allowed vertical column reinforcing to buckle where concrete had spalled off. Although the subsequent analysis will utilize seismic loads generated by a first mode response, the results of the analysis will compare reasonably well with the observed damage, even though the analysis predicts significantly more hinging at the lower floor levels than was observed in the actual building damage. This deviation between predicted and observed response may also reflect the characteristics of the actual earthquake ground motion experienced at the site. Since it is unlikely that an analyst will have available to him or her such detailed response records prior to undertaking a seismic rehabilitation project, this analysis proceeded using the Methodology to the maximum extent possible.

D'12

I.

preliminary Analvsls

Before undertaking a detailed, time consuming analysis of a building, the Methodology (Chapter 5) recommends conducting a preliminary analysis. This case study uses the evaluation statements in FEMA 178. The evaluation statements express a variety of positions, which, if true, suggest that no detailed analysiS is required. When the statements are false, additional analysis is recommended to examine potential seismic deficiencies. For this building, many of the statements are "false, and only the "false" statements are discussed in this section.

3.1

EValuation statements for Basic Building system

ConFlgu,.atlon S.1.1 Weak Story. There is a significant strength discontinuities in the vertical elements in the lateral fore resisting system: column shear strength at fourth/fifth floor. Torsion. The lateral system may not be well balanced and may be subject to torsion because of infill panels at first floor. $.1.2 concrete Moment F,.ames Shearing Stress Check. The building does not satisfy the Quick Check of the average shearing stress in the columns. Shear Failures. The shear capacity of the frame members is not greater than the moment capacity. Stirrup and Tie Hooks. The beam stirrups and column ties are not anchored into the member cores with hooks of 135 degrees or more. Column Tie Spacing. Frame columns have ties spaced greater than d/4 or more throughout their length and at more than 8 db or more at all potential plastic hinge regions. Column Bar Splices. Column bar lap splice lengths are less than 35 db long and are not enclosed by ties spaced at 8 db or less. Beam Bars. At least two longitudinal top and bottom bars do not extend continuously throughout


the oft pos thre lonl witl are spal at 8 at tl join of~

dew sig~

anal

4.

4.1 coni the altel are

MaJ in 0

Met calli case

4.2 Ani first

dete Stati

-


Igth

the length of each frame beam. At least 25 percent of the steel provided at the joints for either positive or negative moment in not continuous through the members. Beam Bar Splices. The lap splices for the longitudinal beam reinforcing are not located within the center half of the member lengths and are in the vicinity of potential plastic hinges. Stirrup Spacing. Some beams have stirrups spaces at d/2 or more throughout their length and at 8 db or more at potential hinge locations. Joint Reinforcing. Column ties do not extend at their typical spacing through al beam-column joints at exterior columns. Based on the number and potential significance of the possible deficiencies ouilined above, a more detailed analysis is recommended to assess the significance of the these deficiencies. This detailed analysis is presented in Section 4 of the case study.

.trength

4.

Detailed Analysis Using the Product 1.2 Methodology

4.1

Introduction

5 uming apter alysis. ts in !SS

a

that no ments d to this and this

Basic

well Luse of

loes not ,ring ;he

lent rups lember have hout at all

This section outlines the basic steps required to conduct a detailed analysis of the structure using the Methodology. There are a number of alternative approaches that can be used, and these are outlined in Chapter 8 of the Methodology. Many of these alternatives are covered in detailed in other sources, but the Capacity Spectrum Method using a static nonlinear analysis, often called a "pushover analysis, " was selected for this case study.

4.2

Elastic Analysis to Establish First Mode Response

An elastic analysis is conducted to establish the first mode response which will be used to determine the pattern of force to be applied in the static nonlinear analysis.

splice

:op and oughout

IIday

Inn APpendix D. Holiday Inn

Table :So Description 01 the prelimInary ElastIc Models ".

':<:

<:;~

:EFFectJve

, . MDmentsDf panel

zane I,VI$et1us ' .Datflplng

: :', ,Ihe/tltif·/ '.

,RigIdity'

COlumns

0.61.

50%

Beams

0.81.

Model 1 5% 10%

Model 2 Columns (4th'5th)

0.051,

Columns (all others)

0.50 I,

Beams

0.50 I,

* Ig

0%

= Gross moment of inenia

Live Load Estimates 4.2.1 Estimates of typical service live loads were applied to the model as discussed in Section 9.2 of the Methodology, rather than the live loads assumed by the model building codes for design. Since a "Hotel" occupancy is not listed in Table 9-1, it was assumed that a "General Office Area" occupancy was a representative occupancy. 4.2_2 Soli-Structure Interaction The dispersed nature of the frame elements and the pile foundation suggested that issues related to soil-structure interaction as discussed in Section 9.3 would not be significant. Other case studies conducted illustrate the application of the Methodology and soil structure interaction.

4.2.$

MOdeling Assumptions

Table 3 presents member stiffness and other modeling assumptions used in the analysis. Two models were constructed to examine the sensitivity of the structure to different parameters. Modell assumes constant effective moments of inertia for the beams and columns. while Model 2 attempts to account for the reduction in shear strength anticipated in the columns between the Fourth and Fifth Floors.

D·1!


The building was modeled as a three-dimensional elastic structure with a fixed base to establish the first mode response of the structure for use in applying the lateral loads used in the pushover analysis. The structural model included both perimeter beam-column frames as well as interior slab-column frames, as reCOmmended in Section 9.4.2.2 of the Methodology. In the longitudinal direction, there are two slab-column frames while in the transverse direction, there are seven such frames. All are elements believed to contribute to the lateral resistance of the building. The total model consists of all structural walls in the building plus the deep spandrel beam-column frame along the north and ~outh perimeter of the building. Figure 7 Illustrates the three-dimensional computer model of the building showing the infill wall panels at the first floor, north elevation. The effective stiffness properties used herein are ,:onsistent with Methodology guidelines, SeCllO? ?~.2. For the perimeter frames, a panel zone rIgidity of 75 percent is used to account for the fact that the beams do not cover the full width of the columns. The analysis included P-delta, orthogonal effects, and accidental torsional effects. The accidental torsional effect was included by displacing the center of mass at each level by 5 perce?t Of. the dimension measured perpendicular to the directIOn of the applied force. The ortho?on~l effect is included through load co.mb~nal1ons by specifying 100 percent of the seismiC demand from one direction to be added to pe~cent of the demand from the orthogonal dlrecllon. All significant modes were included in the analysis. Individual modal responses were combined using the CQC procedure. The seismic base of the building is assumed at the Ground Floor level for the purpose of c~lc~lating the base shear. Building weight dlstnbutlOn used in the analysis consists of

3?

1390 kips at the Roof, 1424 kips at floor levels Seventh through Third, and 1747 kips at the Second Floor. The total seismic weight of the building is 10,257 kips.

4.2.4

Elastic Model Results

The three-dimensional eigenvalue analysis yielded the following values for the building period: 0.60 second for the fundamental period in the longitudinal (east-west) direction, 0.46 second for that in the transverse (north-south) direction, and 0.32 second for the torsional period. Table 4 provides a summary of the building periods obtained from analysis of the building response records obtained during the Northridge earthquake, building periods reported in earlier studies of the building after the San Fernando Earthquake, and the building period obtained using UBC 91 Method A formula (Co = 0.03 assumed). The significant lengthening of the period in the longitudinal direction during peak response of the San Fernando Earthquake as well as after 10 seconds of the Northridge Earthquake is attributed to the strength and stiffness degradation of the perimeter frames, particularly the south perimeter frame.

4.3

static Nonlinear (Pushover) Analysis

4.8.1 Intl"Dductlon The predicted performance of the structure was examined using a two-dimensional, nonlinear model to represent the expected seismic performance of the exterior spandrel beam-column frame, both in its existing state, and after the proposed rehabilitation. An additional equivalent frame model was created to check the capacity of the interior flat-slab-column gravity system to withstand the deformations imposed by the lateral system response.

4.s..~

T

mode guide of dif show. or 19' heigh distri! first r in Se( analy:

T

beam Figur, ORAl Amor gravit the cc displa progr: calcul allow

analy~

are ac

beam~

D·'.


--


els Table 4. Approximate Fundamental Building Period

e

.

'.

'. .

. .'

.

'

.

....

••••••• UBC 91 Method A

pre·1971 san Fernando earthquake ambient vibration

is

..

. ·.····E-WDlr. lLongltlldlnall

. ....

I

N-SDir. .' . (TranSVerse)

0.68 sec

0.68 sec

0.52 sec

0.40 sec

lod in

San Fernando earthquakel1971)

~cond

Early part of earthquake

0.70 sec

0.70 sec

.ion, ble 4

During peak response

1.5 sec

1.6 sec

Northrigde earthquakel1994)

lse ier o j using med). [he )f the ibuted le meter

Ire linear ,olumn

e alent ity of to

ateral

~ay

Inn

Early part 10 • 10 sec

1.5 sec

2.2 sec

Middle part110·20 sec)

2.1 sec

2.2 sec

Towards the end I> 25 sec)

2.4 sec

2.0 sec

4.502 Modeling The specific parameters used in the nonlinear model, in accordance with Methodology guidelines, are tabulated in Table 5. A comparison of different normalized lateral force distributions showed that a first mode shape, inverted triangle, or 1991 UBC lateral force distribution up the height of the building resulted in nearly identical distributions. The force distribution given by the first mode shape from the elastic model discussed in Section 4.2 was used in the static nonlinear analysis. The static nonlinear analysis, using beam-column subassemblies such as that shown in Figure 10, was conducted with the aid of the DRAIN-2DX computer program, Version 1.10. Among the numerous modeling parameters, gravity load induced moments, shear capacity of the columns, maximum base shear, and roof displacement limits were selected to overcome program limitations and simplify the resulting . calculations. Since the present program does not allow the inclusion of element loads in nonlinear analysis, the effect of dead and live load moments are accounted for by reducing the capacity of all beams by 10 percent.


Column Modeling 4.5oS Figures 8 and 9a illustrate a column interaction diagram and a moment-curvature diagram for a typical north and south perimeter frame column between the Fourth and Fifth Floors, respectively. Similar analyses were performed for the columns at the other floor levels. As noted earlier, "expected" values were used for the steel and concrete strength to calculate all capacities. Above the Fourth Floor, columns typically consist of 6-#7 vertical reinforcing with #2 ties (2 sets per location) at 12 in. on center. Below the Fourth Floor, columns consist of 6-#9 vertical with #3 ties (2 sets per location) at 1212 in. on center. Concrete of higher strength was used for the lower floor columns as shown in Table 2. The shear limit state governs the behavior of these columns, i.e. the shear capacity is less than the shear (2 MP/hol) associated with flexural hinging of the column ends. The limiting story shear capacity is dictated by shear limit states of individual columns and includes the contribution of interior as well as exterior frames. This effect was monitored manually during the analysis using the spring element shown in Figure 10.

D·tS


In til colu toth to c( mod resul asSUl

Actual Material Strength

2S% Increase from expected material strength with strength reduction factor equals to 1.0

Initial component Stiffness

75% gross moment of Inertia for Columns; 50% gross moment of inertia for Beams; account for effect of slab.

... not activated modeled as linear spring at the bottom of spandrel beam·column connection. see Figure 10.

"'*

4.S.4 Beam Modeling Figure 9b illustrates the moment curvature diagram of typical Fourth and Fifth Floor spandrel beams in the longitudinal direction. Typical stirrups consists of # 3 ties at 3 inches and 5 in. on center near the two ends and 10 in on center in the middle portion. For reinforcement, these beams have 2-#6 at the bottom, 3-#8 at the top and 2-#6 in the slab adjoining the beam. The flexural limit state governs the behavior of these beams. Because of low positive (tension at the bottom) flexural capacities of these beams, even with the effect of in place dead and live loads, the demand on many of the beams exceeded their capacities. This is consistent with the observed damage in the building where flexural cracks were observed at the bottom of several beams. 4.S.S Beam-Column Joint Modeling Based on probable demands on the beams and columns, it is estimated that the maximum shear demand on the beam-column joints at the lower floor levels were between 8 to I Of,0.5. It was noted in examining the beam column joint in Figure 6c that although the record drawings require several

D-'.

ties in the joint, the as-built condition appears to be somewhat different. At several joints where concrete had spalled off during the earthqual

respl Thes built

4.S.

1 thec for tI . excel and t deter deter modi the iJ prese modi

4.4

4.4. 1 devel Chap discu resul' Capa

4.4.; E of hiJ beam show figun hingi states II at the prese

assoc

------


In the present case, the shear capacity of the columns was assumed to be a uniform value equal to the capacity of the column ties only as a means to conservatively reduce the complexity of the model. For the purposes of scaling the pushover results, the base shear and roof displacement were assumed to be limited to 960 kips and 20 in., respectively, for evaluating the existing building. These limits were never exceeded prior to the building achieving a mechanism. Slab Modeling 4.5.6 The equivalent frame model used to evaluate the contribution of the slab was similar to that used for the spandrel beam-column frame, with the exception that the sum of the column dimension and three times the slab thickness was used to determine gross moment of inertia as well as determine the moment capacity of the slab. This model was generated to compare the capacity of the interior frames, and is rather limited in its present iteration in terms of predicting the failure mode of the various constituent components.

rs to ere ke, no le static Nonlinear Analysis Results Of,O.5 is 4.4 of Introduction . all lIC Y 4.4.1 minor The goal of the static nonlinear analysis is to mn develop a Capacity Spectrum as described in > Chapter 8 of the Methodology. This section ;t with " discusses the results of the analysis and how these :olumns results. were used to construct the required )ped in Capacity Spectrum.

I n shear I 4.4.2

pushover AnalysIs Results I Based on the pushover analysis, the sequence Ich andrei of hinging of beams, column shear failure at the I beam-column connection, and column hinges is I shown in Figure 11. The numbers shown on the i figure correspond to the order in which the beam e 'I hinging, column hinging, or column shear limit need to states occurred, respectively. It can be seen that the hinging generally begins J~tment '[ at the lower floors and progresses upward. Table 6 IS. i presents the base shear and roof displacement !lYSIS. ! associated with the first flexural hinging of a beam

I I

IIday

In~ Appendix D. Holiday Inn

Table 6. Base Shear and Roof Displacement Associated with First Beam and column to Hinge sase Shear at Hinge (kipSJ

Displacemel1t at Hinge (III.)

Beam

155

1,4

Column (Shear!

445

5,6

Column (MOment)

520

8,0

Element " ". , "Name

and column element as well as the first column element to fail in shear. Since the applied shear and roof displacement required for formation of the column shear failure are less than those associated with the flexural limit state in the column it appears that shear, rather than flexure, is the more~obably failure mechanism in columns. This appears to be consistent with the performance of the structure during the Northridge , Earthquake. Over 90 percent of the modal mass participated in the first mode, and from the story drift plot in Figure 12, it can be seen that the initial assumed first mode lateral force distribution is a reasonable assumption for this building. The pushover analysis was terminated after a column shear failure occurred in all columns just below the third level, resulting in a mechanism. This behavior is not entirely consistent with the observed damage from the Northridge Earthquake because the majority of the damage was observed in the floor above. Other studies have demonstrated that it is possible to develop an analytical model that more closely matches the observed damage when ground motion generated by the Northridge Earthquake itself is used. Nevertheless, the significant column shear deficiency appears to be recognized by the analysis and, given the assumed loading distribution, the analytically predicted failure is not unreasonable. Load-displacement results, first yield, major yield and initial deterioration points as defined in the Methodology are shown in Figure 13. As

D·n


shown in the figure, the existing building reaches a mechanism just beyond a total roof displacement of almost 16 in., which corresponds to an overall drift ratio of 2 percent.

capacity Spectrum Load-displacement and modal analysis results were combined to generate the capacity spectrum using the procedure outlined in Chapters 4 and 8 of the Methodology. The steps are: J. Calculate the ratio of base shear versus building weight (V/W). 2. Calculate the modal story participation factor and modal base shear participation factor. 3. Calculate the spectral displacement versus spectral acceleration.

Table 7. Demand RespDnse spectrum parameters seismic zone Factor

Z - .0.40

Near·Source Factor

N - 1.0

seismic Coefficient

c. ~ 0.40 Cv

4.4.8

4.4.4

Demand Response Spectrum The Basic Safety Objective selected for the case study is Building Performance Level "Life Safety" at the Design Earthquake and Building Performance Level "Structural Stability" at the Maximum Earthquake (ME). Descriptions of these performance objectives are found in Chapter 3 of the Methodology and Section 1.3 of this case study. The Design Basis Earthquake was assumed to be described by ground motion with a 10 percent .. chance of being exceeded in a 50 year period. The . Maximum Capable Earthquake was assumed to be described by ground motion with a 10 percent chance of being exceeded in a 100 year period, but not exceeding the maximum single event that can be foreseen within the geologic framework assuming median attenuation. A five percent elastic demand response spectrum, shown in Figure 14, was generated using the procedures outlined in Section 4.4.2.4 of the Methodology with the parameters listed in Table 7 for the Design Earthquake. The Design Earthquake is represented by the 10 percent damped inelastic response spectrum shown in Figure 14 using the spectral reduction factors for the acceleration (SRA) and velocity (SRv) controlled regions of the spectrum calculated using Equations

D·18

= 0.64

8-9 and 8-10 of the Methodology. The building as a whole must be checked for stability, strength degradation, and excessive deformation as described in Section 11.3 of the Methodology. Static inelastic analyses of this building showed no instabilities with respect to gravity loads. All performance point roof displacements in the Design Earthquake are less than the 0.02 x 65.67 ft x 12 inlft = 15.76 in. Life Safety Limit shown in Table 11-2 of the Methodology. Similarly, the Structural Stability Limit is found from the expression 0.33 V;/Pi = 0.33(815/4,661) = 0.058, which translates to Structural Stability Limit of 0.058 x 65.67 ft x 12 inlft = 45.5 in. The resulting demand spectrum in shown in Figure 14. Iterative procedures are needed to find the unique "performance point." The desired performance point is Point B in Figure 14, and it can be seen that the building is not capable of achieving this level of spectral displacement at the given load. For this reason, the rehabilitation scheme discussed in the following section is proposed.

4.5

Time History comparisons

Limited inelastic time-history analyses were executed as an approximate check of the performance point displacements predicted by the Methodology. A group of ten near-field acceleration records, each with components in twO directions, were selected. For each record, the given components were transformed to fault-parallel and fault-normal components. Scale factors were computed so that the average spectral accelerations of the 20 histories would be 0.64g for a structure with a I-second period. That is, the records used as time history input were scaled to


mate (unrf

1

DRA to thl push, roof time· very trend J. 1 I, P

2. 1

d b 1 undel oven a pus effec· force

s. 5.1

1 achie 14. 1 progJ and s objec Meth objec stabil p provi to sal inclu, Basel level. struc; each requi was!

IIppe

-


---

ters

g as

1

,d no

i.67 )wn the

1.058, )f

in find Id it f it the

ere y the in two he 5cale ,ectral 64g for he :d to

~ay In"

match a single representative point on the 5 percent (unreduced) spectrum for the Design Earthquake. Time-history analyses were performed using DRAIN -2DX. Five percent damping was assigned to the first two modes. Figure IS shows the pushover curve with combinations of maximum roof displacement and base shear taken from the time-history analyses. It can be seen that there is a very wide scatter in the results; however two trends appear relatively clearly: I. The time histories predict higher shears, at lower displacements, than suggested by the pushover curve. 2. The time histories predict higher maximum displacements, at lower shears, than suggested by the pushover curve. Thus, the pushover analysis appears to underestimate the maximum shear and overestimate the displacements. This suggests that a pushover analysis, in general, may miss critical effects of higher modes that increase component forces without increasing roof displacement.

S.

Rehabilitation Scheme

5.1

Introduction

The subject building is not capable of achieving the performance point shown in Figure 14. This suggests that a seismic rehabilitation program is required to provide the needed strength and stiffness to satisfy the required performance objectives identified in Section 1.3 of the Methodology. Two alternative performance objectives are considered: life safety and structural stability. A number of alternatives are available to provide the needed strength, ductility, and stiffness to satisfy the required performance characteristics including internal shear walls and external frames. Based on a review of the desired performance level, the existing architectural character of the structure, the level of disruption associated with each scheme, and access required to execute the required construction, the exterior frame system Was selected. Alternative analyses are possible

~ppencllx D, Holiday Inn

using the interior shear wall scheme, however, these are not presented in this paper because they do not further explain the application of the Methodology to the analysis and rehabilitation of concrete structures.

5.2

EXterior Frames

The Basic Safety objective requires that for the Life Safety Building Performance Level, that the designers enhance gravity resistance of the frame columns, limit deformation in ,the frame columns, and reduce the vulnerability of the frame columns to shear failures. Figures 16 through 19 illustrate the conceptual distribution and member sizes of the exterior frame concept shown in plan, elevation, and details. Using the methodology outlined in developing the nonlinear model for the existing frame, Section 4 of this case study, a pushover analysis of the structure was undertaken. A preliminary approach to sizing these frames uses hypothetical spectral pushover curves to find performance points within required deformation limits. Spectral values at the hypothetical performance points are then converted back to absolute values, and the required strength and stiffness of additional frame elements can be determined. Assumptions inherent in the new design include: •

The new mode shape matches the existing mode shape. An modal participation factor P = 1.4 was assumed, and an appropriate alpha for the hypothetical performance point was chosen from the evaluation data.

•

Initial and post-yield stiffness of the hypothetical pushover curve match the existing building.

•

For the hypothetical strengthened building, the initial yield point is appropriate.

•

New frames have lower yield displacements than do the existing frames.

•

The effects of new frame weight and material properties can be ignored.

D·'.


---------------------------------------------------------------------------------

The results of the analysis are presented in Figure 20 showing the new capacity spectrum for the rehabilitated structure. A similar analysis was conducted for the structural stability performance goal. As can be seen from Figures 21 and 22. significantly less rehabilitation work is required to satisfy the requirements of this performance goal. Figure 23 illustrates the capacity spectrum for this more modest rehabilitation effort.

G.

Concluding Remarks

This report presented an application of the Seismic Evaluation and Retrofit of Existing Concrete Buildings to the Holiday Inn in Van Nuys, California. The purpose of this example building study was to illustrate the use of the Methodology document as an example for other engineers to follow. The analysis of the existing structure concluded that it could not satisfy the requirements of the assumed level of seismic performance and a

D·20

----

seismic rehabilitation scheme consisting of exterior concrete frames investigated. The resulting rehabilitated structure satisfies the required seismic performance level. The static inelastic analysis appears to do a reasonable job of identifying critical limit states in the structure and provides a simplified design criteria against which the rehabilitation of the building can be undertaken. Although there are a number of other evaluation methods, the guidelines contained in the Methodology appear to offer the design engineer a well-structured approach to evaluating and seismically rehabilitating existing concrete structures. Additional case studies are available that examine the application of the Methodology to other structures. These should be consulted to obtain a broader picture of how this methodology should be applied and to the1\nge of engineering judgment required to evaluate and seismically rehabilitate existing concrete buildings.

APpendix D, Holiday In"

Appe

------

------------------------------------------------------------------------------------------------

------

-----------------------------------------------------------------------------------------------------


Iday Inn Appendix D. Holiday Inn

D-21


llu,,; ~.

UJIO

North/South Perimeter Frame Clevation Example Building Study Holfday Inn. Van Nuy., CA

D·22

nCUR[

2

Appendix D. Holiday .JIII

-

--

,---------------------------------------------------------------------------------SEISMIC EVALUATION AND RETROI=IT 01= CONCRETE BUILDINCS ,----------------------------------------------------------------------------------

~

~ •c

~

~

~

a a

~

~

a a

~

~

,r:. ..

b~

~

"

!

,

"

.

~'

~

,

b

~

,

,

,

,

,

,

'"

~ ~

,

"

"" ~

~

'" ~

,

... ,.-

16'

SECTION-B

"

SECTION-A

,

" "

,

~~~

~ f5~

SLOPE 1:6

~~~

MAXII.IUt.l~

"I u, "••

'. ,

.,.

, ,

• .'

1-'-7,f-

"

..

" "

.

'

. .. ..

= PRESENT

TIES MAY NOT BE 'N THE AS-8UILT-----1 •

'

.

,

"

. ' ,

,

,

.

"

, "

,

, ,

STRUCTURE

,6'

::.::I!

a;J ~ ~,

~

~

~ "

'/ •

EXPANDABLE

JdI NT

1:-:.

..

~+

"o

ANGLE J"X,J"

MATERIAl

CONT.

BRICK WALL 24"o.c.

SECTION-D

4-14 CONT.

--L.,tt---1-14 CONT. SECTION-C

Beam Sections and Elevation at Column Example Building Study

FIGURE

3

Holiday Inn, Van Nuy.. CA

' Appendix D, Holiday Inn ,llday In!,

D-21


-

~ FI

i

z

~

! I •

I ;

2gb

~'I; ~.

U."'.D

North/South Perimeter rrame Showing Northridge Earthquake

Damage

Exempt. Bundlng Study Hollda, Inn, Van Nu,a, CA

D-24

nCUR[

4

Appendix D, HOliday Inn

N,

---------

~----------------------------------------------------------------------------------

SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS ~----------------------------------------------------------------------------------

FIGURE 5(a) North Frame Elevation Showing Damage

FIGURE 5(b) Close-up of North Frame Column Shear Damage

FIGURE 5(c) Beam Column Joint Damage at North Frame

Damage Photographs

Example Building Study Holiday Inn. Van HUYSt CA

Iiday 1l1li


fIGURE

5

D-25

FIGURE6(a) Column Shear Failure

FIGURE6(b) Column Shear Failure

F/GURE6(c) Beam Column JOint Damage - Note Absence of Specified JOint Reinforcing

Damage Photographs Example Building Study Holiday Inn, Van Nuys, CA

D·26

FIGURE

6 Appendix D. Holiday I"~


ETABS

moc.~

with infUl wall

Existing Structure

IGURE example BuDding Study

F'lGURE

7

Halldl, Inn. Vln NUYI. CA

6


D·27


--

PCACOL V2.30

.

r--'-1000

-\ ------~-------~ -·· ,,

-\

.

.------~.------~-------~--------~------.-

---

: ,

, -------:--------~------------

··

,,

, ,

,,

600Ir~~~~~~~~~~~~=~~~~~_~_

200 I' ------ ,

,,

:

: fs=O.Sfy :

:

I

•

•

I

• •

I

•

,

•

-:-------:-------~--------'

300 ~Mny

(ft-k)

-200

, , -----------.-------~-----------------------., -600 , ........... _- . . _______

"

oi>_ ..... _ - - - ....... - - - - - - - - - - ... - - - - - _ ..... - - . . . .

Existing Column Interaction

F1C~RE

Diagram at 4th and 5th floors eumpt. aulldln, Sbldy

8

Haltday Inn. Van Nuya. CA

D·28

Appendix D. Holiday 1l1li

--


IS)

fm ~

'2

120

I

"0..

:.E

1040

=

~ ~u

:Cii

20. in..

Lw"

~90 ~ ....,

5

3i":O. psi

i'j::

14-. in. O. l
=

Ot;C""..:ity::=l

1.ca

sc~,

1

JO-'" ", j

o .. a

.

92

1$4-

.

.

.

.

m

2:0

:!22

CURVAn.i;<:: (rcd/in x 10-6) Mo",.ni Cutva;t:r. Oia..,.em - 14-" x 20" Coit.:rms FIGURE 9(a) . Existing Column Mame,rt-Curvature Dia...",

in-

~

--""-

!.lor CUd1y

*

_

• __ __ m __•

a+---~~--~

a

~

~

~~

~

CIJRVA~ (rr::(1fI

•

__

~

1~

=

it.

n ...

-

Q.1oips

8.IJ

____

~

.

~

_

~

_

x '~l

FIGURE 9(/l) ExisIing Beam Moment-Curvatura Diagram

Existing Column one S.am Mamlnt-Curvatvr. Diagrams ~.

Building Study HOlIday bin, Von IIuyo. CA


9

D·29


I I I

BEAM COLUMN ELEMENT RIGID END ZONE

____ .J

NODE

BEAM COLUMN ELEMENT

PLASTIC HINGE

ZERO LENGTH HORIZONTAL SPRING ELEMENT (MONITOR SHEAR FAILURE IN COLUMN)

ACTUAL BEAM AND COLUMN SHOWN IN DASHED LINE FOR REFERENCE

Droin-2DX Modeling of Concrete

fiGURE

Beam-Column Joint Ex.mple BuildIng Study Holld_», Inn. Van Nuy., CA

D-:SO

10

Appendix D. Holiday In"

Ap~

~

J

,.

;;

J

.

:

iiC

1

!~

:z: !:!.

ROOF

o

678

5

4

3

2

7TH

I.PIJ___ U~ __ .00 120

61H

51H

• & --91.•

I

30 T28

ii I :1 H Z

Q'"

~~

•

!..

0'"

I-

3RD

.17

~'!'

~'"

::

~ 9.

2ND

I

•

~ ~... I

1ST

,I::

81

115

t ---•• --. ,_ .••. _..••. - •• .• !!'16--· II:: 84 '09· •~ -tl~ 34 103

82

·.'11-·

~~ _ _ .B~13_2. 153

.76

33102

37 104

1:: 81

83

36 105

1:

35 114

1::

42

'12--· '0-8- . '07-- •

I. 13 ••.• • . _. ·;.j;2-·-;~44--·;~43

65.1 15

-;~42

70 21 - -

56• • 19_43• •8_74• •6_~2• • 9-.:?3.1.1O..._75.1.769.1.18 38 40

t t

~6. ~ 149

139

86

95

1

93

85 1:: 78

129 1~~ 1~1 1~ 1~8 53• •127 •.. ~8• • _~2• • _ _5~ • _ _ ~ __ ~ .?_59.

.u7

H

80

.77

20 BB 1 28 49 112 64 1 11 63.1 14 87

2.:c

J[ -g.;: ,.

4TI-I

122

141 106 145

•

.62

1

'17

23 22 4

57

-.134

3~

141

140

II45

I135

4-58

147

99

100

97

96

!l8

92

.101

- - '" - - - - - - • Beam'CoILnTI Plaslic t-linge

m

.

C

II

r:

~

o Z II Z

a

..'" ',.-" .o.

m

o

55.

·46 148

CoILnTI Shear Failure

-" ut

-,--------r--,-----,,------. --_.

j

~

ut

m

,. "oz "..'m"

m

.-III

r:

a

Z ut

n

.. a•

III


~

~...

-

5

,

4

I

Base Shear Level

--.-

-+-

3

......-

2

)E

-+-

V=10Ck V=2O(k V=3Q(J( V=4OCk V=5OOk V=6OOc (d > 2%driiI)

,

I..

1 0

2

6

4

STORY DRIFTON)

Story Orift Ptot. Existing Siructun

Eumt>to

,

Building Sbld,

a

10

"GURE

12

HDlIdo, Inn. Van "UYO. CA

D-S!


AII~


----- 1---------------------------------------------------------------------------

Fare. Displacement Existing SInlc:lu..

CUI"YI.

!DmpIe . . . - . Slucly H - , ..... Ya Nuya, CA

laylnn


FlGtlR£

13

D·:!:!


Capacity Spectrum. Existing Siructurl ~.B_I

Stud,

RGURt

14

HallOy 1M, Va. Hup, CA

D-34


-


-

1000 LEGEND

900

• Fault Normaa. Muimum Sh••,

o Fault NonnaJ. Mulmum Displacement ... Fault PInIUet. Maldmum Shear .6 Fault Parallel. Maximum DIsplacement

800

700

lil Il.

eCap

• Pet

4Cor

-'!C. eCor eNH ;l'He RI. ICS" sr, "'Cap ....... CE

600

eEC7

-- --

... eC7r

g

eSCE

a:

iii::c en W en ~

500

o Per "Rln

400 ',/

300

. .-

.-

.-

A'c.~SCE I

- - - ':>sci,.

,6Cor

SCS

ANH

,/ ,/

ORIn

,/

AeC7

,/

200

Cor A .t OSy!

'ii

100

a

ARln

o a

2

4

6

ap

8

10

12

14

16

18

ROOF DISPlACEMENT (IN)

Time-History/Pushoyer Comparison. Existing Structure

E.urn!>lo

a-. Study

15

HoIIdoy Inn. Von Nup, CA

~ay

Inn


D-35


0

_.

..... ....r

.

9

8

C

Ill" • I

I

~

A

. ..

----------- : ,,,

II II

II _______I";.-;M

~fi

v~~.'r MDIOIt :n -

-

B

, I

--

,

,

~ 7 ~

•

"'+

6

~ 5

,

,

r-,

~

. ~

-'4"

@-

TItCII C:OIICII£Tt

k4~"'.II.o.

-,

~~~

' - ,,,

~.

4

,, ,, ,,, ,,

: 3

l -

2

,

"

:

!I>
-

---------

~

___________ ~

---------- "" ----

---j ~

Typical Floor F'romlng Pion Life Sofety Retrofit Exempli Building Study

,",,"""

.,""""

FIGURE

16

HolldlY Inn. Vln Hurl, CA

D-:sa


-

-


~w

-.

u.N.G.

South Perimeter Frame Elevation : Lit. Safety Retrofit Enmple BuRdlng Stud)'

F'IGURE

17

Hollda, Inn, Van Muva, CA

lay Inn


D·S'


GRID I

I

EXIST1N

G

COLUMN

VNEW

CONCR8~

rRAME COLUMN BEYOND

SEYOND

/

j-:-:' . •. ~ ..

, 8 0

W

,

•

. -L. '.

I:'

"

, ,-,-

'.

.• :

;~.~ J,.~ NEW

.• '.;: . ), . ...• f·

\0

•

.

• .;-

[T'

,'.

•

EXISTIN G B E A M / :

'/i

"'g

b" , ' "

•

~

FRAME BEAM

"c

,

N

~ N N

"'0 0' ~o

Lo..~

I

,...

f"

N

N "0 0

"'

\ '- 8-#9

~

"-8"

TOP & Bon. #4 TIES 0 4' o.c.

:

I

I

SECTION FRAME BEAM SECTION Enmple 8undlng Study Holkl.r Inn. Ven HurSt CA

D-SS

rlGURE

18

Appendix D. Holiday I""

-

-


GRID

I I

EXISTING

BEAM~

EXISTING CONCRETE COLUMN ~

I I I I I I

. L . ",'J.. ~ •

•

I.

___ .

•J

~+ .

.

• • 1---'-

I I I I I I

~

.

I I I I

NEW FRAME BEAM

/

NEW FRAME COLUMN

. ~

8-#9

_____ #4 TIES @ 4" o.c .

11.' ...

to

•J •

'"

..' •

2-#9

. .

GRID

8-#9

~

EXISTI~

BEAM

/

NEW FRAME BEAM

1'-8~

PLAN VIEW FRAME COLUMN OET AIL

Eumpl. BuDding Study

FIGURE

19

Holldl,. Inn. Y_n Nur., CA

lay Inn


D·SS


1.2 I ~I I

1.1 1.0

L.

r-

----:2.5Ca~

1:1

a . .. ""'4I----=----"'tll .f Darrsge Ca1IrCI

-

0.8

I

:~

1:

0.9

:§

I

::. I

I

I::J

armx=O.37g ay=O.35g drTax=3.6"

cI"9

e

i

0.7

I

0.5

~

1!y=2Z' EfI'edIve Darqling =21 %

0.8

0.4 0.3

21% EIfedive Oarr-ping Spednm I

0.2 0.1

-- I

0.0

--~_=_-:-:-:-~----=-_=_.:l- TckcCe

- - ---

r

1L-..J...:::...J:.....:J.~L...-l-LL-...L...!....-.l..--L.--L...wL...-l-.l-...!....-.l..--L---I

o

1

2

3

4

5

6

7

8

9

10 11

12

13

14

15

16

17

18

SPEC"TlW.DlSI'IACEMENT. Sd ON)

Capacity Splctrum. Ufo-Sofoly R.trofit

Eumpt. 8 _ . Study

20

Holiday Inn, 'OIl Nuya, CA

D-4D


---


o

c

8

A

6J'-r

9

[B iif I

I

~

•

/

,

··• :

II

______ ~-iI

·:l/i""xJ(I" ~~

r~r·lHr IKl'tIIIM

=ntCOl.uu..

(l(I[IIIOR

· -""'" ---1f

8

~ 7 ~

•

"

/

6 H - I - - - - ..I!------<;----. -----"'-l1li1-'

\:~ :

\

•

5

i" ··•

~

•;

@--m.

4

• •

--.

L-!

:

19

I"I'P.

~ ----@jm

~ 3

•; 2

.

"

1\:5>«1

~

• -----------

:1[-----------1 ----, ,-

_______ .ll! _____ . _____ :' >-• •

____ ..J•

Typical rloor Framing Pion

FIGURE

Structural Stability Retrofit Example Building Study Holkl8W Inn, V.n Nuy., CA

day Inn


21

D·Qt


iii

i i i i

South Perimeter Frome Elevation : Structural Stability Retrofit

Ex.mple Building Study Holiday Inn. Van HUYI, CA

..,.. D-42

22

Appendix D. Holiday Inll

Ap~


-

TS
1.2 1.1

1.0

2.5Ca

0.9

§ c2l

~

O.S

!

0.7

I

0.5

~

anm:=0.24g ay=O.25g .dTai( 4.5'

=

0.6 I

. t:larr1Jing = 34% ~

0.4 0.3 -'.... Pel fa Ii ElliCe Point (4.5'. 0.24g)

0.2 0.1

: Redu:ed Spe~~;;~~_~,~_:-:-::""",-,,,::,-...;-:.....::...::....:..:j- Td-ccde

-- -

2

3

4

5

6 7 8 9 10 11 12 13 14 15 16 17 18 SPECTRAl.. DlSf'lACEM3'.IT. ScI ON)

Copocily Spectrum. SI",clurQl-Stobilily Retrofit

Exoa"Ie

8.-.

Sludy

FIGURE

23

Holiday ..... V. . Nup, CA

day Inn


D-43


APpendix E

Cost Effectiveness Study prepared by Jimmy R. Yee Consulting Engineers 4850 Alta Drive Sacramento, California 95822

Appendix E. cost effectiveness study

E·1


Table of Contents I. General ..................................................................................................... E-5 2. Example Building Descriptions and Retrofit Schemes ................................................ E-5 2.1 Barrington Medical Building .............................................................. E-5 2.2 Escondido Village Midrise ................................................................ E-6 2.3 CSU at Northridge Administration Building ........................................... E-6 2.4 Holiday Inn at Van Nuys ...................................................... ·............ E-6 3. Example Building Demolition/Replacement Costs .................................................... E-6 4. Example Building Retrofit Costs ......................................................................... E-7 4.1 Definition of Retrofit Costs ............................................................... E-7 4.2 Retrofit Costs and Performance Levels ................................................. E-8 4.3 Itemization of Retrofit Costs ........................................................... , .. E-8 5. Benefits/Costs ............................................................................... ·.............. E-13 6. Comparison with FEMA Projects for Estimation of Seismic Rehabilitation Costs ............ E-14 6.1 Typical Costs of Seismic Rehabilitation of Buildings ............................... E-14 6.2 University of Southern California Medical Center .................................. E-15 7. Ease of Use of the Seismic Retrofit Analysis ........................................................ E-15 7.1 Traditional Approaches ............................... ,.~ ................................ E-15 7.2 Analysis and Retrofit Design Methodology .......................................... E-16 8. Consistency of Application of the Evaluation and Retrofit Methodology ........................ E-17 8.1 Preliminary Evaluation .................................................................. E-17 8.2 Modeling ................................................................................... E-17 8.3 Nonlinear Static Analysis ................................................................ E-17 8.4 Foundation Effects ....................................................................... E-18 9. Cost Effectiveness of the Evaluation and Retrofit Methodology ................................ E-18 10. References ................. , ................................................................... , ......... , E-18

AppendIx E, cost EffectIveness Study

E-5

~----------------------------------

SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS ~-----------------------------------

APpendix E

cast Effectiveness study t.

General

A benefit/cost model study was undertaken to analyze the results generated from the four example building studies reported in Appendices A, B, C, and D in this volume. This study reports on the feasibility of using the structural performance levels for seismic retrofit designs and on the application of the seismic retrofit analysis and design methodology contained in Volume" 1 of this document. The nonstructural performance levels were not considered in the example building analyses and therefore were not addressed in this benefit/cost study. The objectives of this study are to: 1) estimate example building retrofit costs (direct or hard costs), 2) provide a comparison between the example building retrofit estimated costs and cost ranges from traditional retrofit analysis methodologies, 3) develop a relationship between costs and extent of change in the retrofit due to selection of the performance level for the four example buildings, 4) perform a benefit/cost analysis for the performance levels evaluated for each example building, 5) evaluate the ease of use of the seismic retrofit analysis, and 6) identify the consistency of application between engineers. The reader should be aware that most of these retrofits were designed based on known actual earthquake damage. Differences in retrofit schemes and costs may result in cases where the benefit of knowing the actual earthquake damage to a particular building is not available or has not been realized. All of the example buildings occur within the high seismicity areas of California. Therefore, the conclusions drawn from this study may not be fully applicable to other geographic locations.

Appendix E, cost Effectiveness study

2.

Example Building Descriptions and Retrofit Schemes

The four example buildings consist of the Barrington Medical Building in Los Angeles, Stanford University Escondido Village Midrise, CSU Northridge Administration Building, and the Holiday Inn in Van Nuys. For all buildings, seismic retrofits for one to two structural performance levels were completed using the newly developed methodology. A summary of the existing buildings and their retrofit schemes are given below.

Barrington Medical Building The building was designed in 1964 and is located in Los Angeles. The building has approximate plan dimensions of 104' by 130' and is six stories in height with no basement. The floor systems consist of a cast-in-place (CIP) two-way concrete 7-1/2" flat slabs. The apparent lateral force resisting system is a combination of CIP perimeter moment resisting frames with "short" columns ( between spandrel beams) and shear walls. The foundation system consists of CIP drilled concrete piles. Structural Stability Level Retrofit. Supplemental steel column support of the slab and spandrels at the perimeter moment resisting frame columns or alternatively, the strengthening of the perimeter moment frame "short" colu~s by th~ use of a fiber reinforced epoxy composIte materIal 2.1

(FRP).

Life Safety Level Retrofit. New CIP concrete infill waIls, shotcrete strengthening of existing walls, the addition of pile caps and drilled CIP piles, and all items in the Structural Stability Level Retrofit.

E-5


Escondido Village Mldrise The building was designed in 1964 and is located on the Stanford University campus in Stanford, California. The building has approximate plan dimensions of 114' by 57' and is eight stories in height with a basement. The floor systems consist of a 12" voided CIP concrete one way slabs with integral beams. The apparent lateral force resisting system is CIP concrete shear walls. The foundation system consists of spread footings. Life Safety Level Retrofit. Confinement of the concrete columns and shear wall boundary members in the lower three stories by the use of steel plates and fiber reinforced epoxy composite (FRP) material, the welding of short longitudinal bar splices in shear wall boundary elements and the strengthening of the column-slab connection using a steel collar at all levels of the structure. Damage Control Level Retrofit. New full height concrete shear walls and drilled pier foundations, confinement of the concrete columns in the lower three stories by the use of steel plates and fiber reinforced epoxy composite (FRP) material, and the strengthening of the column-slab connection using a steel collar at all levels of the structure.

2.2

CSU at Northridge Administration Building The building was designed in 1964 and is located on the CSU campus in Northridge, California. The building tower has approximate plan dimensions of 227' by 68' and is five stories in height with no basement. The floors consist of 19" deep cast-in-place (CIP) one-way joist and slab systems. The lateral force resisting system is CIP concrete shear walls. The foundation system consists of CIP drilled concrete piles. Immediate Occupancy Level Retrofit. Strengthening of existing shear walls using shotcrete and the strengthening of existing shear wall construction joints. Life Safety Level Retrofit. None required.

2.3

E·a

-

2.4

Holiday Inn at Van NUys The building was designed in 1965 and is located in Van Nuys, California. The building has approximate plan dimensions of 150' by 63' and is seven stories in height with no basement. The floor systems consist of 8" ± thick cast-in-place (CIP) two·way flat plate slabs. The apparent designated lateral force resisting system is ClP perimeter concrete moment resisting frames and interior slab-column frames. The foundation system consists of ClP drilled concrete piles. Structural Stability Level Retrofit. Addition of reinforced concrete ductile moment resisting frames up to the fifth floor at the perimeter of two sides. The transverse direction retrofit is not addresse~ in the example building study. Life Safety Level Retrofit. Addition of reinforced concrete ductile moment resisting frames full height of the building at the perimeter of two sides. The transverse direction retrofit is not addressed in the example building study.

I.

Example Building Demolition/Replacement Costs

The building demolition/replacement cost for each of the example buildings was estimated for comparison with the estimated construction costs for the retrofits. These costs were established through a professional cost engineer in conjunction with general construction contractors. The demolition/replacement costs do not include any soft or indirect costs, or contingencies. For purposes of demolition/replacement costing, direct costs are defined as construction (primary) costs including mechanical, electrical, plumbing and architectural features. The contractor's field/home office expenses, profit margin, and bonds are included in the direct cost. All costs are based upon the Engineering News Record Cost Index of May 1, 1996. The replacement construction is

Appendix E, cost Effectiveness studY

cons mate inclu code for e Tabl,

4. !

were will, tecru build simil

4.1 I this I cons! by rr this! profe vend Conti

losse impa inclu can I non-I

-


Table 1. Building Demolition/Replacement costs ~

has md is , floor IP) ated

BuildIng

Barrington Medical Center stanford University Escondido Midrise CSU Northridge Administration Bldg. Holiday Inn at van NUys

r

:onsists lition ng ftwo

frames 'two

for for :osts d mction ;t

$151.30

$7,485,000

53,400

$140.20

$16,208,000

120,200

$134.80

$8,613,000

69,700

$123.60

Example Building Retrofit Costs

Definition of Retrofit Costs

Estimated retrofit construction costs given in this study are direct costs. Direct costs are the any construction costs (primary) and were determined by material quantities and associated unit prices in , direct this study. Unit prices were established through a :osts professional cost engineer in conjunction with rod vendors, material suppliers, subcontractors, and IIhome contractors. Indirect or soft costs such as housing re losses, business and industry loss, relocation :ed upon impacts, employment loss, and tax impacts are not f May I, included in this study. Direct construction costs can be further divided into earthquake and non-earthquake related costs (Hart, 1994).

Appendix E, Cost Effectiveness study

Estimated ",cost/SF

72,500

A variety of materials and retrofit systems were selected for the four example buildings. This will give the reader information as to what retrofit techniques can be utilized for existing concrete buildings and also the construction costs for similar buildings associated with these techniques.

4.1

.Bulldlng Area '{$FJ

$10,970,000

considered "in-kind" consisting of "like type" materials of the existing facility. No betterment is included, except for meeting the current building code. The estimated demolition/replacement costs for each example building are summarized in Table 1.

4. lent

EStlma,ted ." Dem(UReplacement. cost

Earthquake related costs for this study include the structural work to upgrade the existing lateral force resisting system and/or to provide additiomil vertical stability to the gravity load carrying system. Damage repair is excluded from the direct costs. Non-earthquake related costs include the non-structural work of removing and replacing architectural, mechanical, plunibing and electrical features in order to carry out the structural work. For this study, the non-earthquake related costs are included in the total direct cost for the building but will be separated from earthquake related itemized retrofit costs. Non-earthquake related costs which are excluded from this study are: • Fire and life safety system improvements • Mechanical, electrical and plumbing system improvements • Architectural improvements • Hazardous material abatement • Providing access for the disabled • Retrofit costs include the contractor's field/home office expenses, profit margin, and bonds. All costs are based upon the Engineering News Record Cost Index of May 1, 1996. The retrofit costs do not include any soft or indirect costs, or contingencies. The building is assumed occupied, except for the portions of the building under construction. The influence on the cost of the

E·7

----------------------------------------~-------------------


..~ Table 2. seismic Retrofit com VS. performance Levels

Barrington Medical center

structural Stability Life safety

stanford University Escondido Mldrlse

Life safety Control

CSU Northridge Administration Bldg.

Life safety

None 5170,000

None

Immediate Occupancy

Holiday Inn at van Nuys

Structural stability Life safety

5800,000 $1,530,000

511.50

retrofit by a fast-track or phased construction schedule or by pressures to reopen or occupy the building as soon as possible is not considered in this study.

4.2

Retrofit Costs and Performance Levels

A Summary of the estimated retrofit costs for each structural performance level is given in Table 2. The retrofit costs for the four example buildings show a direct increase in construction cost incremental with the selection of a higher performance level. For the CSU Northridge Administration Building, the existing structure was determined to have satisfied the requirements for the Life Safety and Damage Control performance levels without retrofit. The existing Stanford University Escondido Midrise structure was considered to be already at the Structural Stability performance level. The Holiday Inn's total retrofit cost was based upon the assumption that the retrofit in the transverse direction would be approximately the same as in the longitudinal direction of the building. The relationship between retrofit costs and changes to the structural performance level for each example building are graphically illustrated in

E-8

5550,000

57.60

51,700,000

523.50

5460,000

58.60 529.96 51.40 522.00

Figures 1 through 4. A composite graph including all example buildings is given in Figure 5. CSU Northridge Administration Building, Figure 3, illustrates the situation where the seismic resistance of the original structure is at a high performance level already. The incremental cost for the Immediate Occupancy level is small as a result of this condition. The generation of these Cost-Performance curves allows for an estimation or extrapolation of incremental performance level changes with only a very few points. However, future studies including additional performance levels and example buildings are recommended to determine the accuracy of the extrapolation.

4.3

Itemization of Retrofit costs

The detailed construction cost estimates for the retrofits allowed itemization of the various tasks involved in the retrofit. Although these tasks are specific to the four example buildings, it is expected they will be representative of the kind and cost of retrofits which may be encountered for other similar concrete buildings. Graphs are given in Figures 6 through 9 illustrating the various retrofit tasks and associated unit costs for each example building.

APpendix E, Cost Effectiveness studY

--

-


Figure 2. stanford Escondido village Mldrlse cost·perfOrmance curve

55 studY Appendix E. cost Effectiveness Study

E-9

----------------------~----------~---------------------------

SEISMIC EVALUATION AND RETROFIT DF CONCRETE BUILDINGS

2~~~----------------------------------------~ 16.64 CSU Northridge Administration Bldg. 13.31

9.98

~

6.66

400000

3.33

0+------,------~-----r------r_----~~--_4 0.00

Not Considered

Structural Stability

Limned Safety

Lije Safety

Damage Control

Immediate Occupancy

Performance Level

Figure s. CSU Northridge Administration Building CDst·performance curve

20~

28.70

Holiday Inn at Van Nuys 16~

22.96 11.

10

8c:

1200000

17.22

U

c:

~

2 10 c:

~0 0

800~

11.48

0

U

tl::I ~

10 c: 0

U

400000

5.74

0

0.00 Not Considered

Structural Stability

LImited Safety

Life Safety

Damage Control

Immediate Occupancy

Performance Level

Figure 4. Holiday Inn cost·performance curve

11 1

E-10

APpendix E. Cost Effectiveness stucIY

j

IIPII


40,-------------------------------------------, Building Case Studies LL

~6

1 "o

o

30

20 Legend

-v- Holion, .•:l1n .CSUN Admin -eStanfortCEVM

10

. . .Me

Not Considered

Structural

Stability

Limited Safety

Life Safety

Damage Control

Immediate Occupancy

Performance Level

FIgure 5. Building case studies cost·performance curves

Barrington Medical Center

Add Pile Foundations Supplemental Column Support or

Strengthen Columns (FRP) CIP Cone. Infill Wall Shotcrate Walls Mech .• Elect., Plumb., Arch O.H., Expenses. Profit. Bonds

o

1.00

2.00

For LIfe Safety Performance Level

4.00 5.00 3.00 Construction CosVSF

6.00

7.00

8.00

FIgure 6. Barrington Medical center Retrofit Tasks/costs

s studY

Appendix E. cost EHectlveness study

E-n


Stanford University Escondido Village Midrise Strengthen Rebar Splices Strengthen Columne (FRP) Slab Shear Strengthening at Columns Strengthen cols & walls (steel plata) Mach" Elect. Plumb.. Arch

-F~===c~

D.H., Expanses, Profit, Bonds

5.00

For life Safety Pertonnence Level

7.00

8.00

Construction CosVSF

FIgure 7. stanFord IInlverslty EScondIdo Village Mldrlse RetroFIt TaSks/costs

s. perf orig othe 1.

CSU Northridge Admin Bldg. Strengthen Wall Construction Joint

2.

3. Mach., Elect•• Plumb.,

!

c.H., Expenses, Profft, Bonds

this cost 199: cost (FE

0.0

1.00

2.00

For Immediate Occupancy Performance Level

5.00 3.00 4.00 Construction CosVSF

6.00

7.00

8.00

FIgure 8. CSII Northridge Administration Building RetroFit Tasks/costs

E·12

Appendix E, Cost effectiveness studY

repl

----


-

Holiday Inn at Van Nuys Concrete Moment Frame

Mach .• Elecl., Plumb., Arch

D.H., Expenses, Profit, Bonds

.00

o

1.00

2.00

3.00

For Ufe Safety Performance Level

4.00 5.00 6.00 7.00 Construction CosVSF

B.OO

9.00

10.00

Figure 9. Holiday Inn RetroFit Tasks/Costs

5.

Benefits/Costs

In this study, an increase in the structural performance level from that established for the original building is considered as the "benefit". In other studies, retrofit benefits typically include: 1. Lowered repair costs. 2. Reduced loss of building function and therefore indirect costs. 3. Improved life safety for occupants. Although indirect costs are not addressed in this study, they may be at least equal to the direct costs of repair and retrofit of the building (FEMA, 1992a). For a more thorough discussion of indirect costs, the reader is referred to the references (FEMA, 1988 and FEMA, 1989). It is apparent from reference to the replacement costs that the direct costs of the

studY

AppendIx E, cost EffectIveness study

retrofit are small in comparison to the replacement costs. In this study, the direct costs of retrofit range from 5 percent to 18 percent of the replacement cost, excluding the special case of the CSU at Northridge Administration building. Consideration needs to be given to the indirect costs in the project to obtain a rigorous cost effectiveness analysis. The expected effectiveness, or reduction in damage, of traditional concrete building retrofits for the Life Safety performance level over non-retrofitted concrete buildings averages from 30 percent to 18 percent, for MMI IV to MMI XII level seismic events respectively (FEMA, 1992a). As a result of the more exact analysis methodology proposed herein, a further reduction in damage is expected.

E·IS


eno Me' low

Barrington Medical Center

Large

concrete Frame with InflllWall

Life safety

$29.93

516.54

stanford University

Large

Concrete Shear wall

Life safety

526.13

57.45

Large

concrete

Damage

531.67

527.59

5hearwall

Control

Escondido Mldrlse Stanford university EsCOndido Mldrise

abo retr redt and elef con

mel foUl

CSU Northridge Administration Bldg.

Very Large

concrete Shear Wall

Immediate Occupancy

511.72'

Holiday Inn at Van NUYS

Large

concrete Moment Frame

Life safety

529.93

inel

51.10

retr· Saft retr·

517.30'

NOI

1.

Seismicity Category of Very High, 1996 construction start and 4% inflation.

2. 3. 4.

Immediate Occupancy minus Damage Comrol level costs

Star Wh, reqt stre

Non·structural work costs omitted or not considered. Cost for longitudinal direction retrofit used also for transverse direction.

G.

Comllarlson with FEMA projects for Estimation of Seismic Rehabilitation Costs

6.1

Typical Costs of Seismic Rehabilitation of Buildings

Typical retrofit costs for each example building were determined using the FEMA document Typical Costs of Seismic Rehabilitation of Buildings, Volume I, Second Edition (Hart, 1994) to obtain a comparison with the building specific cost estimates from this study. Cost option 2 of the FEMA methodology was selected to determine the typical retrofit costs. This option accounts for building type, floor area, geographic location, date of retrofit construction. seismic map area. performance objective. the number of buildings and other considerations. The FEMA document

CSt

min waI Nor be < elefi eart alre the

considers only the direct cost of the structural work for retrofit and does not address retrofitting to performance levels below Life Safety. Therefore. cost comparisons at the Structural Stability performance level were not undertaken. Table 3 presents the direct structural retrofit costs estimated for five separate retrofit schemes developed in the four example building studies of bast this project and the corresponding mean costs belc determined using the FEMA document. Thi! Comparison indicates that the estimated costs of reh~ the retrofit schemes developed using the perl Methodology are all lower than the FEMA mean fror costs. in some cases substantially lower. However. Imn one must be very cautious interpreting this limited expJ data. The FEMA cost estimation methodology. , the· based on the averages of large numbers of highly incr variable costs. is specifically not intended for use Dafi in estimating the retrofit cost for individual i Imn buildings and this comparison does not provide

1

E·'4


1~

-

-

:ing

en. ,fit nes s of

of ean fever,

nited

y, ~Iy

'use de

studY


enough data to support a claim that the Methodology will consistently deliver significantly lower cost retrofit designs. On the positive side, however, the comparison above does suggest that the potential exists for retrofit cost reduction using the Methodology. The reduction can be attributed to the ability to define and focus on retrofitting the most critical structural elements. This ability stems from the more complete understanding of earthquake demands, member limit states, failure mechanisms and foundation effects associated with the inelastic/nonlinear anal ysis. Furthermore, the high variability in estimated retrofit costs, notably for the higher than Life Safety performance levels, can be explained by the retrofit measures established for the CSU Northridge Administration Building and the Stanford University Escondido Village Midrise. Whereas the Stanford University Escondido Midrise required major concrete shear wall and foundation strengthening to improve the performance level, the CSU Northridge Administration Building required minimal strengthening of one line of concrete shear wall. The minimal retrofit work for the CSU Northridge Administration Building may, in part, be attributed to knowing the exact structural elements to be retrofitted based on previous earthquake caused damage and the presence of an already adequate lateral force resisting system for the higher performance levels. The FEMA document and database appear to be based on the existing building condition being at or below the Structural Stability performance level. This is reasonable since facilities are not typically rehabilitated if they meet the Life Safety performance level. Therefore, the incremental cost from Life Safety or Damage Control to the Immediate Occupancy performance level is not explicitly addressed by the FEMA document. For the CSU Northridge Administration Building, the incremental cost was obtained by subtracting the Damage Control level retrofit cost from the Immediate Occupancy retrofit cost.

AppenCllx E. Cost effectiveness study

6.2

University of southern California Medical Center

A study was undertaken by FEMA to ascertain the need and extent of retrofit for the Psychiatric Hospital at the Los Angeles County USC Medical Center (FEMA, 1995). Structural Damage was sustained during the 1994 Northridge Earthquake which indicated a need for seismic retrofit. In this case, the structural elements to be retrofitted were damaged by the 1994 Northridge Earthquake. Analyses using recently published research on reinforced concrete shear walls in the post-elastic range helped to confirm that the "short" column or wall pier elements were critical. A construction cost estimate was undertaken in the FEMA study for I) damage repair and 2) the "Optional Structural Hazard Mitigation Scheme". The second scheme was defined as the retrofit of 119 selected exterior wall/columns which would result in a building performance level greater than or equal to Life Safety. From the detailed estimate, the hard or direct cost for the "Optional Structural Hazard Mitigation Scheme" was determined to be $3,098,148. This cost does not include the scope and construction contingency, but does include non-structural costs. This cost figure differs from the subtotal given in the FEMA study and is not the net cost change from the base repair scheme. With a floor area of 115,030 SF, a cost per SF of $26.91 was calculated for this particular scheme. Comparing the $26.91/SF for this building and the average Life Safety level cost from th:is study in Table 2, the FEMA retrofit cost is approximately 50 percent higher.

7.

EaSe of use of the Seismic Retrofit Analvsls

7.1

Traditional Approaches

Traditional detailed seismic retrofit analyses of existing concrete buildings in structural engineering practice have typically incorporated the use of an equivalent static lateral force

E·15

-------------------------------

..


------------------------------------. procedure or linear dynamic analysis (response spectrum) to determine deficiencies. The analysis criteria consisting of a percentage of the force level required in recent editions of the Uniform Building Code (UBC, 1994) or the use of the force level required by an older edition, such as the 1970 UBC, has been adopted by many municipalities and state agencies. In other cases, the detailed analysis procedures established by the Federal Emergency Management Agency in the NEHRP Handbookfor the Seismic Evaluation of Existing Buildings (FEMA 178) or Methodsfor Evaluating the Seismic Resistance of Existing Buildings (ATC-14) have been used or approved in lieu of some percentage of the UBC equivalent static lateral forces. FEMA 178 and ATC-14 address archaic and non-ductile lateral force resisting systems in high seismic zones, whereas the UBC does not. The differences between FEMA 178 and ATC-14 included the fact that FEMA 178 earthquake forces are at strength level and the ATC-14 forces are at allowable stress design (ASD) level. When converted from ASD to Strength methodology, the force reduction factors, R, does not always agree. There is also a concern whether these global force reduction factors are too conservative for existing buildings. The above criteria are all very similar to the methodology used for design of new building structures, so those familiar with current building codes could easily convert to these procedures for existing buildings. We would expect that a new methodology for seismic retrofit analysis, which includes inelastic effects modeled explicitly and considerations of restricted component ductility, would be more time consuming in understanding and implementing.

7.2

Analysis and Retrofit Design Methodology Results from a questionnaire given to the structural engineers performing the building case studies indicate that the new methodology, using the Capacity Spectrum Method, is more time consuming than those in previous guidelines, such as the detailed analysis procedures of ATC-14,

!

FEMA 178, or the UBC. The extensive amount of 1 eml time indicated to implement the methodology may j rela can be, at least in part, due to the time to learn and utili understand the developing methodology and to .!' as d familiarize themselves with the presently available nonlinear analysis computer programs. The ! pusl demands of time required to: I) generate the non various demand curves, 2) determine the nonlinear elas capacity curve by manual iteration using elastic info analysis computer programs or semi-automatic und iteration using nonlinear computer programs, and desi 3) find the performance point, including the time part to learn the tools of the methodology, appear to to b result in an increase in retrofit design time by a anal factor of two to four times that of traditional mO( analyses. Also, the influence of the structural eval engineer's background and experience with abil performance-based earthquake engineering of post concrete structures is also a consideration. resi: Another important factor in the analysis time to b is the limitation of the computer software. One wid software program used by all the building case rna) study participants is the DRAIN 2-D (two allo dimensional) series of non-linear analysis retT! programs developed at the University of California, Berkeley (PoweJI, 1992). Three 8. dimensional nonlinear analysis programs with a similar origin to the DRAIN 2-D series are also available (Maison, 1992) to explicitly account for torsion. These programs do not have pre- or post-processors to easily examine data and do not explicitly account fOf all the types of concrete rece member failures or desired force-deformation for I relationships. As an illustration, the global metl capacity curve for each building was established by the participants using incremental pushover 8.1 curves to account for the degradation of members with restricted ductility. This procedure turned out for 1 to be very lengthy. Post-processors were typically rneti generated by the use of spreadsheets. Even with eval the spreadsheets, examination and transformation the I of the data is a time consuming process. The rem participants all indicated they are unaware of any FE~ commercial software program which completely automates the nonlinear static analysis process,

I

1 :j

E-'&

Appendix E, COst Effectiveness studY

--It of

nay I

able near c md me

o a

me

a


emulates the desired member force-deformation relationship and transforms the data into that which can be easily evaluated. Only one participant utilized a three dimensional linear analysis program as the primary tool for the non-linear static pushover analysis. However, the methodology's incorporation of nonlinear analysis gains insight unavailable from an elastic code based type of analysis. More information is obtained and a more complete understanding is gained using the new analysis and design methodology. As an illustration, one participant noted that buildings which may appear to be highly deficient when evaluated with elastic analysis methods can be demonstrated to be moderately deficient, or even adequate when evaluated with the nonlinear analysis methods. The ability of the nonlinear analysis to provide post-yield force redistribution in the lateral force resisting system was considered by all participants to be an extremely important benefit. In summary, with the methodology, critical structural elements may be identified with greater confidence thus allowing for easier determination of the degrees of retrofit necessary for the desired performance level.

B.

;0

for not

ed r

bers d out :aUy ith tion any ely s,

studY

Consistency of Application of the Evaluation and Retrofit Methodology

Only portions of structural calculations were received from the building case study participants for use in evaluating consistency. These indicate the methodology was generally followed. 8.1

preliminary EValuation AU, except for one participant, used FEMA 178 for the preliminary seismic evaluation phase of the methodology. The FEMA 178 preliminary evaluation procedure appears to have found most of the obvious deficiencies in the buildings. The remaining participant did not indicate the use of FEMA 178, but instead used an alternate analysis

Appendix E. Cost Effectiveness study

methodology (Inelastic Demand Ratio) for the preliminary evaluation. 8.2

Modeling Building models for the nonlinear analysis using DRAIN-2DX varied between participants. CSU Northridge Administration building was modeled using one line of coupled shear walls. The Holiday Inn model consisted of complete moment resisting frames in the longitudinal direction. Stanford University Escondido Valley Midrise was modeled using multiple stick elements to represent the shear walls and moment resisting frames. The Barrington Medical Cen.ter model consisted of complete moment resisting frames. Interior shear walls were modeled as single beam-column elements. The effects of the different types of model representations on their results has not been established. In all the buildings, except for one, the interior gravity load resisting beam and column frames were included in the analysis. Contrary to the methodology, it appears that these frames could have been omitted in several of the shear wall building models. All of the participants took into account the effects of member strength degradation in their model. NOnlinear static Analysis Three of the participants selected the level 3 Capacity Spectrum Method which uses the shape of the first mode response as the basis of the lateral load distribution. The remaining participant selected the level 2 Capacity Spectrum Method which follows the code type triangular force distribution. In several of the buildings, the selection of the level 3 and under Capacity Spectrum Method was found to result in the underestimation of the higher mode effects when compared to nonlinear time history analyses. Assuming the nonlinear time history analysis to be correct (or at least better able to predict maximum forces since the effects of higher modes are represented), an increase in maximum member

8.3

E·n


shears could result in increased retrofit work on shear walls and-moment resisting frames from that given in this study. However, careful review of time-history analysis results is required to assess whether peak forces are appropriate retrofit design criteria.

8.4

Foundation Effects The effects of foundation flexibility was studied by three out of the four participants. The participant, which did not consider foundation flexibility, concluded it would have little effect on their particular retrofit design. The effect of foundation flexibility was found to be of significance for the Barrington Medical Center. Considering estimated foundation flexibility, the capacity curve was controlled more by pile slip at the foundations than by frame hinging and shear wall failure. In this case, consideration of foundation flexibility could lead to a reduction in retrofit work and therefore construction costs. The Stanford University Escondido Village Midrise model showed minimal change due to foundation flexibility considerations. Whereas the initial inelastic behavior of the structure was dominated by foundation rotation and rocking, subsequent behavior was still controlled by the superstructure components yielding.

9.

Cost Effectiveness of the Evaluation and Retrofit Methodology

This limited study indicates the potential for identifying construction cost savings in retrofits provided by the choice of.different performance levels combined with the new evaluation and retrofit methodology. Comparison between expected Life Safety retrofit mean costs from a recent FEMA project (Hart, 1994) and that of this study shows an average 40 percent decrease in direct costs with the use of the new evaluation and retrofit methodology.

E·'8

6.

The increase in design time presently associated with the new methodology, which equates to an increase in engineering fees, can be rationalized by the savings in construction cost and increased confidence in predicting performance levels. In addition, as the methodology is used more, enhancements to computer analysis and design software should result. These software enhancements will automate the tasks, incorporate failure mechanisms and force-deformation relationships now not addressed and thereby significantly decrease the design time associated with the new methodology. Until such time as the nonlinear computer programs are coded specific to -this process, the design time and therefore engineering fees may be much higher than the previous approaches have traditionally required.

'0. 1.

7.

8.

References ATC, 1987, Evaluating the Seismic Resistance of Existing Buildings (ATC-14), Applied Technology Council, Redwood City, California. \

!

2.

1 FEMA, 1988, Typical Costs for Seismic Rehabilitation of Existing Buildings, Volume - 1I Jl- Supporting Documentation (FEMA 157), Federal Emergency Management Agency (FEMA), September 1988.

3.

FEMA, 1989, Establishing Programs and

Priorities JOT the Seismic Rehabilitation oj Buildings - A Handbook (FEMA 174), Federal Emergency Management Agency (FEMA), May 1989. 4.

FEMA, 1992, NEHRP Handbookfor the

Seismic Evaluation of Existing Buildings (FEMA 178), Federal Emergency Management Agency (FEMA), June 1992. 5.

FEMA, 1992a, A Benefit-Cost Modelfor the

Seismic Rehabilitation of BUildings, Volume 1: A User's Manual (FEMA 227), Federal Emergency Management Agency (FEMA), April 1992.


lI


6. be and

'ate

d the c to

FEMA, 1995, Los Angeles County / University of Southern California Medical Center, Psychiatric Hospital - First Appeal Response Findings, LACO 2641, FEMA IO08-DR CA 037-91033, Federal Emergency Management Agency (FEMA), October 16, 1995.

7.

Hart, 1994, Typical Costs of Seismic Rehabilitation of Buildings, Volume I SUMMARY, Second Edition, FEMA, July 1994.

8.

Maison, 1992, PC-ANSR, A Computer Program for Nonlinear Structural Analysis,

developed by B.F. Maison, Berkeley, California. 9.

Powell, 1992, Drain 2DX, Static and Dynamic Analysis of Inelastic Plane Structures, developed by Allahabadi, Prakash, and Powell, University of California, Berkeley.

10.

UBC, 1994, Uniform Building Code, Volume 2, International Conference of Building Officials, Whittier, California.

11.

UBC, 1970, Uniform Building Code, International Conference of Building Officials, Whittier, California.

I.

~),

:ity,

Ime

57),

d

if

2. . the Ime

'al \),

ItudY

Appendix E. cost effectiveness Study

E·1.

--------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

APpendix F

Supplemental Information on Foundation Effects

Appendix F. supplemental Information on Foundation Effects

F·1


Table of Contents 1. Introduction ............................................................................................... F-5 2. Seismic Performance of Building Foundations ................... , ................................... F-6 3. Historical Perspective on Foundation Design ......................................................... F-7 4. Pertinent Research ......................................................................................... F-8 5. Key Conclusions ........................................................................................... F-9 6. References ............................................................................................... F-9 Resource Summary 1 ....................................................................................... F-12 Resource Summary 2 ....................................................................................... F-24 Resource Summary 3 ..................................................•.................................... F-26 Resource Summary 4 ....................................................................................... F-30 Resource Summary 5 ....................................................................................... F-35 Resource Summary 6 ................ : ...................................................................... F-37 Resource Summary 7 ....................................................................................... F-40 Resource Summary 8 ....................................................................................... F-43 Resource Summary 9 ....................................................................................... F-48 Resource Summary 10 ..................................................................................... F-52 Resource Summary II ..................................................................................... F-54 Resource Summary 12 .................................................................... , ................ F-61 Resource Summary 13 ..................................................................................... F-64 Resource Summary 14 ..................................................................................... F-69 Resource Summary 15 ..................................................................................... F-73 Resource Summary 16 ..................................................................................... F-82 Resource Summary 17 ..................................................................................... F-84 Resource Summary 18 ..................................................................................... F-88

Appendix F. Supplemental Information on Foundation Effects

F·!


APpendix F

Supplemental Information on Foundation Effects 1.

Introduction This Appendix is a supplement to Volume 1 of

Seismic Evaluation and Retrofit of Existing Concrete Buildings (Report No. SSC 96-01) Product 1.211.3 of the Proposition 122 Seismic Retrofit Practices Improvement Program. The methodology includes a Chapter 10 on the effects of foundations on seismic response. This supplement compiles and summarizes selected information and research on the seismic performance of building foundations. The guidance in the general methodology for foundation effects is based in part on the summarized material. The user of the general methodology can refer to this material as a resource for the practical evaluation and retrofit of concrete buildings. The methodology focuses on the structural analysis of concrete buildings to evaluate anticipated performance during earthquakes. Since seismic forces are proportional to mass it is logical to consider them a property of the building itself. Accelerations which combine with the masses to generate the seismic forces also are related to the stiffness properties of the building. While the magnitude and distribution of these forces are affected by the properties of the building, in reality the initiating cause of seismic forces is the shaking motions of the ground beneath it. As the building responds to the ground movement, forces are limited by the stiffness and capacity of the foundation and soils materials. The types of mechanisms at the foundation which limit seismic forces include rocking or uplift, crushing of soil material, and sliding. Neglecting these effects can lead to two quite different, yet equally undesirable, results.

1. The use of traditional, force-based analysis procedures often results in the prediction of unrealistically large forces beneath existing buildings, particularly with a fixed base modeling assumption. The engineer is left with a fundamental dilemma. The rocking, crushing and sliding of the foundation implied by the large forces, in many instances, can dissipate energy relatively harmlessly and protect the superstructure from damage. Stiffening and strengthening existing foundations to avoid these behaviors is costly. The result of the retrofit conceivably may be to transfer the energy dissipating damage to the more sensitive structure above. It is possible, and for some buildings probable, that retrofitting of foundations will result in poorer performance for higher cost. 2. Foundation effects typically reduce the force demand on the primary lateral resisting elements such as shear walls. At the same time, however, the rotational flexibility of the base of the shear walls often result in larger lateral displacements of the entire structure. The larger drifts can lead to failures in the beams, columns, or slabs of vertical load carrying system. There is evidence of this type of damage from past earthquakes. Traditional fixed base analysis techniques do not identify this potential for serious damage. The difficulty for the practicing engineer is that there have not been procedures and methodologies for design that treat foundations and soil properties explicitly. Geotechnical engineers provide soil capacities and displacements developed to limit long term settlements due to


F·5

-------------------------------------------------------------, SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

vertical loads. Seismic forces and displacements in soils and foundations are intrinsically nonlinear and difficult to include in traditional linear analyses. Recent proposals to change the approach to evaluation and retrofit design for existing buildings are displacement-based and provide an opportunity to treat foundations much more rationally (ATC, 1995). Within the methodology presented in Volume 1, Chapter 10 on foundation effects presents procedures to include the foundation directly in the structural model for analysis in an effort to capture this behavior. The objective is to facilitate a more accurate prediction of building behavior than that provided by analysis techniques which neglect foundation effects. A group of practicing structural and geotechnical engineers (Geo-Structural Working Group) developed the procedures in Chapter 10 based on their collective experience and knowledge in several basic areas including: • Performance of building foundations in past earthquakes, • Past and present structural and geotechnical design practice for buildings, • Theoretical and empirical research on foundation performance. The following sections of this supplement summarize the current state of know ledge in each of these key areas and offer suggestions for future development to improve understanding of foundation effects. Attached also are brief summaries of papers, articles, research reports, and other pertinent documents.

2.

Seismic Performance of Building Foundations

Society's knowledge of earthquakes and their effects on the environment is empirically based to a large degree. This has lead to an intense interest on the part of the scientific and engineering community in reconnaissance after major earthquakes to document the performance of building and their various systems, elements and components. The result has been an increasingly

F·G

complete record of performance providing valuable guidance for both researchers and designers. Naturally, however, the tendency has been to concentrate on visible damage to buildings and structures. Explicit by its very nature, "nonperformance" is more dramatic than "performance". The more subtle issues of why some systems and elements may have performed well and how this performance may have affected the overall response of buildings receive relatively less attention. One example is building foundation systems. Admittedly, there are abundant, welldocumented examples of damage to foundations in earthquakes. These examples, however, are predominantly the result of permanent ground displacements caused by liquefaction or subsequent block sliding, seismic compaction, and other effects independent of the ground shaking response of the building itself. Damage to foundations resulting from building response to seismic shaking is difficult to document since most foundations are not easily accessible for inspection. Damage to structural components of deep foundations generally consists of distress in pile or pier caps, or at the tops of the piles or piers. themselves where loads are transferred to the caps. Evidence of downward or upward movement of piles and piers themselves is rare. A notable exception is the behavior of piles in highly sensitive clays in Mexico City. Structural damage to shallow foundations components is also rare and usually concentrated at the connection of the structure to the foundation. Soil bearing failures in compression due to increased contact pressures caused by seismic overturning forces are very difficult to document. Mexico City is, again, the exception. Some poorly designed bearing foundations with very marginal factors of safety under static loads appear to have contributed to major structural failures. Evidence of uplifting of bearing foundations where seismic forces exceed compensating dead and live loads is more common. The direct consequences of uplift normally have not been life threatening. The


glot Strul the:

. I.

resi! (SE. allo· bew bear (lCI buill defil bear squa desi: rudi actu joinl San Civi Ass( seisl reco the j

(Ani

diffe eartl inch base Buil inch ener reco buil(

whic stru( reco pred Wou

sam,

-laS

lings Iny

led cted ively Ition

ns in

quent ponse

of. ~

in

piers caps. of

nage 'e and

:0

lent. oorly inal lave !nce .mic ,ds is plift

ffeetS


global consequences of foundat~on uplift on o~erall structural behavior are not readIly dlscermble In the field without supplemental theoretical analysis.

S.

Historical Perspective on Foundation Design

The first provisions for design of buildings to resist earthquakes appeared in U.S. codes in 1927 (SEAOC, 1990). These original provisions allowed lower lateral loads on structures with better foundation conditions in terms of allowable bearing pressures. The Uniform Building Code (ICBO, 1994) of 1935 recognized lower forces for buildings founded on "good" soils, which were defined as those which would have an allowable bearing pressure of greater than 2,000 pounds per square foot. These initial provisions for seismic design of buildings were based on very rudimentary understanding of how structures actually respond to ground shaking. In 1948, a joint committee on lateral forces, c~nsistin~ of the San Francisco Section of the Amencan SocIety of Civil Engineers and the Structural Engineers Association of Northern California, recommended seismic design provisions which included recognition that forces in buildings were rela~ed. to the fundamental period of vibration of the bUlldmg (Anderson, 1952). The observed performance of different types of structural systems during actual earthquakes led the engineering community to include modification to design force requirements based on building systems in the 1960 Uniform Building Code. These "K" factors were meant to include consideration of the structure to absorb energy without failure. The K factor was the first recognition of ductility in seismic performance of buildings. The 1960 code also introduced a factor "J" which reduced design level overturning forces in structures for two reasons. First of all, it was recognized that the story shears that were predicted to occur in the building at each level would not necessarily reach their maximum at the same time, thereby resulting in an over-estimation

of overturning effects. Secondly, it was suspected that building foundations tended to move slightly under earthquake loads, relieving some of the forces that may otherwise have been generated. The J factor, however, was eliminated from the codes in the early 70s due to damage which occurred in structures during the Caracas earthquake. In the early 1970s procedures were developed to modify the base shear coefficient by a soils factor" S" to reflect the fact that local soils conditions affect seismic ground motion. The S factors were included in the 1976 Uniform Building Code. Effectively, the S factor extends the maximum plateau of the design spectrum to include buildings with longer periods founded on relatively softer soils. These factors are based on empirical data and not directly upon site soil conditions. In the late 70s, provisions were developed to approximate the effect of th~ . interaction of buildings and their foundatIons with the supporting soil (ATC, 1978). Seismic design provisions for buildings are normally incorporated into building codes which have sections relating to the various materials (wood, steel, concrete, masonry) used for construction. These chapters provide design specifications and material properties for all of the elements of the structure. The Uniform Building Code provides a table of allowable working loads for bearing and passive resistance for several generic soils types. These may be used in lieu of specific recommendations by a geotec~ical . engineer. Geotechnical engineers routmely provIde structural engineers with allowable soil loads for specific building sites and foundation systems. These allowable working loads were normally permitted to be increased by a factor of 1/3 for short-term loads such as wind or seismic. The allowable loads usually are based on considerations to control long-term vertical displacements due to dead plus real live loads . Ultimate capacities and stiffnesses for short term loads are not provided except for special cases. In the elastic analysis of structures for working loads,


F·7


the structure is part of a larger system that includes the foundation and a supporting medium with their own strength and stiffness properties. This has been termed inertial interaction effect. Technically, the spatial variability of ground motion over the area of the building is also a parameter. This effect, known as kinematic interaction, is the difference between free field ground motion and the actual input to the structure and its foundation. Inertial interaction is generally more important than kinematic for typical buildings and foundation systems. The procedures in Chapter 10 of Volume 1 neglect kinematic effects. Past theoretical and empirical research indicates that foundation strength and stiffness has a significant impact on the response of buildings to seismic demand. Studies of performance of specific buildings during past earthquakes document the importance of including the foundation in the structural analysis model for design (Rutenberg, 1982; Wallace, 1990). Housner (1963) initiated the practical theoretical consideration of foundation rocking which has been developed further by others (Priestly, 1978; Psycharis, 1982). Large scale shaking table tests have confirmed the influence of uplift on seismic response (Hucklebridge, 1978). Simplified theoretical modeling techniques demonstrate good correlation with more rigorous solutions for both single and multiple degree of freedom systems as well as empirical data (Bartlett, 1976; Chopra, 1985; Vim, 1985). These have been extended to study the practical behavior characteristics of typical buildings (Nakaki, 1987). Pertinent data on the nonlinear stiffness and strength properties of soils materials themselves are rare. Theoretical elastic stiffness properties are fairly well documented for shallow foundations (Gazetas, 1991). The inelastic behavior of piles and piers subject to seismic demand has received some attention recently in the research community (Pender, 1993; Martin, 1995). Research on the behavior of bridge foundations provides valuable

WO I

Appendix F, Supplemental Information on Foundation Effects

App,

it often is assumed that the foundation of a building is fixed obviating the need to consider stiffnesses or displacements due to short term loads. The current NEHRP provisions (BSSC, 1995) specify the analysis and design of buildings for seismic loads according to ultimate strength concepts rather than working stress. These provisions also include a chapter on foundation design requirements. This chapter specifies that soil materials have sufficient capacity to support all the loads supported by the foundations including seismic loads. The use of ultimate capacity for soils materials as opposed to the working stress values previously utilized for design represents a major change. The provisions note that the determination of soil capacities shall be based on acceptable levels of strain considering the short duration and dynamic properties of the loading. Specific guidance on the determination of ultimate capacities and deflection is not provided. In another significant departure from previous procedures, the provisions require investigation of site hazards, including slope instability, liquefaction, and surface rupture as a result of earthquake motions. The treatment of geotechnical and foundation materials in seismic design procedures for buildings has been implicit in most cases. Specific guidance on the design and behavior characteristics, similar to that included for structural materials and systems, has not been included.

4.

Pertinent Research

Procedures for including the effects of the interaction of structures with their foundations first appeared in building standards in the 1970s (ATC, 1978). Veletsos (1988) developed the basis for these procedures by separating the interaction into two components. Seismic response of a structures traditionally has been evaluated assuming a fixed base and considering the motion of the base to be equal to the free-field ground motion. In reality

F·B

insig 1989

5. 1 effec ofbt conc

•

(

s t

•, (

I I

•

• •

• • 5. revil pro\ met! mea pert' reso

j

-


insights for buildings as well (Lam, 1991; Levine, 1989; Martin, 1995),

m

5.

ture Illy Ires

The prevailing situation with respect to the effects of foundations on the seismic performance of buildings is characterized by several key conclusions: • Geotechnical and foundation response can significantly influence the performance of buildings during earthquakes, •

has ~s

to

•

• :al

78; sts nic ;ood oth ; as

Key Conclusions

Costs of rehabilitation of existing buildings warrant realistic assessment of seismic performance of components including foundations and underlying soils, Traditional seismic design procedures do not reflect realistic consideration of geotechnical and foundation effects, Proposed methodologies require estimates of force-displacement relationships for foundation elements and geotechnical components subject to seismic loads,

•

Existing empirical data on geotechnical materials is inadequate to provide sufficient design information for all cases,

•

Research to investigate force-displacement behavior of geotechnical materials and foundation assemblies is needed,

• to

Damage reconnaissance for earthquakes should include documentation of geotechnical and foundation effects on buildings,

Anderson, Blume, Degenkolb, Hammill, Knapik, Marchand, Powers, Rinne, Sedgwick, and Sjoberg, 1952 "Lateral Forces of Earthquake and Wind," Transactions of ASCE, Vol. 117. Applied Technology Council, 1995, Guidelines and Commentary for the Seismic Rehabilitation of Buildings (ATC 33.03), 75 % Draft, Redwood City, California~ Applied Technology Council, 1978, Tentative Provisions for the Development of Seismic Regulationsfor Buildings (ATC 3- 06), Redwood City, California. Bartlett, P. E., 1976, "Foundation Rocking on a Clay Soil", University of Auckland, School of Engineering, Report No. 154, M. E. Thesis. See Resource Summary Number 1. Building Seismic Safety Council, 1995 (and other editions) NEHRP Recommended Provisions for the Development of Seismic Regulations for New Buildings, Federal Emergency Management Agency Publication 222, Washington, DC. Chopra, A. K., and Yim, S. C-S, 1985, "Simplified Earthquake Analysis of Structures with Foundation Uplift", American Society of Civil Engineers, Journal of Structural Engineering, Vol. 111, No.4, pp. 906-930. See Resource Summary Number 2.

ld

G.

'es :s are es ved unity Ie Ible

The Geostructural Working Group has reviewed selected references, as noted below, to provide further insight to engineers using the methodology, The Resource Summaries are not meant to comprise a complete review of all pertinent literature, since many other useful resources are available. The Geostructural Working Group chose these as particularly useful

ffeetS


IS

References

based on their own individual practical experience. The intent is to assist users of the methodology by providing a starting point in their search for more detailed information.

Gazetas, G., "Foundation Vibrations", Foundation Engineering Handbook. See Resource Summary Number 3. Housner, G.W., 1963, "The Behavior ofinverted Pendulum Structures During Earthquakes", Bulletin of the Seismological Society of

F'9

.SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

America, Vol. 53, No.2, See Resource Summary Number 4. HuckJebridge, A. A. and Clough, R. W., 1978, "Seismic Response of Uplifting Building Frames", American Society of Civil Engineers, Journal of the Structural Division, Vol. 104, No. ST8, pp. 1211-1229. See Resource Summary Number 5. International Conference of Building Officials, Uniform Building Code, Whittier, California, multiple editions. Lam P.I., Martin G.R., and Imbsen R., 1991, "Modeling Bridge Foundations for Seismic Design and Retrofitting", Transportation Research Record 1290, Proceedings of the Third Bridge Engineering Conference, Denver, Colorado. See Resource Summary Number 6. Levine M. B. and Scott R. F., 1989, "Dynamic Response Verification of Simplified BridgeFoundation Model" Journal of Geotechnical Engineering, ASCE, Vol. 115, No.2, See Resource Summary Number 7. Martin, G. R. and Lam, 1. P., 1995," Seismic Design of Pile Foundations: Structural and Geotechnical Issues", Third International Conference on Recent Advances in Geotechnical Earthquake Engineering and Soil Dynamics, St. Louis, Missouri, See Resource Summary Number 8. Martin, G. R. and Yan, L., 1995, "Modeling Passive Earth Pressure for Bridge Abutments", ASCE Conference Geotechnical Special Publication # 55, Earthquake Induced Movements and Seismic Remediation of Existing Foundations and Abutments, San Diego, CA, See Resource Summary Number 9. Martin, G. R., 1988, "Geotechnical Aspects of Earthquake Engineering", Journal of the Australian Geomechanics Society, Special

F·1D

Issue, 5th ANZ Geomechanics Conference, See Resource Summary Number 10. Nakaki, D. K., and Hart, G. C., 1987, "Uplifting Response of Structures Subjected to Earthquake Motions", U. S.-Japan Coordinated Program for Masonry Building Research, Report No. 2.1-3, Ewing/Kariotis/Englekirk & Hart, See Resource Summary Number 11. Pender, M. J., 1993, "Aseismic Pile Foundation Design Analysis", Bulletin of the New Zealand National Society for Earthquake Engineering, Volume 26, No. I, See Resource Summary Number 12. Priestly, J.N., Evison, R.J., and Carr, AJ., 1978, "Seismic Response of Structures Free to Rock on Their Foundations", Bulletin of the New Zealand National Society for Earthquake Engineering, See Resource Summary Number 13.

Wal

Psycharis, I. N., 1982, "Dynamic Behavior of Rocking Structures Allowed to Uplift", Report No. EERL-81-02, Earthquake Engineering Research Laboratory, Cali fornia Institute of Technology, Pasadena, CA, See Resource Summary Number 14. Rutenberg, A., Jennings, P. C. and Housner, G. W., 1982, "The Response of Veterans Hospital Building 41 in the San Fernando Earthquake", Earthquake Engineering and Structural Dynamics, Volume 10, pp. 359379. See Resource Summary Number 15. Structural Engineers Association of California,

Recommended Lateral Force Requirements and Commentary, Sacramento, California, multiple editions. Veletsos, A.S., Prasad, A.M., and Tang, Y., 1988, "Design Approaches for Soil-Structure Interaction", Proceedings of the Ninth World Conference on Earthquake Engineering, Tokyo, See Resource Summary Number 16.

Appendix F. Supplemental Information on Foundation Effects:. I\pPE

j

-ce, 'ling

ling


Wallace, 1. W., Moehle, J. P., and MartinezCruzado, J., 1990, "Implications for the Design of Shear Wall Buildings Using Data from Recent Earthquakes", Proceedings of Fourth U.S. National Conference on Earthquake Engineering, Palm Springs, CA, pp. 359-368. See Resource Summary Number 17 •

Yim, S. C-S and Chopra, A. K., 1985, "Simplified Earthquake Analysis of Multistory Structures with Foundation Uplift", American Society of Civil Engineers, Journal of Structural Engineering, Vol. Ill, No. 12, pp. 2708-2731. See Resource Summary Number 18.

.on

1978, 'the

f

ornia See

G. o ~d

5915. I,

nts ia,

Icture "orld

ffeets

Appendix F, supplemental Information on Foundation Effects

P-'I1

---------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

MF. as

Resource Summary' Bartlett, P. E, Foundation Rocking on a Clay Soil University of Auckland, School of Engineering, Report No. 154, M. E. Thesis, November 1976 This is a comprehensive theoretical and experimental study of the moment-rotation relationship of a spread footing on a clay soil. The effect of uplift and soil plasticity are included. Generally good correlations between theoretical and experimental results were observed. The effect of foundation rocking and yielding was found to lengthen the natural period of vibration of the structure and can be expected to lead to a reduction in internal structural forces for earthquake motions. This work concentrates on the development of relationships between moment and rotation for a rigid foundation rocking on a clay soil. These relationships were found to be both amplitude and path dependent. Overturning Moments on Spread Footings. The interaction of foundations with the supporting soil has been modeled in two distinct ways. The solution for a rigid foundation supported on a linear elastic half-space model is based on an assumption of complete continuity of the subgrade. This means that the deformation of a particular point beneath the footing is dependent on all the loads acting at all points on the contact surface. This formulation is complex and rigorous solutions for moment rotation behavior are confined to a few simple foundation shapes. As shown in Figure I, the contact stress beneath the circular plate is not necessarily proportional to displacement. The extension of the linear foundation model to include inelastic soil behavior is theoretically possible but practically very complex. A simpler model, particularly for inelastic solutions, is the modified Winkler model. In this

F-12

1 with

j model, the contact pressure beneath any point ther. beneath the footing is assumed to be proportional . conti to the deformation of the soil (see Figure 2). If a reael model of a rectangular footing is subject to vertical .i the r, load as shown in Figure 3, the resulting downward ! folio' displacement is ~

v

ro = ks LB In the above expression 'Yo is the initial contact stress, ks is a stiffness coefficient for the soil material, and L and B are the footing dimensions. Using this basic model, the author develops a moment-rotation relationship for overturning forces on the footing in terms of the initial contact pressure as follows

I less t fullc FOr!

2, yi. .full r FigUi capac

r =....!iJL. o

F.., ks

In this expression, qu is the ultimate strength of the soil material and Fv is a factor safety against bearing failure. This factor safety normally varies by design between 2.0 and 3.0. For some structures on particularly settlement-prone soils the factor safety may be somewhat higher to control vertical settlements. The moment-rotation relationship with an initial factor of safety against bearing failure of less than 2.0 is shown in Figure 4. The footing maintains contact over its entire width so long as the eccentricity lies within the middle third of the footing. As the eccentricity increases, the moment,


1 is illt some Conta limit uplifl rotati

. Appe

J

1__-----------------------------------

Ii


-

-- 1----------------------------------------------MF. at which uplift is initiated may be calculated

and the elastic limit is

as

(JE=(JF(2(;.-I)r

VB

M F =-

6

The corresponding rotation, OF. of the footing with a full contact moment of inertia, 10 , is

-~ 8F6 k, I.

mal If a :rtical Iward

After uplift the contact pressure at the base of the footing is as shown in Figure 5. As rotation continues the elastic limit of the soil is eventually reached in shown in Figure 6. It can be shown that the rotation of the building, at this point, is as follows F

2

(JE = (JF7 mtact I ons.

Dntact

It can be shown that if the factor of safety is less than 2, then the elastic limit is less than the full contact limit and yield will occur before uplift. For the case where the factor of safety is exactly 2, yielding and uplift occur simultaneously. The full plastic capacity of the footing is illustrated in Figure 7. The expression for the plastic moment capacity, Mp , is M = V

ngth maIly me ils the trol I

of .ng Ig as f the ,ment,

ffects

p

B(I_J...)

2

F.

The situation for a factor of safety less than 2 is illustrated in Figure 8. Where the rotation is sO\llewhere between the elastic limit and the full contact limit. If rotation continues, a full contact limit will be reached and the footing will begin to uplift. It can be shown that full contact limit on rotation is

(J - q.

F.

r 2k,B (F.-I)

Figure 9 summarizes graphically the basic expressions for overturning moment and rotation for various values of initial factor of safety. These values have been normalized to the full contact limit state. Table 1 illustrates the change in stiffness for a footing as it rocks or the soil yields beneath it. The author summarizes three mechanisms for energy storage of dissipation in the model of the rigid foundation rocking on a clay soil. Prior to plastic deformation of the soil, energy may be stored as elastic strain in the soil material (Es ). If the structure uplifts, potential energy is stored in the structure itself (Ep). Energy may also be dissipated by plastic yielding of the soil (Ed ). This is illustrated in Figure 10 for an initial factor of safety of 3. If the factor of safety falls below 2, the center of rotation actually moves down into the soil under the action of gravity and rotations exceed the elastic limit. In this case the structure is doing work and this must be added to the rotational energy input to give the total energy. This condition is illustrated in Figure 11 for an initial factor of safety of 1.5. In general, it can be seen from this development of energy conservation that soils with low factors of safety generally will dissipate a lot of energy in plastic deformation of the soil. Alternatively, for soils or foundations . with high initial factors of safety the energy dissipated in the soil will be relatively smaIl and the potential energy transferred to the structure by uplift will be relatively high. Repeated Loadings on Winkler Models. The author uses the basic relationship developed between moment and rotation to study the effects of repeated and reversed rotations on the modified Winkler model. The exact solutions, although simple to express, are highly path-dependent and amplitude-dependent. This is due to the fact that inelastic displacements build up as the foundation

"ppendlx F. Supplemental Information on Foundation Effects

F-1S


rocks back and forth as illustrated in Figure 12. Consequently the author developed a computerized solution to generate moment-rotation relationships for repeated, reversed loading. The results are illustrated in Figure 13 for a factor of safety of 3. Finite Element Model. Using the basic theory developed previously, the author formulates a finite element approach to the solution of momentrotation relationships for spread footings. This allows for the study of rocking at arbitrary amplitudes of plastic behavior beneath the footing. The basic relationships for the finite element soil components are shown Figure 14. These are used in a model that is schematically shown in Figure 15. The results of the final formulation as shown in Table 2 agree well with the basic theory. Experimental Studies. In order to verify the theoretical development, the author conducted a series of relatively large-scale laboratory tests on a model of a spread footing rocking on a clay subgrade. The equipment involved the following: a. The foundation medium consisting of remolded clay enclosed in a rigid steel-walled container. b. A rigid steel footing restrained in a horizontal direction by tie rods. c. A system applying constant vertical load to the footing. d. An independent system applying dynamic rocking displacements to the footing. e. A system of instrumentation to measure displacements and pressures. The design and use of the test system is described in detail in the thesis. Two types of .footings were investigated as shown in Figure 16. Type B examines rocking about the weak axis and Type A about the strong axis of the footing. The results of these experimental studies confirmed the general qualitative theoretical result. The behavior of the soil material for various initial factors of safety was as predicted by the theoretical work. A quantitative comparison of test results versus theory is illustrated in Figure 17. It can be seen that the experimental forces generally fall below that predicted by theory. This is due to the

F-'4

fact that the soil stiffness and subsequent strength upon reloading is amplitude-dependent in reality. Subsequent experimental and theoretical work allowing amplitude variation in the soil stiffness parameters demonstrates that the experimental and theoretical results converge. Conclusions and Implications. For static loads, a moment-rotation relationship for a rigid footing on clay soils is linear only while the footing remains in full contact with the soil in its elastic state. Rotations in excess of the full contact limit or in excess of the elastic limit cause a softening or reduction of the stiffness of the footing. The parameter determining which limit is reached first is the initial factor of safety against bearing failure. If this is greater than 2, the footing separates from the soil before yield occurs on the compression side. If the factor of safety is exactly 2, uplift and yield occurs simultaneously. If the factor of safety is less than 2 the soil yields plastically before uplift can occur. Systems with factors of safety less than 2 tend to be energy dissipative as the center of gravity actually moves down into the soil. The moment rotation behavior for repeated reverse rotations is similarly dependent on initial factor of safety but is additionally strongly amplitude-dependent. Solutions for steady state rocking and a factor of safety greater than or equal to 3 generally show four types of characteristic behavior depending on amplitude as follows: a. If the amplitude is less than the full contact limit for the footing, behavior is generally linearly elastic . b. For amplitudes greater than the full contact limit but less than the elastic limit, the behavior is elastic but nonlinear when the amplitude exceeds the full contact condition. c. For amplitudes greater than the elastic limit but less than a critical value, limited soil yielding at the footing edges effectively reduces the amplitude range over which the footing rocks in full contact until a modified full contact limit is exceeded. Thereafter the behavior is nonlinear elastic.


d. I ~

r

e r

s l

t I

1 solut exce: case mom the ~

- 1-----------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

:th

y. s and

d

.ts tact

It is

st :urs is ly. Ids th

d. If the amplitude exceeds a critical value, plasticity effects are such that the footing never regains full contact, resulting in nonlinear elastic behavior over the entire amplitude range with a permanent reduction in rotational stiffness. This can be visualized as the soil underneath the footing becomes rounded-off at the edges thereby offering less overturning resistance. The factors of safety less than 3 computer solutions for variable rocking amplitudes show excellent agreement with theoretical results. In the case of a factor of safety of 1.5, the steady state moment-rotation relationship was shown to take the form of a stable hysteresis loop with plastic

yielding and continuous energy dissipation for all rocking amplitudes in excess of the elastic limit. The implications of the foundation rocking inferred from this research including the effect of partial separation and soil yielding indicate that forces will generally be reduced in the structure at the expense of larger total displacements. This researcher concentrated on clay soil for which the ultimate bearing strength is not dependent on the width of the footing. On future studies for granular soil materials where the ultimate strength is dependent upon the width of the footing are warranted. AdditionaIly, the subject research concentrated on surface footings. Future study should evaluate the effect of foundation embedment.

Table. 1. Reduction In stiFFness With Racking Amplitude

Footing Rotation

Factor of Safety

Secant Stiffness

0/0

Fv

k/k

F

ves

Tangential Stiffness ,

, Reduction

k/k' ro

ro

1

1.5 2 3

0.74 1.0 1.0

26 0 0

0.32 1 1

68 0 0

0.42 0.68 0.79

58 32

O. OS'

2

1.5 2 3

0.13 0.35

95 87 65

0.29 0.48 0.60

71

0.01 0.03 0.13

99 97 87

ial

:qual

\

Reduction

t 3

:t

1.-5 2 3

.

21

52 40

Table 2. Comparison oF Moments Computed Numerically With EXact Values

In.

lit 0/

OF

Ie

THEORY

ed he

FINSTRIPS \ diff.

1 104.2 104.0 -0.19

2 165.2 165.1 -.06

4 197.3 197.4 .05

Appendix F, supplemental InFormation on Foundation efFects

'fectS

6 203.5 203.0 -0.25

8 205.6 205.2 -0.19

10 206.6 207.0 0.19

F-15


contact

pressure

q

I

e"T ",,,,"

qu

- - - -

2

-._---1-.,;..-.;

ks

D~------------

________

soil deformation

Figure 1. Angular DIsplacement oF RigId CIrcular Plate (after Weissmann & White, 1961)

Y

Figure 2. Assumed statIc Sol/Behavior

v

~ T 11811111

1:'""

.-

1

,,,"

o~,"

,,

,;

"'~

//'i!i.1&/1

qo

1'

8/ 2

B/2

·1 '

·1

FIgure S. InitIal contact Pressures

F-'&

Appendix F. supplemental Information on Foundation EffectS

1 ,


[a]

---J,I

-----e

I' [b]

0'.

x'

.... -_ .... --- .. _---

v1

Yc 0

B

P

B

2

2

I

T

Ja

x

I

y

Figure 4. faJ Contact pressure,. fbJ Geometry

[a) B

I'

I'

B' 0

:I ksBO

Flflllre 5. faJ contact Pressure,. fbJ Geometry In Uplift case

.ffects



+.

[OJ

'-

,i j

I I

0

;

~ qu

B'

[ bJ

-9

f

B"

~~ ~: ----T-m-----------f~-iiilliQ,.,,--c

I

..

,-I

a+s"

'~

Figure 6. raj CDntact pre••ure; rbJ Geometry fDr Yield ca.e

I.

B"

,I

Figure 7. Fully Pla6tlc ca.e

F-1S


j

API

, 1--------------------------------__ __ SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

e' Figure 8, Full Contact

,

2

Mp/

~E_

1----

- - /-

l/

-'-~- I-

Y

M7 it-1l .1!!

oE

-

--

Fy

= "5

Fy = 2

/J

~~ if

~

o

Z

!J

II o

. , 2 Normalised

4

3 Rotation

5

~

OF

Figure 9. static Moment'Rotatlon RelatlonIlhlp lor a Rigid Rectangular FOoting

EffectS


F·19


2

!! ! ~

~

.x

Energy stored as

M

•e •

in

P. E.

I,,

struc.ture

...

,l,

~E

..'" ~

<:

w

a Normalised

Rotation

Figure 10. Energy Balance, F. = S

/

/

~

.x

'I'0

1/ Energy

2 Work done b

struc.ture

y

•

...

..'" ~

<: W

/

/'

V

.

/

.. .-

~'\ ~stored

~

I

sipated in

soil

".

.-

";.-

'\

.- .-

".

,'\' ·I~n:r~~ so 2

Normalised

".

in

a

J

Rotat ions

Figure 11. Energy //aIance, F.

F-20

/,

=1.5

Appendix F. Supplemental Information on Foundation EffectS t

I

1------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

Figure 12_ Yield Zone Geometry (end of sr" quarter cycle) 2

1·6

1-2

;/ / ,'/ II

---- ~,'L

f\_.. _.VT!\V2T!\\

3 .;"" I

I

1

~3 if

-9.

INPUT

ROCKING

2. I , 18~ '1.

11.

0-8

...z

0'4

--- -

1st

--

2nd

cycle +

I,

subseqUE'nt

';III

X

o

X

12 1 11

o o

-..

~

UI Vl

. .J

~

'

,, '

cycles

UI

3

-0'4

, f' ,

I

t

~61

,,

.j/('

o

z

.

, I

It

-0· 8

lOJ ~lR" /' 17

"

I,

10

-1' 2

-1' 6

-2 -8

VI

q

17:, '7

V. l7~

9~ ~----' -6

-2

o

'NORMAUSEO

2 ROTATION

4

6

8

9/

9F

Figure 1S. NOrmalized Moment-Rotatlon Relationships For Repeated Reversed Loadings with F. = S

FfeetS


F-21


PIJI

. ..". .. ~

qu

~

";;

2

c

3

0

u

....

C E ti

4

qu

)'k

YSIJ I

5

.element

soil

displacement

Figure 14. Element soli Behavior

x
2 3

- ... __ ._._---- .. --...

J •••••••••• N

.

"x· 1 .....

il 0 ~

I

I

,

I

I

.1 Figure 1S. Division Into Finite strips

F-22


FIIIU

1 __--______________________________________ SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINCS

500 mm

E E 0

!ocking

/'

~I

axis

rocki~g

-

Ul

'I

250 mm

axis

E E 0 0

'"

N

TYPE

B

TYPE

A

Figure 16. Foundation lor Test specimens MOMENT! N - m)

vMphheo y)

2000

_ _ _

J

- --;< -~~ ;.- -

--2j V ~

~

1000

",

---

-- -

experime 101 envelop e

thpory

i

~ ~ 1

o

0·01

0·02

ROTATION I radians)

FIgure 17. Comparison DI ExperImental and tehoretlcal Moment-Rotational BehavIor lor Type A FDotlng, Fv= S

ffeets

Appendix F. Sup(llementallnformatlon on FOundation Effects

F·2S


(0)

Resource Summary 2 Chopra, A. K., and Vim, S. C-S Simplified Earthquake Analysis of Structures with Foundation Uplift American Society of Civil Engineers, Journal of Structural Engineering, Vol. 111, No.4, April, 1985. pp. 906-930 In this work, analysis procedures for analyzing single degree of freedom elastic systems considering soil flexibility and foundation uplift are developed. The system considered is shown in Figure I. It consists of a single degree of freedom

Fig. Initi

both cases were obtained from the rocking stiffness and damping coefficients for a rigid massless footing on elastic half-space. Time-history analyses were performed for the

Bel>

Und

~=~

......... RIGID ~'ION

- - - TWO-EL'MlMT fOUNDATION _ _ _ ,NII..I:III FOUNDATION

allo' base stm vibr secc peri to h sma

_ . _ . - I:OUtY.f.LlHTYWO-ll.I:MrNt ,.

fOUNDATION

"c·ll'll1b .............; : , , - ; - - - - - -

/

~b/./il-......,~~.-.-.-.-.-.

•

,.,

'"

~

r

,.,

Figure 1. Slmclure Supported on Three Different Foundation Soil Idealizations: (a) Rigid Foundation; (b) Two-Element Foundation; (e) Winkler Foundation

structure bonded to a rigid massless foundation mat which, in turn, is supported on the foundation soil. It is assumed that slippage between the foundation mat and the supporting soil is not possible. The equations of motion for this system considering uplift were first developed for the case in which the foundation soil is considered to be infinitely rigid. The formulation of the equations of motions was then extended to the case with flexible foundation soils. As shown in Figure I, the flexibility of the foundation soil was represented by either a two spring-dashpot system or the distributed Winkler spring-dashpot system. The values for the spring and dashpot constants for

-----....,{,/.................

"

Figure 2. Moment-Rotation Relations for Unbonded Foundation Mat Supported on Different Foundation Soil Idealizations

north-south component of 1940 El Centro ground motion. The results of the analyses were presented in the form of base shear response spectra. Figure 2 shows the results of the analyses for the case of rigid soils for slenderness ratios (height of the structure divided by the half width) ranging from 5 to 20. Also shown in this figure is the curve for the case when uplift is prevented and the critical base shear values below which there is no uplifting of the structure. From this figure it can be observed that there is a significant reduction in the earthquake base shear when the structure is

Appendix F. supplemental Information on FOundation effectS

App,

j

-


la)

UNDAMPED STRUCTURE

(b) DAMPED STRUCTURE, (_005

••••••• 8OHDlOCONTACT, UPl..I" PR(VfJnlD _ UJrIaoHD(D CONT4CT.U~IFT PEJllIII'TTlO

,

,

, .,

I! )L-'-'-_l--o---'--':-~' '0 5 i,

L.....L---'---'_-'---'-'

0

2

,"

1

4

$

Figure 3. Response of Structures (h/b=lO) to

! Initial Velocity for Two Conditions o!Conwct the

...

Between Foundation Mat and Supporting Soil: (a) Undamped Structure; (b) Damped Structure, g=O.OS

.. GAYlON !l.EIItIN'

-.

allowed to uplift. The beneficial reduction in the base shear due to uplift is more pronounced for structures with higher slenderness ratios and for vibration periods between about 0.3 to 2.0 seconds. However, it was noted that for very short periods of vibration, the foundation uplift can lead to higher base shear demand in structures with smaller slenderness ratio. The base shear response spectra for the case

with flexible foundation soil were then developed. Figure 3 shows these spectra curves for the slenderness ratio of 10 for the Winkler foundation with two different foundation soil properties. Comparison of results presented in Figures 2 and 3 shows that consideration of foundation flexibility leads to further reductions in the earthquake base shears. The authors also developed simplified approximate expressions for estimating the earthquake base shear of the uplifting structures. In these approximate expressions, the maximum base shear is related to the system parameters and the peak spectral acceleration for a corresponding fixed base structure and does not require time history solution of the equations of motion. The base shear spectra obtained from this simplified approach are plotted in Figure 2 for the case of rigid soil and in Figure 3 for flexible soils showing a reasonably good agreement with the exact solution. These simplified analysis procedures can, therefore, provide estimates of maximum base shear and deformations to a reasonable degree of accuracy for practical structural design and can also be used to quickly perform parametric studies of the effect of foundation flexibility and uplift on seismic response.

~ded

tion

,und ented gure e of rom 5 'or cal lifting in the

Hects


F-25


Resource Summary! I

Gazetas, G.

Foundation Vibrations Foundation Engineering Handbook, Fang, H.Y., Editor, VanNostrand-Reinhold, New York, 1991,40 pgs.

o

Introduction. The focus of this chapter in a foundations engineering handbook is on the dynamic aspect of machine vibrations as they relate to foundations. However, most of the relationships that are presented are useful to determine the properties of building foundations for use in analysis of seismic loads. The chapter focuses on stiffness and damping properties for the analysis of rigid plate vibrating on an elastic halfspace. Some information also is provided for deep foundations. Effective Soil Properties. The chapter provides basic relationships for soil stiffness and damping properties. The shear modulus, G, and the mass density, p, of the soil are related to the shear wave velocity, Vs, as follows

v = {Q

, VI'

As illustrated in Figure I, the shear modulus, G , and the damping factor, f3 , are strain

F-Z&

dependent. The chapter presents a number of empirical relationships for calculating the initial shearing modulus for different types of soil material for a low strains. Also included are field and laboratory test procedures for determining shear moduli. Theoretical Development of Vibration of Equations. The chapter presents basic theoretical equations of motion for oscillations of foundation. These include vertical oscillation and generalization to all modes of oscillation, including coupled rocking. Presentation of Data and Accompanying Example. Most of the chapter is devoted to a series of very useful tables providing static and dynamic stiffness and damping properties for various types of foundations and soil conditions. Figure 2 shows several examples of the conditions for which tables are included. Table I is an example of the type of tables that are included in the document. There are also a number of illustrative examples.

T

ON,

'"


App

j


Table 1. EXample oF Table oF Sol/-FOundation PropertIes Dynllmlr: StiNne" /( - K '*(w) St.,1c Stiffness K

Dynamic Srit/ness Genual Shape Vibrlltion Mode Vertical, 1

(found8lion-,oil cont&c1 surfac. is of area A.

andhasacircumscribedrectangl.2L bV2B;L > Bl·

2GL

K, _ - - (0.73 + 1.64Xo.715) 1 -,

wilhX ..

1

:ld

HOffronl.l, y

L-B

X;", 4,54GB 1 -,

~·

2GL

K, _ 9GB

2-,

2-,

lal8f.1 direction)

cal on.

Horizontal, x (in the longitudinal direction)

Rocking. rx (around longitudinal x allis)

Rocking.", (.round 181er.1 allis)

°"·0,,2)'

It,

zo

k.(i. v;.o) .

is ploned in Graph a

4L'

K,_-- (2 + 2.50x088)

(in the

Co.fficiMl k (Gan.,., IlMfM:

Squa,.

k,=k,(i;1I0)

B)

G (L)"'( 2.4 +0.5-B) L

K. ___ /g1 5 _ 1 -¥' 8 wilh

X._K,

1t.:lI: 1

K,... 3.6GB'

Ie,. :. 1 - 0.20'0

1 -,

G [ (')'''] B

s. in

feets

Torsional

K, ..

~_n[ 4 + 11( 1 _~)10]

~,-t,(LI8.Y;lIo)

i. plotted in Graph c

C, ... C,gIB:llol is ploned in Graph d

pV.A,

C,• .. (pVL.IbA-) ·l,. ~,. - l,.(L/8; .0)

is plotted in Graphs e and f

/(,., _ K,.

C,., .. (pVL./~) 'l,y

{ ,<0.45

fr,.,:. 1 - 0.30s o

l,y"" C,.,(L/8; so)

v::: 0.50:

is plotted in Gr.ph g

(LFlO

fr" ::: 1 - 0.25s o j

ions

C, = (pVt.A.) . t,

ell::llt

'1J,,(/6y ) .reamoment of inartia ofthefoundation-soil contact surface around th, Mer) axis

71 K'Y""'.,..-:-;fh 3

(Gene,.' Shllpes)

C, - (pV,A e ) . t,

is ploned in Graph b

02 ( 1-K,,=K,----GL 0.75-v L

Radilltion Ollshpot Coefficient C

K,"" S.3GBl

with J. "" flu + I. v baing the pol., moment of the soil-foundation contact lurflCe

Ie,:. 1 - 0.14so

C, ... (pV.Jb ) 'l,

l,,,,,l,(L/B;.o)

Not. that as LIB ..... «I (stnp footing) the tn-.tiall valua of K. and K,_ 0: the "aluu comPllt«lllOm the two given lo,mulll COtrHpol"ld 10. footing Wllh LIB a. 20. '.o-wIIIV,.


F-27


l6.

T

I.

I,

G
I I I I I I

y,

Y

I

,

'0"

•

r-;:

10-"

10-"

y, 10·a

,

10. 2

0.'

D.S

"

"

.,,'

10"

0.01

0.'

20

So Pll'Wlt 10

,

"

"AVlf1Jge. CUr'll for ~ GRAVELLY soill ,

0.001

~

.,,,

,.,.."

'-......::, ""'<,

SANOS

CLAYS

0.001

10-"

10.1

0.01

"" 0.'

20

• 10 SANDSANQ

GRAVELS

0

0.001

0.01

Yc ,percent

0.1

0

0.001

0·01 Yc' percent

, 0.1

Figure 1. Effect of Nonlinear stress-Strain on Shear Modulus, C, and Damping, p

F-28

Appendix F, Supplemental Information on Foundation effectS

App

.J


T

T

1

Pion

"

1

L>'

I ~ ~mOII'.'1

,i,ill

'"und.liO~

t?>??2??C"

...... ,'

2

'"

..":'

-:,:.'

. .-..... : ..

..

..

', '""

.... ;....

',Homogl"eou. "'HolI.paco; ,,' G • Y • P

".",::;

'.,"

,'-.

.

", '

..

... G •

",.

I

3

,

,

.

....

4

,.---.... ';.,'

,1

H

~ ~ ~-~: ~ ~-~ '..: ~ .' '5' ~';'~'I: ~'~ ',:. G. v • P

M"i..L,,,

-

rigid

•

.,',

-"

-: :' ..

"

:..

-,,'

.

fO,,"Olion

• --------~--.·r1,~.--,,-o\~~~,~,~,-

. ".~ :.

I I·

in"omog~nlO~~ HaLllpo~.

. Gell

: F.

:

11\I'Iomog.notclin

0, Homoge",ou. ,: Oepolit,'

-I i: ,.I 1..

'I: .

'\"

.\11

E

E•

fi;" -,'

I'"

-: j ~;:

II' L•

•

._~S~::>

Figure 2, Soil and Foundation systems for Various Tables of properties

Effects


F·29


Resource Summary 4 Housner, G.W. The Behavior of Inverted Pendulum Structures During Earthquakes Bulletin of the Seismological Society of America, Vol. 53, No.2,

February, 1963 Introduction. This is a classic paper on the theoretical behavior of rocking structures. During the Chilean Earthquake in May of 1960, several water towers resembling inverted pendulums rocked about their foundations. Other water towers that appeared likely to rock about their foundations were similarly damaged. Evidence of rocking of apparently unstable structures was also observed during the Arvin-Tehachapi Earthquake in California in 1952. In an effort to explain this behavior, Housner develops theoretical equations describing the rocking of blocks during earthquakes. The basic model that he used is shown in Figure 1. Free Vibrations. Housner reduces the equation of motion for the block shown in Figure 1 to the expression 9=a-(a-90)coshpt In this equation the parameter p is

p=~WR

P

[

1

--e; 1 I--

a This equation is shown in Figure 2. In general it can be seen that the period lengthens with increasing angle of rotation and shortens to near zero as the initial angle of rotation diminishes. Housner goes on to calculate the dissipation of energy which occurs when the block impacts its base during each half-cycle of rotation. For this development he assumes that all of the reaction to the impact is inelastic which means that there is no "bouncing" due to elastic response of the block or its foundation. With this assumption, the reduction in energy for each impact may be expressed as r=

10 where 10 is the rotational moment of inertia about point O. This expression represents the motion of a block if it is raised or rotated to an initial angle, 90. From this expression, the period of vibration as a function of the initial angle of rotation is as follows

F-IO

I

T=-cos 4 h-

(

mR2 ) l-lo(l-COS2a)

The effect of this reduction is shown in Figure 3. The parameter t/I on Figure 3 represents the ratio of the initial angle of rotation, 90, to the characteristic angle of the block itself, cc.. From Figure 3 it can be seen that for large amplitudes of rocking, the energy of vibration decreases rapidly but for smaller amplitudes, the energy dissipates very slowly. When including the effects of energy dissipation the expressions for the period of vibration is modified to

APpendIx F. SUPPlemental Information on Foundation EffectS

Hou ofa of ti exru ove) acce osci com

whil gra' requ the I

for 1 com due dura grou The) over acce Hou desc acce max the i two

1 SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

Overturniug by Constant Acceleration. Housner develops equations to describe the effects of a constant acceleration, a, lasting for a period of time, t I, acting at the center of mass of the example block. The block mayor may not overturn depending on the magnitude of the acceleration and its duration. For small angles of oscillation it can be shown that the necessary conditions for motion to be initiated is that

at g> a which specifies the fraction of the acceleration of gravity required to begin tilting of the block. The required acceleration in time required to overturn the block is expressed in the equation leral ar

~WR

cosh-t,=l+ 10

(

2~ ~-l ga

mof ts is n to is no k or ction s

igure

om les of lidly ltes lergy

ffeets

1

)

ga

This relationship is illustrated in Figure 4. Housner carefully points out that the analysis for a constant acceleration pulse is not a realistic comparison for earthquake ground motion. This is due to the fact that constant acceleration of finite duration followed by a constant velocity of the ground does not occur during an earthquake. Therefore, it is not meaningful to discuss overturning of blocks in terms of a percent gravity acceleration. Overturning By Sinusoidal Acceleration. Housner continues to develop a theoretical description of the rocking block to a sinusoidal acceleration load. If the variable a represents the maximum acceleration and (i) is the frequency of the sine wave, then the relationship between the two parameters required to overturn the block is

Another version of this equation is:

~= 1 +~(21r)2 WR To

ga

In the second expression, To is the period of the ground acceleration. These relationships are illustrated in Figure 5. Overturning by Earthquake Motion. Overturning can be caused by successive smaller pulses at certain frequencies that may occur during actual earthquakes. Housner illustrates this effect by developing conservation of energy and momentum equations describing the response of a rocking block to earthquake motion. As a result he develops an expression relating the geometry of the block to the spectral velocity which would result in a 50% chance of the block overturning. This expression is:

a- S ,

-.fiR

~MR2

T

For a relatively slender structures, this equation reduces to: S

a=-'-

.fiR

This equation may be interpreted as stating that for a given structural velocity, Sv , a block forming an aspect angle a will have an approximately 50% chance of being overturned. The significance of results of this development is that the stability of the block is dependent on its size, as expressed by the factor R. Thus, when comparing two blocks of the same relative proportions, the larger block is more stable than the smaller. This is explained by the fact that the ground motion is the same for both and the effects of the mass of the block in providing stability is greater for the larger block. This effect is summarized in terms of the half-width of the block by the following three equations


F-II


Earthquake motion:

b=Svl Square pulse:

Sine pulse: b- a.-,==.;;h== g

l+;(~J

It can be seen from these equations that the half-width of the block required for stability does not increase linearly with its height. The general

conclusion is that tall, slender structures may be more stable than might be supposed intuitively. Housner goes on to study the inverted pendulum water tank structures from the Chilean earthquake. He does some comparative calculations which verify that such structures could have rocked about their foundations during a strong ground shaking. Summary. The key conclusion of this work is that it is misleading to infer stability from considerations of a constant horizontal force acting on a rigid block. In fact, taller slender structures may be more stable than such considerations would imply. It should be noted that this theoretical work was extended with an experimental study by other researchers. This is summarized in Research Summary No. 13.

cosh p,T = 1 1 .. - 8,la

5 4

b

j

.... e~

I

b

II

3

I~WR 4 I..

J

/

2

cg ~

h,

V

IW O· Figure 1. A Rocking BlOCk

F·:S2

V /'"

eX..

0.2

V

0.4

V

V

J

V

0.6

0.8

1.0

0 Figure 2. period T 01 a Block Rocking With Amplitude 60


App

-


Je

'an

1.0

;ould 0.8

rk is

\

0.6

:ting

r=0.7

\

~I.O ~ "N"" ~ 0.2 0.4

ies

...............

is

~

-- --

2

r=::::::

3

4

5

6

7

8

NUMBER OF IMPACTS Figure S. Amplitude
5

4

3

t ffiif

1'-':

2

\ \

1.0

i

'"

1.2

........ ............ 1.4

1.6

1.8

2.0

~

go<

Figure 4. Con$tant Acceleration of "a" Of Duration t, Required for overturning

'feets I Appendix F. supplemental Information on Foundation Effects

---------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

R

2.2

V

2.0 /'

1.8

"'-'"

/'

1.4 1.2

-

/"

V

,,

V

02 0.4 0.6 0.8 1.0 1.2

~1.i~

'i

wer infh thei eval anal upJi

~

o

I

.~;I

,/

g~ 1.6

LO

V

1.4

1.6 1.8 2.0

seal FIgure 5. SInusoIdal AcceleratIon pulse "a" sIn all RequIred for OverturnIng

mOl

Uni· Eng the I base of II was the botl1 whil pert upli: buill inve fixel eom pro~

selel with end! mod tensi direl neol eolu

Appendix F, supplemental Information on Foundation EffectS

ApPI j

-


Resource Summary 5 Hucklebridge, A. A. and Clough, R. W.

Seismic Response of Uplifting Building Frame American Society of Civil Engineers, Journal of the Structural Division, Vol. 104, No. ST8, August 1978. pp. 1211-1229 The primary objectives of this investigation were to make an experimental assessment of the influence of allowing the columns to uplift from their foundations during an earthquake and to evaluate whether commonly available structural analyses programs can accurately predict this uplifting nonlinear response. Shake table tests on an approximately 1I3rd scale model representing a 9-story, 3-bay steel moment frame building were performed at the University of California, Berkeley's Earthquake Engineering Research Center. Figures 1 and 2 show the key features of the test model and the uplifting base detail. The slenderness ratio, defined as ratio of the height of the structure and the half width, was 3.1. The test model structure was subjected to the 1971 Pacoima Dam S74W earthquake motion in both the fixed base condition and the condition in which the columns were allowed to uplift. This is perhaps the only work where the effect of allowing uplift of the columns on seismic response of building structures has been experimentally investigated. The response of the test structure, for both fixed base and uplifting condition, was also computed analytically using the nonlinear analysis program DRAIN-2D. In these computer analyses, selected beam and column elements were modeled with concentrated bilinear plastic hinges at member ends. The uplift at the base of the columns was modeled with nonlinear truss elements having zero tensile force capacity and stiffness in the upward direction and having compressive stiffness of the neoprene impact pads at the base of test structure . columns in the downward direction. These uplift

elements were assumed to be rigid in the fixed base computer model. The key observations of this work are summarized below: • Allowing uplift generally had a beneficial effect on the seismic response of the test structure. • Test structure exhibited a very low damping of about 0.7 percent in uplifting/rocking mode of response as compared to a damping of about 3.2 percent for fixed base response. • Fundamental period of the structure softened from about 0.5 sec. for fixed base response to about 0.76 sec. uplifting/rocking response. • Vertical earthquake input had little influence on the building response. • A tangent stiffness proportional damping matrix predicted the response of uplifting structure very well. • Analytical results from the DRAIN-2D nonlinear analyses predicted the measured response very well. Based on these observations, the authors concluded that allowing transient uplift during an earthquake does not imply imminent toppling of a practical building structure even during a maximum credible earthquake, and, in fact, allowing uplift can lead to considerable reduction in earthquake loading or ductility demand or both. The authors also made the case for incorporating seismic uplift of structure in a rational design to achieve enhanced ductility and earthquake performance at reduced costs, especially for reinforced concrete structures.

ffectS \, Appendix F, supplemental Information on Foundation Effects

F-35


R

__ K -

•

"

,

r7""1 "

...

r-;r-,

,.

.u

....

,

~

....

..,...,

,

.-;;-

-

...... I.'

~

Figure 1. Test MDdel Schematic

.

I •••• 1

..

.....

,

,

pap' abul

::

,

,

J:"r'"

.-

-

~

....,. ,

~

r=-"'

fI'"'

i

acc( ofs spre The calil engi

"

~

•

:"7

,

~

~

r--,:---'.

"

.-

~

,

"I

.J 'r.

-

~,

I

.

//.0:~"-

.....:,,~~y

--1'

9 STORY STEEL TEST FRA"E

that seisl dire anal brid beel behi bacl expi pres abUl appl

valu gem are I

-.·.11 ,

use on a

coe! hori roc~

Figure 2. Upllflng CDlumn Base Detail

F-S.

Appendix F, Supplemental Information on Foundation effectS

func of tl emb

l SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

Resource Summary 6 Lam P.I., Martin G.R., and Imbsen R. Modeling Bridge Foundations for Seismic Design and Retrofi~ting . ., Transportation Research Record 1290, Proceedings of the ThIrd Bndge Engmeenng Conference, Denver, Colorado, March, 1991. This paper presents simplified procedures with accompanying design charts for the development of stiffness coefficients for abutments, piles and spread footing foundations for highway bridges. The procedures presented in the paper were calibrated to design practice adopted by bridge engineers. Several examples are presented in the paper to illustrate various sensitivity issues in abutment and foundation design. In its discussion about abutments, it is stated that the abutments attract a large portion of the seismic force, especially in the longitudinal direction. The problem is not an easy one to analyze because of the uniqueness of individual bridges and the soil conditions at each site. It has been recognized that there is highly nonlinear behavior in abutments due to failure of the backfills and from structural nonlinearity at expansion joints. An iterative procedure is presented in the paper to determine the appropriate abutment stiffness in a linear dynamic response approximation of a very nonlinear system. Typical values of initial stiffness used by Cal trans and a generalized equation to estimate abutment stiffness are presented. The analysis of spread footings involves the use of stiffness equations for a rigid footing sitting on a semi-infinite elastic half space. The stiffness coefficients are presented for vertical and horizontal translation and for the torsional and rocking rotations. The stiffness coefficients are functions of the shear modulus and Poisson's ratio of the soil, the size, shape, and depth of embedment of the foundation. The stiffnesses of

feets

an embedded regular footing can be found from the equations given below:

Vertical Translation Horizontal Translation Torsional Rotation Rockl ng Rotation

IX~8GR/(2'v)

IX~16G R'/3

IX ~ 8 G R' I 3 (1 • v)

where G and v are shear modulus and Poisson's ratio for an elastic half-space material; R is the equivalent radius as shown in Figure 1. Figures 2 and 3 present the IX and ~ factors, which represent the embedment and shape correction factors, respectively. The stiffness of the rectangular footing can then be found by multiplying the stiffness found from the table above by IX and ~. Pile foundations are the most commonly used foundation system for support of bridge structures. The equivalent coupled foundations stiffness matrix model is the most general method of representing the foundation stiffness. The paper outlines the steps to determine this type of foundation stiffness. The paper also discusses the variations in some parameters that can cause the results to very sensitive to those assumptions. Considerable engineering judgment must be applied to estimate the stiffness coefficients for a pile foundation. The paper gives a broad overview of the assumptions and variations in foundations for bridges, but there is still application to foundations of buildings.


F-:n


.' I

•

J "

I

I

lL

EQJIVAUHT RADIUS: TRANSLATIONAl.:

ROTATIONAl.:

R _

r alItZl.)l ]"-,._ _ _ _ _ b.-AXIS ROCKINGI t:

3"

.

Figure 1. PrDcedure for Determining Equivalent Radius Df a Rectangular Footing

p-:sa


App


I.

1 c·

2

~ ~

W L

'" iii

LJII

Figure 2. Shape Factors For Rectangular Footings, ex

...0.0

0.1

0.2

0.3

O.Y

0.5

D/R Flgure:S. Embedment Factors For Footings, fJ

feet5 . Appendix F, supplemental Information on Foundation Effects

F·SS


Resource summary ,

-

Dynamic Response Verification of Simplified Bridge-Foundation Model Journal of Geotechnical Engineering, ASCE, Vol. 115, No.2, pp. 246-260, February, 1989

F-40

I t-.·'.:'~i

Levine M. B. and Scott R. F.

This paper presents a simplified method to evaluate the rotational stiffness' of bridge foundations, and the application of the method to calculate the dynamic response frequencies of the Meloland Road overpass (MRO) during the 1979 Imperial Valley earthquake of magnitude 6.4. A comparison of the calculated and recorded response frequencies of the bridge indicates that simple foundation models such as those presented in the paper, allow modeling of the response of structures with reasonable accuracy for practical engineering purposes. The MRO consists of a continuous reinforced concrete box-girder road deck monolithically supported on open-end diaphragm abutments and a single, reinforced concrete column pier as shown in Figure 1. Each abutment is supported by a single row of seven timber piles. The central column footing is supported by a square grid of 25 timber piles. The soil conditions at the site are relatively uniform and consist of medium stiff to stiff clays to a depth of at least 60 feet. The dynamic soil properties used to develop the foundation models were estimated based on data from a standard site investigation program which did not include direct measurement of dynamic properties. The dynamic response of the bridge was calculated using a simple finite element model consisting of ten 3-D beam elements. The abutment and column foundations were represented by rotational springs along the three main axes of the bridge. Translational springs were not necessary since the bridge was analyzed

ltk~~ r~d'

using the recorded motions at the column footing and abutments as, input motions. The models used to calculate the rotational springs at the column base are shown in Figure 2 while those used for the abutments are shown in Figure 3. In these figures, the x, y, and z axes correspond to the longitudinal, vertical, and transverse directions on the bridge, respectively. To calculate the rotational springs of the column footing the soil-pile foundation was represented by a uniform Winkler foundation with vertical stiffness I<.,y and horizontal stiffness I<.,h. The rotational springs of the abutments were calculated by representing the fill with equivalent spring stiffness' ki, and kiy , and the piles with equivalent vertical springs kay and lateral and torsional springs K" and Ka•. The dynamic spring constants were assumed to be equal to the equivalent static rotational spring constants. The soil spring constants were calculated from the coefficient of subgrade reaction and the modulus of elasticity of the soil which were estimated from in situ standard penetration tests and laboratory unconfined compression tests. The bridge model was used to calculate the response frequencies of the structure for the first transverse and vertical symmetric modes of vibration of the structure. The response frequencies were also calculated using models with fixed and pinned ends at the foundations. The calculated frequencies are compared with the measured frequencies below.


,"O'i,-

~(,-.-~

Tra, ver

-

SVIT

with at tl1 mea calc subs pinr vert mea the 1 mea

App

1


Parametric analyses were performed to evaluate the effects of various assumptions for the rotational springs on the calculated frequencies. These analyses indicate that the transverse mode frequency is affected primarily by the rotational springs at the abutments with respect to the y axis and is relatively insensitive to the springs at the column footing. The vertical mode frequency is affected only by the rotational springs at the abutments with respect to the z axis. The measured frequency for this mode would suggest that the abutments behave as if they were pinned with respect to the z axis. The comparisons between the observed modal frequencies of the MRO and those of the model . with rotational springs proved that, even with simplifying assumptions and a very basic uncomplicated approach and model, it was possible to estimate the dynamic response of the structure to a level of accuracy which is suitable for practical engineering purposes.

.calculated Frequencies (HZ/·

i:1;~t/Plnned . Ends

Transverse vertical Symmetric

Ig

ed r

on

lith

3.72 4.74

Ends

3.34

8.27

3.13

5.12

6.80

4.72

It may be seen that the frequencies calculated with the bridge model using the rotational springs at the foundation are in good agreement with the measured frequencies (within 10 percent). The calculated frequencies assuming fixed ends are substantially higher than those measured. The pinned·end assumption yields a frequency for the vertical mode in excellent agreement with the measured value. The corresponding frequency for the transverse mode is substantially lower than that measured.

!nt

ng

F

10'·1l1.1~)

'08

C~

..~-

10'·'JI.1m)

·III"·..··~)

==j 7"

a•• '2

us rom y

AINI. 3

ELEvATION

r_

TYPICAL

o

,

with

I.

••

Akt.1

rst

"J.'~)

Figure

_ r ; •._ :

1.

SECTION

~

MeIDland Road overpass-General Plan

fects • Appendix F. supplemental Information on Foundation Effects

F·41


L

n

_.':" ....

... ....

I I

ke•

I

ry

7

-, M-J ..... J.8" I

IS' 14.6 m)

(,

keo

... ......

I

I

• ....

•

mar

I

~ 14.6

I.SI

...

role soil: grOl

m)

disTl sum

(b)

Fillure fl. Models Used to calCulate Rotational Sprlnlls at Column BaSe; aJ About x and z Axes; b} About y AXis

ti

~.

3e-,o.I5 m 1

'

.

r-----

8 ' 3.'-------, I1.S 1ft,

I H_l"-

'.)M,

:J' •

K..

11111/1111

'

'01

.

101

J-.

'"

resp soil char The: the! the!

to.&"'1

1

that

are I

,!)/oI,

char sour sour undf coO( at th

influ eart! exiSI

requ estal whel

Fillure S. Models Used to calculate Rotational Sprlnlls at Abutments; fa} About z AXis; fb} About x Axis; rc} About y AXis

resp' influ acce pres. met! and,

F-42


ApPI


Resource Summary 8 Martin, G. R. Geotechnical Aspects of Earthquake Engineering Journal of the Australian Geomechanics Society, Special Issue, 5th ANZ Geomechanics Conference, August 1988 This is a state-of-the-practice overview of the many aspects of geotechnical engineering plays a role in earthquake engineering. The behavior of soils has contributed to major structural damage, ground, embankment and slope failures, and disruption of lifeline facilities and systems. A brief summary of significant recent earthquake events that led to greater understanding of soil behavior are given in the paper. The paper presents a discussion of site response. The significant influence of local site soil conditions on the acceleration and frequency characteristics of the ground motions is discussed. These factors include the earthquake magnitude, the source mechanism of the earthquake including the speed and direction of the rupture, the geologic characteristics of the wave propagation path from source to site, the distance of the site from the source of energy release, the geologic topography underlying the site soils, and the local soil conditions (soil type, stiffness, layering and depth) at the site. A discussion is given about the influence of local soil conditions. Design for earthquakes have a strong reliance on the use of existing strong motion records. Scaling is often required of a sufficient number of records to establish a smoothed spectra for design. However, when suitable re~ords are not available, a site response analysis may be needed to evaluate the influence of the local soil conditions on the ground acceleration and spectral characteristics. The paper presents a discussion of available analytical methods that require modeling of the shear strain and damping characteristics of the soils. A brief

feets

discussion about the developing role of centrifuge testing to understand dynamic soil behavior is also presented. The paper also presents the state-of-thepractice in the understanding and prediction of ground settlement. The paper discusses the application of laboratory testing to prediction and how multi-directional shaking should be accounted for. The paper also has an extensive discussion about ground liquefaction. First the issue of liquefaction on level ground is addressed. The paper discusses the development of laboratory tests to predict liquefaction behavior, but points out many of the problems associated with good quality undisturbed soil samples. The paper goes on to discuss the Seed and Idriss simplified procedure for evaluating field liquefaction potential based on in situ liquefaction strength curves determined from a laboratory test program. The paper also discusses the empirical approach that uses the standard penetration test (SPT) results to predict the potential for liquefaction. The paper also discusses the mechanistic analytical approach to determine the liquefaction potential; this requires an understanding of the progressive pore water increases during cyclic loading of sand. The paper describes the approach and several analytical codes available to perform such an analysis. The paper also discusses earthquake induced settlement of saturated sands. The paper comments that our understanding about post-liquefaction behavior is still far from complete and needs further research. The paper also has a discussion about the response of earth structures, such as earth dams,


F·IIS


embankments and slopes. The paper points out that the importance of dynamic response was not fully recognized until the early 1960s when it was the practice to represent the effects of earthquake loading by an equivalent static seismic coefficient independent of the height and stiffness of the earth structure. A discussion of several analytical methods developed using equivalent linear response are given; these programs are QUAD-4 and LUSH. The paper also discusses embankment and slope stability. This section of the paper first discusses earthquake induced deformations of limited extent. The paper discusses the Newmark simple method which assumes that rigid plastic behavior only occurs when the accelerations exceed a well defined yield acceleration. This method was extended by Makdisi and Seed for earth dam analysis. The paper also describes other methods based on stress path dynamic testing in the laboratory. This section of the paper goes on to discuss liquefaction induced instability. The paper stresses that the prediction of deformation during earthquake loading of saturated cohesionless earth structures is clearly a difficult problem due to the added complexity of time varying changes in effective stress. The problem of determining the undrained residual strength of the soil for the analysis is highlighted. The paper also discusses the topic of retaining structures. With regard to the design and analysis of gravity or cantilever retaining walls, the paper comments that the Mononobe-Okabe pseudo static approach is widely used to compute earth pressures induced by earthquakes. The paper notes that if the peak ground acceleration is used for the lateral coefficient, the size of the retaining structure often becomes excessively great. For a more economical design, a small earthquake

F-U

induced lateral deformation is preferable. The paper describes the Richards and Elms method which is a displacement method that proposes to use an acceleration less than the peak value. A discussion about the effects of liquefaction on retaining structures is presented. The consequences of pore pressure build-up are discussed and possible mitigation strategies are presented. The last topic is pile foundations and the paper discusses the vulnerability of pile foundations to ground liquefaction. Degradation of the lateral soil support stiffness (p-y curves) may occur from either pore pressure increases from the earthquake free field response, or from localized pore water pressure increases in the vicinity of the pile head generated by relative displacements caused by structural inertia loads. If this occurs, either total or partial loss lateral stiffness support may result as shown in Figure 1. Figure 2 illustrates the degradation of lateral stiffness with pore pressure increase for a pile embedded in sand subjected to earthquake loading. Figure 3 illustrates the effects of liquefaction on pile bending moments and deflections. The paper describes how this problem might be analyzed. A discussion of bridge foundations is presented where it is noted that although the first option may be to improve liquefaction resistance through replacement or in situ densification of the soils, piles may be economically designed. Potential for liquefaction exists if the piles are ductile and founded well below the zone of liquefaction. A case history is presented to illustrate this. This paper presents a broad overview of the effects of soils on structures. Although the paper is somewhat dated, it points out the need for research and gives insight into how soil related problems can be analyzed.

APpendix F, Supplemental Information on Foundation Effects

-

.li

r addV

__-------------------------------------------------------------------------SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

1"",1.", MO"""" O"d S~eo. rO'ce, I,om StrUC'VI,

Olutlace"'cnl due to ."',I'oCho"

Q,lalt",

t

Beftd,nq Mo""~"t' dominated by Lolrlol

1"le,octo" LOOd\

_VI> Chi

Pdf 0"D10C'''''"'

o. rll'''' I Tl'o""

t

~I

...,./\ 11.-1 ~V I

I'tte I'"ld

O·'OloC."'~"1

Tu".

ItOft

Be"d'"q Mo"'en', do"".o'td by

Eo' '''ova ~t

~'u I"eld

Olloloetmeft',

O"oIOCt'""ru O.

T.m,

~

I

H."o".,

Figure 1. Soll·Plle InteractIon MechanIsms DUring seismic loadIng


F·45


su qOJ1!d' S! Jdd dq

nor

1.0

Sl,lIness S' r/6 16'0-18,", I

SPdJ

06

°IJ

0

3Jn!

...

U'I

U'I

-... 0

10' r'ee and Head Poles

F

0.6

c

fo'

U'I

F --.

0

a:

Pll(

- 04 02

-

lIn! I1!le p1!: Jdl dl[1ln

'AE( 14(&0

,.;

nos I F'e e Hend Smol16

6

01 J:ld1!1

'll(O .. (AO PIL [

S:lOU;

o~----~----~------~----~----~ 4 '10 o 6 6 2

TIme

01

1

sec

Figure 2. lateral stlHness Degradation WIth pore Water Pressure Increase

F-4&

wdlc

Appendix F, supplemental Information on Foundation EffectS

--


Be ll d' ft 9 Molftelll (_IOrlb .7. 0 ?

Oellecl'oll y 1,"1

3 0

....

30.

.....

,,," •

I ,

I

c

~O

,,

I

'I

40

... ...

,,• , ,,

10·

20

."

~

I

"

... ...

, ,,

... •

• otrr '1f"1 n '""l.aTOy( at .. /"I~I'I. aCllI(" I

.. • , sees

70

T ,~,

•,,

• tn" It-

.~. ~~~, _"'_. n

,

, ,, I , .,,, ,

-.. --1\".,. .. ... c ......

T

.. ,

,.-"

,., •. n 4 , .. '!"I(I> ,
_ " . ·n - - --111\ • •' " .

80

f;

\

c

60·

•• ft

•

'fr. ...... ,

, I

I I

Figure $. Effects of Liquefaction on Pile Bending Moments and Deflections

-


F·4'


Resource summary 9 Martin, G. R. and Lam, I. P. Seismic Design of Pile Foundations: Structural and Geotechnical Issues Third International Conference on Recent Advances in Geotechnical Earthquake Engineering and Soil Dynamics, St. Louis, Missouri, April, 19\ In this state of the art paper design concepts and issues related to the seismic design of piles foundation systems are presented. Design aspects discussed include: I. Questions related to modification of free field earthquake ground motions by pile foundation systems. 2. Methods for determining the stiffness characteristics of pile foundation systems for incorporation in earthquake structural response analysis. 3. Questions related to degradation of lateral stiffness arising from liquefaction. 4. Questions related to potential design concerns arising from seismic overload of foundations causing permanent ground displacements. A brief introduction is provided on approaches used for a rigorous analysis of the dynamic response of soil-pile-structure systems to incoming seismic waves in a fully coupled manner taking into account both kinematic and inertial interaction (Figure 1). It is noted that due to the complexity of nonlinear coupled models. the Winkler model represented by series of independent or uncoupled lateral and axial springs simulating soil-pile interaction. provides the most convenient means of analysis. It is noted that for most pile foundation systems. piles may be assumed to deform in a compatible manner with the free field ground motions and the effects of kinematic interaction in modifying input ground motions to structures can be neglected. In addition. it is noted that due to the relatively low frequency range of earthquake

F-48

inertial loading. stiffness functions for pile foundations are in most cases frequency independent. In discussing analysis methods and design issues. the paper first addresses lateral loading of single piles using the conventional p-y curve approach. The sensitivity to boundary conditions is discussed together with simplified methods for linearizing the nonlinear behavior in terms of a pilehead stiffness matrix. Useful charts for determining the components of pilehead stiffness matrices. such as that shown in Figure 2. are provided. Current thinking on the effects of liquefaction on lateral stiffness is also presented. Methods for analyzing the axial load stiffness characteristics of piles are then discussed. The importance of axial stiffness in determining rotational stiffness of pile groups is emphasized. Both computer program approaches and simplified methods for determining lateral and axial pilehead stiffness characteristics are presented. The paper then presents the methodology for combining the lateral and axial stiffness of single piles to determine a pile group stiffness matrix for structural analysis. An example is provided illustrating the methodology. using the idealized pile group shown in Figure 3. The influence of the pile cap on lateral stiffness is also discussed along with the influence of battered piles on load deformation behavior of pile groups. The question of moment-rotation capacity of pile groups is discussed in detail in relation to seismic retrofit problems. The paper notes that a key element in retrofit design relates to the provision of adequate

Appendix F. Supplemental Information on Foundation Effec:ts

fou mo

sUF rot dOl cor

paF stn ten ana det. cha

App.


"' 199

foundation capacity to resist the base overturning moment arising from the inertial forces of the superstructure. It has been shown that the momentrotational characteristics of a pile group can have a dominating effect on the response of a structure, as compared to lateral stiffness characteristics. The paper points out that as performance criteria for structures are now more often being evaluated in terms of nonlinear time history or pushover analyses, geotechnical engineers are being asked to determine nonlinear load-deformation characteristics of foundation systems and the

consequences of pile foundations exceeding pile capacity. An example of a pile group system subjected to axial and moment loading and the consequences of allowing permanent foundation deformation arising from transient pile foundation yield is examined. Overall, the paper provides a useful summary of design issues and approaches related to the needs of structural modeling including pile foundation systems.

If

s is

TIiE WHOL.E SYSTEM

s

k.

1. KJnematic SeLsmlc Response:

It. .lield malion

...n.u....

'n.

.....

_

... ' ..........

1I,IO"U,c iWl

_

...

-

u,II)ooU,c-

gtO~

inp.,A malion

s.il.mlC w:veS

,r

2.. Pile Group Oynamic: Impedances (~ dislriOU'oIOtI 01 ina~ loading to individual piles,

3. Supe,-sINctl,lr. Inlni;,1 Aespanse

Ie for I the ng

ion

II. J. . . .

•

k. hrw'--~,;..,

m,

I T•

Iouncilhon i(lput molion

n

Figure 1. General procedure for SeismIc sol/·Plle FOundatlon·structure Interaction

~te

eets


F·es


10 9

','

10 10 10" 10 12 BENDING STIffNESS, EI (lB-IN')

10 13

C't.,

T FIgure 2. Coefllclent for FIxed Pile Head Lateral Stiffness

F-SD

APpendix F. Supplemental Information on FOundation EffectS


Ij1bJc J

rns Srjffns;s$ Sglutign Single

Pi'''

Pile Gml!~

42

9 x 42 =378

U1Cr.II T=sloation

Ic" = kc .11:iplin)

Venial Tr.a.nsLation (kil"in) .

9 x 1.200 = 10.SOO

1.200

k~.

Rocking RollltiOQ

l'o-~ +

193.000

k..zkH ("...·kiplr>.d)

a

TorsiouJ RoULion K... (in.ldpl=l)

c.oss-Coupling kIJ=k,.=..k::14=·k.:. (kip)

4.42.48' + 4x42.(4S·.... S·) = 1.16 x 10' 9 x ·2.250 = .20.250

·2Z0

,. ·S·

".

.., :-

@

, ....

@

1··S"'" '-

@

4'

r.!f K.. S.l

= 1.74 x 10' + 1.66. 10' = 1.83 x 10'

e e .-. ,:. e;. e- '-- ': Diameler .- '

4'

S

11'

S

I

-

~'"i 1;,--';-'-'r--,"":'-.:!,l"I

...

,-

IS', I...

"

rF f--L c--

70'

y

=

120 pcf

~.

=:

30·

,. Diameler

pipe pile 0.25' wall Ihickness filled with

concrete

-

'-

Figure S. Idealized Pile croup In sand

:eets


F-S1


Resource Summary 1D Martin, G. R. and Van, L. Modeling Passive Earth Pressure for Bridge Abutments ASCE Conference - Geotechnical Special Publication # 55, Earthquake Induced Movements and Seismic Remediation of Existing Foundations and Abutments, San Diego, CA, October 1995 As a component of a bridge structure, abutments not only act as a retaining wall for backfill soils, but also serve the additional function of providing resistance to deformation to earthquake induced longitudinal inertial loads from the bridge deck. Quantifying an abutment stiffness and ultimate passive capacity is an important issue in modeling bridge structures for earthquake loading. In this paper, design procedures are briefly reviewed and the results of a numerical study modeling the passive earth pressure characteristics of bridge abutments are presented. Although the paper relates to bridge abutments, the results are also applicable to building structures in relation to the lateral passive capacity ability to be mobilized by footings or building basement walls.

F-S2

The paper utilizes a finite difference numerical approach to analyze the load-deformation and passive load capacity of abutments simulated as a rigid wall, (as shown in Figure 1) for both cohesive and cohesionless soil backfill materials. Classical passive pressure solutions are first presented and are followed by numerical parametric studies to examine the influence of wall height and material properties on mobilized passive pressures. The effects of wall friction on results are also examined in the parametric studies. The computer program FLAC used for analysis is shown to provide reasonable numerical results when compared to classical solutions. Overall the paper presents useful charts which can be of value in assessing the problem of lateral passive capacity for building foundation components, such as that shown in Figure 2.


App

•


r.'

I';~

I I I I

r";"

.,

,

l/

,

I I I I I

f

I I I I I I

I

I

I I I I

I I I I I

I

1

I

I

I I

I

I

I I I I I I

I I I I 1

I I I

I I I

I I

I

I

I

I

I

I I

I I

I

1

I I

I

-

/I I I I I I I I \\I I I I I

ical

I I I I III I I

III

>

II

I

1

I I

I I

I

I

;

I

I I

I I

II

!

I

! !

I I

,

I I

,

I

I

I I I I I I

i

i

a Figure 1

s. .vall

·n lies.

; is

Cal

I

CowanlfJ Pr.:o,:.:-n!

14.0 T

SLud~·. uIH = 1ort:!

15:8 i

~:8 ±

4.0 ~

2.0

0.0

•• - •. ~ ..•...•• _~ __ • ___ • _.4

-L;_ _ _ _~_ _ _ _ _ _ _ _~_ _ __ ; _ - - - _ _ _ < ~

o

5

IS

10

20

25

lI"ali Hei::ht 1ft)

(bl 14.0

,

,

C,)Ulllnth

T

12.0 I

•

!

. Prll!:~~n\ 'stut.:y.UlH = If)~(';

•

10.0 • s.O T 6.0 T ' . - ~ . 4.0 2.0 0.0 +-----;-----+----~

.. ..... • .... - - -. - - .. -~ ____;_---...;

f

o

5

10

W,II

15

20

25

Hei~ht (ft)

=

Figure 2. Normalized Total passive Force Np vs. wall Height H:y = 120 pcf, c = 1.0 ksf, rp= 0; E 200 ksf, v = 0.45faJ 6 = 0; and fbJ 6 = S!r

fects


F-51


imp SI6

Resource summary"

Dar.

Nakaki, D. K., and Hart, G. C. Uplifting Response of Structures Subjected to Earthquake Motions U. S.-Japan Coordinated Program for Masonry Building Researcb, Report No. 2.1-3, Ewing/Kariotis/Englekirk & Hart, August 1987 In this study, the effect of foundation flexibility and uplift on the response of slender shear wall type structures is investigated. The slenderness ratio (defined as the height divided by half the width) of structures studied has been limited to 3.5. This slenderness ratio is representative of shear wall structures of about 10 story height. Effect of inelastic behavior of the structure on response was also considered. The structure·foundation system considered in this investigation is shown in Figure I. As shown, the flexibility and damping of the soil was modeled by elastic Winkler springs with viscous damping for energy dissipation: The superstructure was represented as an inelastic, single degree of freedom system on a rigid foundation. Relative horizontal displacement between the base of the structure and the Winkler elements was not considered. The vertical Winkler foundation springs were considered effective only for the compressive forces and have zero tensile capacity. The properties of the distributed Winkler springs and dashpots were computed from the rocking spring and dashpot for a rigid foundation resting on an elastic half space. Using this approach the Winkler spring and dashpot constants, ko and Co respectively, are

ko

= 3 Ka/(2 b 3)

Co = 3 Ca/(2 b 3) In the above equations, b is the half width of foundation and Ka and Ca, the stiffness and

F-S4

damping coefficients of a rigid circular massless foundation on an elastic half-space, are Ka = 8 G r 3/[3(I-y)] Ca = a 8 G r4/[3(I-y)v s] where G is the shear modulus of the soil, r is the radius of foundation, y is the Poisson's ratio, Vs is the shear wave velocity and, a is a dimensionless coefficient dependent on the frequency of excitation, the radius of foundation, and the shear wave velocity. The value of ex was taken as 0.2 in this study. The study was aimed at concrete and concrete· masonry shear wall type structures. Therefore, the inelastic behavior of the superstructure was represented by stiffness degrading hysteresis mqdel developed by Newmark and Riddell. Two distinct limit states were defined for the system under consideration. The first limit state is associated with the initiation of uplift and marks the first change in the stiffness of the system. The second limit state corresponds to the yielding of the super structure. P-A effects, since they were considered to be small, were not included. Direct time-history analyses were performed for two types of ground motion: a long duration strong motion and a short duration impulsive type motion. The long duration motion was represented by the SOOE component of 1940 EI Centro earthquake. This record has a peak horizontal ground acceleration (PGA) of 0.35g. The PGA in the vertical direction is 0.2Ig. The short duration,


of il stud the, eart For pres spec and appl iIlus wid, 30'

MOl stud wall sum sign rod fixe, Whc fixe, muc som high had Wit! was prod peri, the I moti stTU( than

ApPI


is

s Ir

in te he

)

he

d

pe ted

impulsive type motion was represented by the Sl6E component (PGA = l.l7g) of the Pacoima Dam record scaled down to 50 percent. Influence of including vertical ground motions was also studied. Parametric studies were performed to study the effect of foundation flexibility and uplift on the earthquake response and the potential for damage. For this purpose, the results of the analyses were presented both in the form of the ductility demand spectra and time history plots of the input energy and the dissipated and absorbed energy. Practical application of the concepts developed was then illustrated by analyzing a 9-story, 78' high and 27' wide concrete masonry shear wall supported on a 30' long by 4' wide by 3' thick concrete footing. Monte Carlo simulations were also performed to study the influence of yield capacity of the shear wall on response. The key findings of this investigation are summarized below: As shown in Figures 2 and 3, uplift resulted in significant reduction in ductility demands when the rocking frequency of the system was less than the fixed base natural frequency of the structure. When the rocking frequency was greater than the fixed base frequency of the structure, uplift had much smaller effect on ductility demands and, in some cases, allowing uplift produced slightly higher ductility demands. The frequency content of the ground motion had significant effect on the ductility demand. With uplift allowed, the EI Centro motion, which was richer in short period motions, tended to produce higher ductility demands for shorter period structures founded on stiffer soils; whereas, the Pacoima Dam motion, richer in longer period motions, was more severe for longer period structures. When the rocking frequency was less than the fixed base natural frequency of the

structure, including the vertical ground motion, it typically increased the higher mode coupling thus increasing the ductility demands. For cases where the rocking frequency was greater than the fixed natural frequency of the structure, inclusion of vertical ground motion typically had little effect on overall response. Time history energy plots for a system with fixed base natural period of 0.4 sec., see Figure 4, showed that the hysteretic energy loss in a system with uplift permitted is smaller than without uplift thus implying less earthquake damage in the uplifting structure. However, the energy spectra plots, Figure 5, showed that allowing uplift, in most cases, resulted in larger hysteretic energy loss than without uplift. This was especially true for the Pacoima Dam motion which was rich in long period motion and for periods greater than about 0.5 sec. Thus, it was concluded that on the basis of energy dissipated, it can not be conclusively stated that allowing uplift will reduce the damage sustained by the structure and, also, ductility demand alone may not be a true indicator of damageability. Including the vertical ground motion generally increased the energy dissipated by hysteresis. The shear wall studied had a fixed natural period of vibration of about 0.5 sec. The results of the analyses for this case study showed that allowing the structure to uplift from its foundation resulted in an essentially elastic response. The authors, however, cautioned that the reduction in the inelastic deformations in the structure comes at the expense of uplift displacements at the foundation level which must be accounted for in the design. Monte Carlo studies on the variability of yield capacity of the shear wall showed that the foundation rotation and uplift were more sensitive to yield capacity of the wall than the displacement response.

in In,


F-SS


he

\I

II

e h

v

e

IB

mF

i c

kO

Co

I

, b

b

YG'L XG Figure 1. Structure-FOundation system Madel

P-s&


FI9

App


IZ.O

1t.0

·

10.0

·

t.O

·

-

Q

'.0

I~

7.0 -

S Ii ;:) Q

\ \ \ \

uplift

\ \

-

-

uplift prcycn'ted

\

\

'.0 -

,\

5.0 -

\ \

4.0 U

-

\

-

1.0 -

,, - -

...

--- ...

1.0 0.0

• 0.0

0.1

•

0.4

•

0.1

•

0.1

mal

1.0

•

1.1

0

IA

(SEC)

Figure 2. Ductility Demand for structures Allowing Uplift and With Uplift Prevented for the EI Centro Motion, 1=$.5, P 0.9, fJ 0.7, 01. 161r rad. sec., ~s 0.05, ~. 0.20

=

,

=

=

lets , Appendix F, supplemental Information on Foundation Effects

=

=

F-S7

•


5.0

4.0

/

,.- ....

/

" ....

/

3.0

/ I

" ....

I \

Z.O

I

" "- ....

....

\ I

..

uplift

1.0

-

-

-

uplift prevented

0.0 0.0

0.1

0.4

a.1

0.' I"tRQ)

1.0

1.4

(SEC)

Figure S. DuctIlity Demand for structures Allowing Uplift and With Uplift prevented for the Scaled pacoIma Dam MotIon, .It = s.s, P = 0.9, f3 = 0.7, 01. = 1IiTC rad. sec., ~. = o.os, ~. = 0.20

F-S8


...


000.0

/

"00.0

i

upfin preveRled

300.0

.....

.. ! !;

200.0

6 ~ ~

100.0

0.0 0.0

2.0

4.0

0.0

0.0

nUE

10.0

1:l.0

'4.0

1 •. 0

18.0

(SEC)

Ina Figure 4. Hysteretic Energy with Up/1ft and with Wl/ft Prevented for the scaled pacoima Dam Motion, t=O.4 SEC., Cy=D.261, A=S.S, p=O.9, /3=0.7, mr= 161c rad. sec., ~1=O.05, ~r=O.20

'ect5

F-S!J


---------- _._-

-- -----.------

--.----- --


Figure 5. Input Energy and Hysteretic Energy per unit Mass spectra for the Scaled pacoima Dam Motion, ).=5.S, P =0.9, fJ = 0.7, mv= 161r rad. sec., ~s=O.OS, ~v=0.20

P-60


•


Resource summary 12 »>'>YL»

Pender, M. J. Aseismic Pile Foundation Design Analysis Bulletin of the New Zealand National Society for Earthquake Engineering, Volume 26, No.1, March 1993

This paper presents a comprehensive presentation of methods of assessing, for preliminary design purposes, the stiffness and capacity of pile foundations under seismic forces. Emphasis is placed on expressions for pile stiffness and capacity in the form of simple formulae and illustrations on the use of formulae through a number of worked examples. Comparisons between data from field testing of foundations and analysis methods are also presented. Specific topics covered in the paper include the following: • Observed seismic response and damage to pile foundations during past earthquakes

•

Discussions on the role of kinematic soil-pile interaction and dynamic response of pile groups The case history documentation on the response of pile foundation systems to seismic loading provides a comprehensive overview. One interesting documented case history relates to the response of the Imperial County, California, Services Building (Figure 1) during the 1979 Imperial Valley magnitude 6.3 earthquake. Forced vibration tests on pile caps following building demolition provided the means for earthquake response studies including the effects of foundation interaction. The objectives of the paper are focused heavily on design analysis, particularly preliminary design. The approach presented is suitable for use with spread sheets albeit more sophisticated methods and computer approaches are also discussed. The emphasis on assembling an extensive set of simple formulae and the use with examples make this paper particularly readable for design engineers. Nearly all the methods discussed focus on the common idealization that the soil-pile system will respond in an equivalent linear elastic manner to applied loading. In this respect a useful compilation of correlations between Young's modulus and the coefficient of sub grade reaction with standard penetration blowcount are provided. The reality of nonlinear soil behavior which occurs during strong seismic loading is discussed briefly, and illustrated using the results of field load tests. The presentation also identifies limitations of the various analysis methods and topics requiring

•

Models for pile lateral stiffness including Winkler and elastic continuum models

•

Models for pile vertical stiffness including Winkler and elastic continuum models and discussion on battered piles

•

Methods for evaluating stiffness of pile groups including vertical, rotational and lateral stiffness components

•

Discussions on the influence of nonlinear soil behavior on soil·pile interaction including case studies

•

Discussions on correlations between subgrade soil properties and penetration resistance from field tests

In,

•

Discussions on results of dynamic tests on prototype scale piles and pile groups and

Ffects

APpendix F, supplemental Information on Foundation EHects

F-G'I


further research. A detailed valuation of the relative role of inertial and kinematic interaction of pile foundation subjected to dynamic loading, clarifies the relative importance of the two effects. The significance of pile group effects is also addressed using analytical data and results of field load tests, as for example shown in Figure 2. Questions not addressed in detail in the paper

include the effects of liquefaction and potential degradation in stiffness of soils adjacent to the pile due to cyclic loading. Overall the paper provides good insight as to the mechanics of soil-pile interaction and is perhaps one of the most comprehensive state-ofthe-practice publications available on seismic pile foundation design.

(ASr 'ACt SHEAII WALL WEST rACE SH[AR WALL

• •

•

•

• •

• •

• •

• •

TYPICAL FLOOR PLAN WEST ELEVATION

,

II

~

4 )

2 G

\ LONG

SECTION

EAST ELEVATION

Figure 1. COnfiguration Of the Imperial County Services Building (after Hall, 19B4J

F-62

Appendix F, Suppiementallnformation on Foundation Effects

FIG

App


)iie

o 30

Je

on

Cr.m.J

...

.~

.!! 20

n:

,/

'"

/0

O'"b

.~"

0 /:"'.,..,

/'"

'if',---'''''-' .g....~.-.. -

~

0

/0 ,,/

0

10 ..J ~

a. > «

/~/

0

/ 0/

~ .-

/J\/ / cS v

~

c..

/

/. /' -8-'

~.c

.",

0.0

0--<>

Single Pile O--OLeadinq Row, Group <>-..... Middle Row, Group o--oBack Row, Group

y

1.5 1.0 0.5 Deflection 01 Load Poinl, inches

2.0

30 0

'"

Q.

~ycle I

:;z

.~ 20

A

Il..

.... QI Il.. "C

C

0

..J

10

C. >

.~

;a

0

500

o Single Pile o Leading Row, Group '" Middle Row, Group o Back Row, Group

1000

1500

Maximum Bending Momenl, Inch-kips

Figure 2. Pile Head DeFiectlan and Maximum Moment Against Average Farce lor a Sxs group and a Single Pile 1 klp=4.54 kN, 1 In. =25.4 mm(after Brown et al., 198B)

ffects


F·SS


Resource Summary 15

2.

Priestly, J.N., Evison, R.J., and Carr, A.J.

Seismic Response of Structures Free to Rock on Their Foundations Bulletin of the New Zealand National Society for Earthquake Engineering, September, 1978 This work is an experimental extension of a theoretical development of the behavior of rocking blocks by Housner (see Resource Summary No.4 and Figure 1). The paper begins by discussing the fact that rocking of structures may be beneficial to their seismic response. The seismic code in New Zealand, at least at the time the paper was written, implicitly recognizes this by allowing the forces on foundations systems to be limited in the recognition of the possibility of rocking. The writers concur with this approach since they feel that rocking and possible deformation of soil beneath footings have improved overall seismic performance for some structures. They point out, however, the possibility that substructure rocking and possible deformation of foundation materials can cause damage at the foundation and first floor slab levels. The authors note that, in spite of its beneficial effects, that rocking has received very little attention by other researchers. One exception is Housner and the other that they note is the work of Bartlett (see Resource Summary No.1). Review of Basic Theory. The authors begin by reviewing the theoretical development by Housner. They expand Housner's work in one interesting way. Housner expressed the energy loss due to the inelastic collusion of the block with its base by an energy reduction factor 2

MR2

r= ( I-T(I-cos2a)

)

where M is the mass of the block, 10 is its mass moment of inertia about the point of rotation, and

P-&4

3.

a is the aspect angle for the block (see Figure 1). This enabled the prediction of the peak of rotational displacement as a function of the . number, n , of impacts as det per cen

where 6

(l

a

a

tP. =-"- and 1/10 =-l!. The authors point out that the fraction of critical damping, A. ,for single degree of freedom oscillator may be expressed as A=_I In (1/10) It n tP. Using this equation for fraction of critical damping and the previous energy reduction factor from Housner, the two variables can be related as illustrated in Figure 2. Response Spectrum Design Approach. If a roCking block is represented as a single degree of system with constant damping, the period of vibration depends on the amplitude of rocking, and a trial and error approach can be used with a response spectrum to determine peak displacement during earthquake motion. The following procedure is outlined: 1. Use the no-rocking natural period of vibration and damping of the structure combined with the acceleration response spectrum of the design earthquake to calculate the elastic


4.

5.

6.

pro wei

Thi con pre 0.4 mo

a. b.


Ilg,

response and check that this will induce rocking. 2. Using the relationship between Housner's r factor and critical damping. A, shown in Figure 2, calculate the equivalent viscous damping of the rocking system. 3. Using Housner's equation

1

4 -1 [ 1-1~ T=pcosh

where

p=~WR

Jr as I

Jf and ent

on

Itts

10 determine the relationship between the rocking period, T, and the amplitude of rocking at the center of mass, 00 (see Figure 3(a». 4. Estimate the maximum rocking displacement, Ll1, and determine the corresponding period of vibration, TI. 5. The maximum displacement response, Ll2, of the equivalent elastic system can then be found from the displacement response spectra as shown in Figure 3(b). A new period, T2, can then be used to refine the estimate of displacement using Figure 3(a). 6. This iterative procedure usually converges within three to four cycles yielding an estimate of the peak displacement of the single degree of freedom system. Model Study. In order to verify the theoretical procedure outlined above, experimental studies were conducted on a model as shown in Figure 4. This model was a one-sixth scale simulation of a concrete shear wall building with a prototypical pre-rocking fundamental period of approximately 0.40 seconds. The foundation conditions for the model were varied as follows: a. model supported directly on shaking table. b. model supported on 25 mm. rubber pads at the corners of the foundation resulting in a three-

fold increase in natural, pre-rocking period of the structure. c. similar to b. except that the pad was placed beneath the entire footing as opposed to the four corners. In addition to these foundation conditions, some free-rocking tests were carried out in the field on the model supported by well-compacted clay soil. The purpose of these foundation conditions was to investigate the general effects of foundation compliance on the results. The model was subjected to four different types of dynamic tests. I. The natural decay of the system for free vibration at amplitudes less than that required to initiate rocking was documented. 2. The natural decay of the rocking motion was investigated by uplifting the model beyond its rocking limit and releasing it. The relationship of amplitude of rocking to natural period and the rate of decay provided experimental data to compare with Housner's basic theoretical equation. 3. The model was subject to forced sinusoidal base accelerations to study the response to different frequencies. 4. The response of the structure to actual seismic excitation was studied using a scaled record of the 1940 EI Centro Earthquake. The authors report only on the results of the free rocking and seismic testing in their paper. The results of the experimental studies correlate excellently with Housner's theoretical relationship between rocking frequency and amplitude. This is illustrated in Figure 5. It should be noted that the models on the rubber pads had a maximum rocking frequency of about 3.7 hz and, above that level, Housner's theory does not apply. The experimental results show that Housner's assumption of completely inelastic energy dissipation due to impact may be incorrect in some cases. For the model, Housner's theoretical equations would have predicted a reduction factor, r, of 0.70. The authors adjusted the results of the natural decay of rocking of the model to match


F·BS

----- --- -' -------... ----'--- '"


Housner's general relationship. To do so they had to use a reduction factor of 0.87. The results are shown on Figure 5(a). The result of the earthquake excitation from the essential records indicate that they response spectrum approach proposed by the authors provide a very good estimate of the maximum displacement of a rocking block. The predicted maximum displacement was 50 mm. and the corresponding experimental value was 45 mm. Design Example. The authors provide design example for a masonry structure. The procedure used is that of the New Zealand building code. The authors determined that the code value accelerations will result in rocking of the structure. This is confirmed by a rocking elastic response spectra calculation using procedures developed in the paper. These calculations indicate that for the EI Centro records, a roof displacement of approximately 160 mm. would occur in the sample structure for a period of 1.6 seconds. The authors note that this amount of rocking is not necessarily excessive but that possible damage to the slab on grade and the foundations might be induced. Conclusions. The authors offer the following general conclusions: 1. Response of a rocking structure is similar to base isolation in that lateral accelerations are limited to that which induced rocking. Structural damage might be reduced by designing structural elements which remain elastic until the rocking acceleration is exceeded.

F·GG

2. Housner's theoretical equations for frequency and amplitude of a rocking block are verified. Foundation conditions in this study did not appear to have a significant influence on the rocking response; however, Housner's assumption that rocking impacts are inelastic was found to not be valid from the models in this study. 3. Extension of Housner's theory to a simple method to predict maximum displacement utilizing response spectra has been developed and verified by limited shaking table testing. 4. This approach might be extended to structures other than buildings including bridges and nonstructural applications such as the rocking of stacked containers. 5. Further research is required to verify the scope and the feasibility of the methods developed in this work. In particular high equivalent viscous damping predicted for squat rocking structures needs verification. It should be noted in reviewing this material that the foundation conditions for the model were probably highly elastic. This may explain why Housner's assumption of inelastic impact was not born out. Bartlett's work, which was reviewed in Resource Summary No.1, may provide an interested opportunity to estimate the foundation yielding at the base of rocking blocks. It is Iikel y that this inelastic effect is dependent on the initial contact pressure as a percentage of the ultimate bearing capacity of the foundation material.


h

1 -

Ap,


cy :d.

ic in

h

+

,. 15

'"

c

~ 10

~

h

ed res

.3

L

!I

,

- --I

O~:~

____~______~~--__------~__~~~

05

10n-

0.6

)f

:ope j in

all

0.7

Energy Reduction

Q

9

1.0

Factor. r.

FIgure 2. ApproxImate RelatIonship Between EquIvalent VIscous DampIng and Energy ReductIon Factor

FIgure 1. Housner's RockIng Block

;ous nes

're lot

in

ly :ial

,

,, ,, ,,, ,,, , I

II

" a:

,, '' ,,

,

, I , I

I

.2

, n estimate~ I

L

,st

4.:

Amplitude

(al

ROCKING

I'

T. T, Period

dtl : 4l, of

Rocking

CHARACTERISTIC

(bl

TRIPARTITE

RESPONSE

SPECTRA

FIgure S. EstImate 01 MaxImum DIsplacement From Response spectra

'ects


F·G'


I lnstrum.ntation

--= : DC-LVDT

~

o : Accrl.rom.trr

an ef: de rei !hI fOI

bl!

pe -INPUT-

Fil

-QUTPUT-

COi

di! Fillure 4. SChematIC tJI MtJdel tJn Shake Table

Ih~

me ani

anI SPI Theory (tbusner)

I

•

"O~

a:

Exp.

•

" 0.6

Exp . rigid

:

•

'$

0.<4

""''$..,.

02

rubber

pad

base

Exp,

rubber pad

Exp.

SOl'

soi by the

base

-----.-

nor

"~

.

4 a 12 W °O~------------------------~----~

No

(a)

Exp. rigid baS(>

r: 0.B7

Theory

OB G!.

Wi Iha api Wi

DECAY

01

OF'

B

Impacts (n)

ROCKING

24

16

Horizontal Ib)

FREQuENCY

vs

OisPlacemQ'nt

32 (mm)

AMPLI TUDE

Figure 5. ThetJry VI. Experiment ItJr MtJdel

F-G8

Appendix F. Supplemental Information on FtJundatlon Effects

iml ene use effi the Slrl bot)


Resource Summary 14 Psycharis, I. N. Dynamic Behavior of Rocking Structures Allowed to Uplift Report No. EERL-81-02, Earthquake Engineering Research Laboratory, California Institute of Technology, Pasadena, CA, 1982

-32

nml

iCts

In this report, the equations of motion for the analysis of simple structures considering the effects of foundation flexibility and uplift are developed. The presentation begins with the recapitulation of the classical work of Housner for the case of a rigid block allowed to rock on a rigid foundation. The equations of motions for the rigid block resting on flexible foundation and with uplift permitted were then developed. As shown in Figures 1 and 2, two types of soil model were considered: the two spring model and the distributed Winkler spring model. It was observed that the equations of motions for the two spring model are much simpler than the Winkler model and are more useful for simplified practical analyses. Therefore, expressions for the two spring model in terms of the parameters of the Winkler springs were developed. It was observed that the equivalent two springs model can approximate the response of a structure with a Winkler foundation reasonably well. The effect of energy dissipation and nonlinear soil behavior in foundation soils was approximated by three different mechanisms: viscous dampers in the foundation model; elastic-perfectly plastic nonlinear foundation spring elements; and an impact mechanism that allows dissipation of energy during impact. It was concluded that the use of viscous dampers was the most practical and efficient way of modeling the energy dissipation in the foundation soils. The equations of motions for a flexible structure allowed to uplift were then developed, both for a single degree of freedom system (Figure

3) and a multistory structure (Figure 4). For the multistory structure, an approximate analysis procedure considering only the first vibration mode was also developed. The approach developed was used to analyze the response of the Millilken Library building subject to the S16E component of the Pacoima Dam record obtained during the 1971 San Fernando earthquake. The building is a 9story concrete building with a one story basement, see Figure 5. The key conclusions and observations of this investigation are summarized below: • Allowing uplift leads to a softer vibrating system, i.e. the first period of the uplifting system is always greater than the first period of the structure with uplift prevented. •

Higher periods of vibration are not significantly affected by uplift.

•

When damping is considered in the foundation springs, uplift tends to reduce the critical damping in the first mode of vibration.

•

Allowing uplift can have significant effect on the seismic response of structures. For the rigid superstructure, uplift tends to reduce the rocklng accelerations while the angle of rotation can either be larger or smaller. For a flexible superstructure, uplift always tends to increase the angle of rotation of the foundation. However, no conclusive statements could be made about the response of the structure. For the Millilken Library building studied, it was observed that the building displacements were not significantly affected


F-GI


and the accelerations and base shears were slightly increased due to uplift.

d~

,

___

•

The simplified fundamental mode approach for the analyses of multistory structures provides reasonably good approximation of response.

~

IE_-_ o

01_ .•.

!

1--

--..

I

r y

FIg

MO

b

mg

of!i-{ __ (...,

Y

-tt

M

•

t.)·Three-dhnen,tonlll clse

0

k

k

(b) Equivalent tNo-dimensional 'Probl.

FIgure 1. Tipping Block on a "TWo-Spring" FOundatIon 'E---_.. __ a

-----

.•. ..

/('-

~~~.~..~--.-~

..Jy b

T

b

mg

h

y,L··"

(a) Full-contact case

y,L··"

(b) After lift-off

FIgure 2. RockIng Block on ContInuous ElastIc FoundatIon' (WInkler made/)

'-70


App


I

x

u

1/ iii/

r! J

;h for

ides \se.

m

1 ,

1.1-.:

h

Flgure:s. Simple Oscillator on a TwO-SprIng MOdel of a ROcking Foundation

~'

M

k

r - -;.- - - -

k

;::,I --u'";"I" - - -""'l~?»~~

;--r--T--- I

Hn

\

1

(n I

I

'

~-~--~-~-~~~~ ,----,-I

,

I

,

I

{'

I I I

I' " : { , , ~.'\P'>_

I

. L __ -/ I.._-

:

r I;

I

I .'

I

e.:y- e

1/

I; "

c

'Y.l,. .

"

Figure 4. Rocking n-story structure on Two-Sprlng FOundation

fects

APpendix F, Supplemental Information on Foundation Effects

F·71


I

D

•

B

oj

w di I! bl nl

SU

TI e~

bl

....--\l=.;;:.,.jl---II ~ .....

of

Ie'

ge Ff

''';

i--~=';;:"'Io-_-II~UII so'

.

'

lhl

is

ap fir lat pic Fo fOI

c

Th

FIgure 5. Milliken library BuildIng: fa) FoundatIon Plan and N-5 section; fb) TypIcal Floor Plan; fC) a N·S SectIon VIew; fd) VIew oF Building LookIng Northwe$t

F·72


the Ce tra bel vel


Resource Summary 15 Rutenberg, A., Jennings, P. C. and Housner, G. W. The Response of Veterans Hospital Building 41 in the San Fernando Earthquake Earthquake Engineering and Structural Dynamics, John Wiley and Sons, Ltd., Volume 10, 1982 Description of Building and Earthquake Damage. This paper documents an analytical study of the Veterans Administration Hospital Building which was located in San Fernando, California, directly over the fault plane of the February 9th, 1971, earthquake. Although several V.A. Hospital buildings at the site collapsed, as did the neighboring Olive View Hospital, Building 41 survived the earthquake with only minor damage. This was in spite of the fact that the building experienced very intense ground shaking. The building was designed for a lateral force coefficient of only 10 per cent. This study consists of several levels of analysis in an attempt to reconcile the good behavior of this building during the San Fernando earthquake. The subject building was designed in 1937 by the Veterans Administration Engineering Office. It is four stories in height with plan dimensions of approximately 200 by 50 feet (see Figure 1). The first floor is partially below grade. The vertical and lateral load carrying system consisted primarily of pierced reinforced concrete shear walls and frames. Foundations were continuous with isolated spread footings under the walls and interior columns. There were approximately twice as many walls in the longitudinal direction as the transverse. Consequently, the analyses concentrates on the transverse direction. Wall thicknesses varied between 10 and 16 inches and reinforcement, both vertically and horizontally exceeded .002 times the

Fects

cross-sectional area of the walls. Boundary reinforcement was provided around openings and the general level of detailing of the reinforcing was quite good from an earthquake performance standpoint. These details were also evident in the spandrels between walls, allowing for coupling between adjacent wall panels. For the purposes of the analyses and this study, an aged nominal concrete strength of 4,000 psi was assumed. The yield strength of the plain and deformed reinforcing bars was assumed to be 30 ksi. Soil conditions at the site consisted of alluvial deposits and the allowable design bearing stresses were up to 8,000 Ibs. per square foot including earthquake effects. The material was reasonably competent and had virtually no potential for liquefaction. For the purposes of the analyses, a shear wave velocity of 1000 feet per second was assumed for all strain levels. The unit weight for the soil is assumed to be 100 lbs. per cubic foot. These assumptions yield a shear modulus for the soil material of 3600 ksi. The building suffered very minor damage during the earthquake. Some of the shear walls exhibited hairline diagonal cracks. There was a small, continuous crack running in the basement and first floor slab. Some settlement of the exterior grade was noted on the upslope side of the building. In the analyses, this settlement was taken as a possible sign of foundation movement. However, it could also have been caused by the


F·7J


consolidation of poorly compacted fill adjacent to the basement wall. Major yielding of the coupling beam between walls probably did not take place based on field observations. The overall cracking in the concrete walls was indicative of the fact that in some locations the diagonal strength of the concrete may have just been exceeded. Ground Motion. There were no records of ground motion at the site or in the near vicinity. It was, therefore, impossible to reconstruct the high frequency components of ground motion at the Veterans Administration Hospital site. The nearest accelerograms were at the Pacoima Dam (1.25 mi.) and at the Orion Holiday Inn (8 mi.). From an analysis of these records it is estimated that a peak spectral acceleration in the range of 0.1 to 0.3 seconds was on the order of 0.7 to 1.5 g. Equivalent Lateral Force Analysis. An analysis based on a 1976 Uniform Building Code including modifications for soil structure interaction in accordance with ATC 3-06 was performed for the building in its transverse direction. Considering the symmetry of the building, three of the six transverse walls in Figure 2 were included in the analysis. Several alternatives were considered for distribution of shear to the individual walls. These are summarized in Tables 1 and 2. Also noted in the tables are the fundamental periods for the various analysis assumptions including a rigid base and one including foundation compliance. It can be seen that the rigid base period of approximately 0.14 second was substantially lower than of 0.20 second estimated from the foundations that were assumed to be flexible. The code equation for period indicates an even a longer period of 0.36 second. The 1976 Code would have required a base shear coefficient of 0.28 due to the essential nature of the facility and the local soil conditions. Calculations for this study indicate that the first yield in the building would occur at a base shear coefficient of between 0.15 and 0.20 (see Table 3 and Figure 3). The total capacity of the building assuming redistribution after initial yielding would have been in the order of 0.30 and 0.40 g. The base

F-'4

overturning moment for stability would have limited the capacity to 0.45 to 0.50 g (see Table 3). Comparisons with these general capacity estimates with the inferred level of ground shaking implies a degree of damage much greater that that which actually occurred. In order to further resolve this inconsistency subsequent analyses were performed. Linear Dynamic Analysis. Investigators performed a linear three-dimensional dynamic analysis of the building using the program ETABS. In this analysis the structure was assumed to be supported on a rigid foundation. This analysis indicated that the response of this building was primarily in its first mode of vibration. This analysis resulted in a higher prediction for first yield in the building (see Table 4) due to the more accurate inclusion of the coupling beams between the shear walls. The capacities shown in Table 4 are in terms of spectral acceleration. Reducing these by the modal participation factor of approximately 85 - 90% would indicate a capacity similar to that of the code analysis. As a result, the Linear Dynamic Analysis corroborates the basic strength of the building. but fails to explain the low level of damage for the high level of seismic shaking. For this reason the analysis was extended to include nonlinear effects. Nonlinear Analysis Including Foundation Effects. The researchers developed a simplified two-dimensional model of one of the shear walls using the computer program DRAIN-2D (See Figure 4). In addition to nonlinear elements for the shear wall and coupling beam, the model included a nonlinear representation of the soil material beneath the footings. This foundation model could accommodate both uplift and plastic yielding of the soil material (see Figure 5). This model was subject to two earthquake records. The Pacoima Dam (PD) and the Holiday Inn (HI). The records were modified to yield approximately the same fixed-base shear as a constant acceleration spectrum at 0.9g. Numerous analyses were made of the model with variations


ir

SI

a!

T

t1:

VI

Sl

VI

c!

in

c(

ill to ar cc

F H

O.

th TI

n

pr

Wi

Ap


ng at

led

ing s lre

ity the : iOW

jed

Is

. the jed uld f the

lay

The authors find that Wall A (see Figures 2 and 3) would not have experienced forces greater than its capacity for the majority of the load cases in the nonlinear analyses. They also note that the hysteretic energy dissipated in the vertical soil springs was small. The results indicate that the effects of uplift tend to predominate over the strength of the soil at the compression toe. Although the displacements for the vertically flexible, weak soils were larger than for stiffer, stronger soils, the forces in the shear walls themselves were generally less. A long period acceleration pulse, evident in the Pacoima Dam record, did not appear to have a large effect on the response of the building. Even though the period lengthened to approximately 0.35 seconds, it still was substantially less than the 2.0 second period of the impulse. Conclusions. Based on their analyses, the authors concluded that the observed successful performance of Veterans Administration Hospital Building #41 during the San Fernando Earthquake was the result of its initial large strength and good design details, as well the unanticipated beneficial effects of nonlinear soil/structure interaction. For this reason they conclude that partial uplift and yielding in foundation soils generally tend to reduce seismic forces in the structure and should not necessarily be avoided by designers. They also found that many of the simplified assumptions that they made for modeling appear to have little effect when compared to the results from more sophisticated models .

in a number of different parameters. These are summarized in Table 5. In this table, the case listed as CSA is the constant spectral acceleration of 0.9g. The damping coefficients a and f3 are both from the conventional Rayleigh damping relationship of C=aM+!3K

The soil properties are tabulated for lateral and vertical movement. The variables Cx and Cy are the stiffnesses for the soil components in the lateral and vertical direction respectively. Yx and Yy are the capacities of the soil components. The results of all the analyses are summarized in Tables 6, 7, and 8. In Table 6, the fixed base comparison between the three input motions illustrates the adjustment of the two actual records to approximately 0.9g base shear. The interactive analysis shown are for linear soil properties in compression and tension with no uplift allowed. From the interactive results, it is apparent that the Holiday Inn record has higher accelerations in the 0.20 second period range than those associated with the fixed base period of approximately 0.14 second. This amplification is not apparent in the Pacoima Dam (PD 2) record. Examination of the results in Tables 7 and 8 provide a number of interesting conclusions. The wall shears and base shears for the nonlinear analyses are consistently lower than that predicted by the linear analyses. An even greater percentage reduction was found for overturning moments in the walls. The effect of vertical acceleration was not found to be significant for these ground motions.

Table 1. Lateral Force Distribution Over Building Height (approximate analysis)

Level

F.' (kip)

Shear (kip)

Moment (kip/ft)

PH

605 2,770 2,710 1,875 1,040

605 3,375 6,085 7;960 9,000

0-0 9,500 50,000 123,000 219,000 354,000

Roof 3 2 I

0

ous ns

'ects

• These (orces are arbitrarily scaled so the base shear equals the weight of the building.


F·75


Table 2. uteral Shear Farce DIstrIbution Among walls (percentJ; and natural perlad (s)

Wide flanges

Narrow flanses WaD

B

B+S

B+S+F

B

B+S

B+S+F

A B C Period

25

25 29

21 29 50 0-233 0-243 (I)

31 25 44

27 29 44 0-123

27 29 44 0-204 0-213 (I)

27 48

T,

46

0-140

OriEina! analf,is (19 7)

B

29 40 31

B - Bending derormation: S - Shear derormation. F - Foundalion compliance. (I) Including mass and rotary inerli. or roundation.

Table S. ute,.al Farce capacIty of BuildIng 41 (app,.oxlmate analysIs)

Percent or weight

Overturning

First yield

Cumulative-

(minimum)

It redislribulion

15-20

3S-45

45-50

50-55

Overturning

• Yield and redistribution.

Table 4. Spect,.al AcceleratIon levels at Different stages of Res/stance

Per cent 9

F-7G

First yield

Yield and redi
Cumulative capacilY

25-30

3S-45

45-50


Ap

-

SEISMIC EVALUATION AND RETROI=IT 01= CONCRETE BUILDINGS

Table 5. soil and Superstructure parameters for cases Analyzed SoH properties Lateral

Vertical

Damping

E.Q.

fects

. Case

scale

CSA HI.I HI.2 HI.3 HI.4 HI.5 HI.6 HI.7 HI.8 HI.9 111.10

0-9 2·0 2·0 2·0 2-0 2·0 2·0 2-0 2·0 2·0 2·0

HI.! I

2·0

III.! 2 Ill.! 3 111.14 pO.! PO.2 PO.3 PO.4

I'S 2·0 2·0 0·4 0'4 0·4 0·4

PO.S 1'1).6 rO.8 PO.!!

0·4 0-4 (}4 O'S

rz,{I

'" '" §

'"II

.

"" ...,

'" II

to

a-O P. - (}00228 ~=O

P=

0,0028

'" '"

~

.'" II

"" ..., II

to

c,

Y,

c,

y.

(Idp/ftl)

(ksl)

a,

'"ex:: 400

ex::

400

I'S I'S HI HI o-S I'S I'S 1·$

1,000 1·000 1·000 0·333 0-333 0-333 0'2S0 0-333 0-333 0·333 0-333

1,200 1,200 1,200 1,200 1,200 1,200 500 500 500

400

1'5

0-333

400 400 400

I'S I'S I'S

'"

'"'" 1'5 1·5

0-333 0-333 0·333 1,000 1-000 (}333 0-333

1,200 1,200 1,200

3-75 3-75

0·333 0,250 (}333 (}333

1,200 1,200 500 1,200

3-75 3-33 3-75 3-7S

400 400 400 400 400 400 167

400 400 400 400 400 400 400

ex;

'"

HI

. 1,0 I'S 1,5


(kiplftl)

(kll)

a,

co 3-75 3-75 3-75 3-33 3,00 3-75 3-75 3-75

1-000 1-000 1-000 0-333 (}333 0,333 0-250 0-333 (}333 (}333 (}333

500

3-75

(}333

500 SOO 500

3-75 3-7S 3·75

, Variations in super. structure properties

"c

0

Z

E = 0-8E.

1 u

c

0

Z (}333 (}333 (}333 1·000 1-000 (}333 (}333 I, = O·SI • 4\op beams (}333 (}250 "0c (}333 Z (}333

j

F·77

._-


Table 6. l.Inear Response: CDmparatlve Results Interactive

FIXed base

CSA'"

HI.I

PD.I

HI.2

PD.2

562

559

571

802

580

6,858

6,730

6,991

8,043

5,728

724

661

698

1,0;35

859

(kips) Base shear (kips)

1,124

1,118

1,142

1,787

1,331

Foundation

1,561

1,367

1,449

2,507

2,048

0021

0020

0023

0066

0-48

0011

0-09

0006

0-07

Wall sbear (kips) Wall moment . (kip-ft) Wall axial

compression

axial force (kips) Lateral roof displacement (in) Downward displacement (in) (2) Upward displacement (in) (2/

(1) Constant spcclral acceleration at 0-99. (2) At extemal column line; includes static: settlement of about 0-03 in.

Table 7. Nonlinear Response: comparative Results Response Wall shear (kips) Wall moment (kip-ft) Wall axial compression (kips) Base shear (kips) Foundation axial force (I) (kips) Lateral roof deBection (in) Downward displacement (2) (in) Uplift (2) (in)

HI.3

PD.3

HI.4

PD.4

HI.5

PD.5

HI.6

PD_6

HI.8

PD.8

HI.7

HL9

481 4,754 901

367 3,499

SOO

469 4,795 890

354 3,692 748

507 4,997 898

370 3,583 712

481 4,830 940

337 3,264 644

455 4,574 944

358 3,833 611

518 5,156 944

4,0.12

1,223 1.911

839 1,591

1,245 1,948

828 1,573

1,195 2,018

842 1,525

1,157 1,985

750 1,428

1,068 2,047

875 1,462

1,313 2,039

1.139 2,112

0095

0-46

HlI

0047 _ 0079

0046

1-00

0-46

1'33

0071

1·07

1-24

0016

0-12

0016

0012

0018

0011

0023

0013

0-42

0025

0·19

0-.$4

0040

0-10

0042

0011

0-27

0-09

0034

O{)S

0-50

0015

0·45

0-46

407 1,004

(1) Upper bound.

(2) At external column line; includes static settlement of abouI 0,03 in.

F·78


App


Table 8. Nonlinear Response: ComparatIve Results (contInued)

Wall shear (kips) Wall moment (kip·rt) Wall axial compression (kips) Base shear (kips) Foundation axial rorce (1) (kips) Lateral roor deftection (in) Downward displacement (2) (in) Uplift (2) (in) (I)

.HI.3

PD.3

HI.10

HI.1I

HI.12

PD.12

-H1.13

PD.13

481 4,754

367 3,499

525 5,298

539 6,364

371 3,605

410 4,012

458 4,617

359 3,525

901

800

864

1,611

685

785

925

806

1,223 1,911

839 1,591

1,297 1,907

1,536 3,158

887 1,570

976 1,761

1,134 1,927

821 1,618

095

0·46

0·94

2-29

0-48

0'65

0·84

0-44

016

0-12

0-16

0'31

0'12

0'15

015

012

040

010

035

1-21

011

023

031

010

@

~

Upper bound. lin~

(2) At eXlernat column

@

@l

,0

includes slatic scnlement or about 0-03 in.

2,

20

IIT®

@

@ 22

@

21

@ @

@

(Bl

32

3S

19 '\.. CO&.U"'''' JoIU"I[It

e

34

'''liST FLOO" ONL.Y

e

.,

••

" @

@ I!§I '0

@

'\..I:OLU ... '" "u"'SEA

11.9

•

407 .042 ,004

~

<

..;

~

U

CJ

0

~

"'" 1

,139 ~112

'24

0 •• ® .@

11

0lC>~ 2 111,2 '@@

@

,.

41

IS

~

n

0

I~

@@@@

X @

WlE ....

S

"

@

0

••

cgJD ••

0

'0 ®

.

@

® @@ 2•

~

@

"

Iii!

0

:l

,",U"SEIt

0

@

@

0

@ ID @

® ••

so

®31

.

®@

@

52

V·

<:;;>

$I

®

®

®

.0

@

®

~

•• ••

Figure 1. TypIcal Floor Level and Plan oF Assumed Framing System

!ets


F·79


fl

-

DO DO DO

D0 D0 D0 D0

0 0 0 0 0 0

0 0 0 0

-A-

-for.

11

0 0

00

wALL 8

WALL A

0 0 0 0 WALL

e

Figure 2. Slmplllled COupled Shear Wall Model

F

A

!

e

1

mill I -IOks!

l--~-rw

IFill AREA IN CONTACT WITH SOIL

SECTION A- A

Figure S. Equilibrium at Incipient overturning, static AnalysIs

F-8D


App


.. II.~ C4-

C3

o---,c

!.2~ ~.';J.ll

9.2~

-..:o'lsbl

C6

I !

C6

C6

C'

C'

r,t:: sir ,

I

2

I

'"

"q

C3

C6

0-'"

CO

C.

zero dlstonce

"q

C5

: -'" C/

?'

sir ......... &,

C4

I-

S:2 C,

"-

u

"q

'-HUNG" 'STRUCTURE DETAIL

'"

,-:U.J&,

DETAIL 2

Figure 4. schematic Representation of Nonlinear Model

z

o

u;

...z ~

8 SPRING I

z

0

u;

... Z

~

8

(0)

SPRING 2

z

i

a

(b)

I.S II.O,O.S)

0

u;

... Z

a

~

COMBINED ACTION

Figure 5. fal TrIlinear Soli Behavior Modelled WIth TWO parallel BilInear SprIngs; fbI BilInear BehavIor Of Horizontal Soli Spring

fects


F-B1


Resource Summary ,& Veletsos, A.S., Prasad, A.M., and Tang, Y. Design Approaches for Soil-Structure Interaction Proceedings of the Ninth World Conference on Earthquake Engineering, Tokyo, August, 1988 In this state-of-the-art report, soil structure interaction concep~ expressed in terms of kinematic and inertial interaction are discussed with reference to a simple linear structure of mass, m, and height, h, supported ona rigid mat foundation of mass, mo. at the surface of a homogeneous elastic half space. The authors note that seismic response of a structure is frequently evaluated considering the motion of its base to be equal to the stipulated free-field ground motion at a reference or control point, normally taken at the ground surface. However, an exact analysis requires that the structure be considered to be part of a larger system which includes the foundation and a supporting medium and that spatial variability of ground motion and properties of soils need to be considered. For such an analysis a two step approach is defined. First, the motion of the foundation is evaluated considering the foundation and the structure to be massless. The resulting foundation input motion (FIM) generally includes torsional and rocking components in addition to translational components. Next the response to the FIM of the actual foundation-structure system with mass is evaluated using actual properties of a supporting medium and providing for the dynamic interaction. The difference in the response of a superstructure computed for the FIM and the freefield control point motions represents kinematic interaction effect. The difference of the responses computed with and without regard for the flexibility of the supporting medium is known as the inertial interaction effect. The total soil-

F-82

structure interaction is given by the sum of both effects. For the simplified structure considered, the effects of inertial and kinematic interaction are examined in the paper. Expressions are given for increases in the natural period of the structure on a flexible support system and for the effects of foundation damping, including radiation damping arising from foundation interaction. To address the question as to whether soil-structure interaction increases or decreases the maximum response of the structure, a number of cases are considered where it becomes clear that the answer is a function of the response quantity under examination and the characteristics of the ground motion and the system itself. Kinematic interaction effects are discussed including the effects of wave passage and ground motion incoherence. The relative importance of kinematic and inertial interaction is examined using response spectra plots, such as those shown in Figure I, where the ratios of pseudo spectral velocity divided by peak ground acceleration are shown as a function of the natural frequency of the structure. Because the presented material and analysis results assume a simplified idealized building and a rigid mat foundation system resting on a semiinfinite elastic medium, it is difficult to draw specific quantitative conclusions regarding the significance if kinematic and internal interaction effects for actual building structures, where more complex foundation systems also involve nonlinear or inelastic soil response to earthquake ground motions. However, it is clear that for many


b

iJ iJ II

s: n e'

k 1< SI

ri bi

Ap

9

-


-

building systems, the effects of kinematic interaction are small in relation to inertial interaction effects and that inertial interaction has the greatest influence on the response of structural systems in the medium and high frequency spectral regions. The authors note that inertial interaction effects are generally more important than that of kinematic interaction and the interaction effects for low frequency highly compliant structures are negligible because such systems "see" the supporting half space as a very stiff effectively rigid medium. The above research and conclusions form the . basis for existing NEHRP Guidelines (for the

seismic design of buildings) on the effects of soilstructure interaction. However, the authors note the limitation of these studies and the approach used, and that additional research is needed to evaluate the behavior of structures for embedded foundation and pile foundation systems. Additionally interaction effects for structures responding in a nonlinear range of deformation are recommended. Clearly, the behavior of individual foundation elements supporting column loads under earthquake loading and nonlinear response can be major factors in determining structural loads, and more research in these areas is needed to formulate improved guidelines.

)r

lOa

19 the

ld tion lye

2.--------------------;

5

he ak the

v,

x.

~

v,

1;, ---No 551

nd i-

0.1

0.05

0.1

·_·····Klonly --Total 551 0.1

, f •• CPS

10

30

"

I

10

30

fa' cps

n

Ire lear

ects

Figure 1_ Response Spectra For Interacting structures with nlR = 2; ObliquelY Incident EI Centro Record, sin a=0_4, 1:=0_05 sec.


F·BS


Resource Summary 17 Wallace, J.W., Moehle, J.P., and Martinez-Cruzado, J. Implications for the Design of Shear Wall Buildings Using Data from Recent Earthquakes Proceedings of Fourth U.S. National Conference on Earthquake Engineering, Palm Springs, CA, May 1990 This paper presents the results of a study of the measured response and observed seismic performance of reinforced concrete load bearing shear wall buildings during past earthquakes. First, the response of two 10-story concrete shear wall buildings in California during earthquakes was analyzed. Both buildings were designed and built in early 1970s. Figure 1 shows the plan configurations of the two buildings. Both buildings have been instrumented with strong motion instruments by the California Strong Motion Instrumentation Program. Because of the symmetry of the first building, a 2-dimensional computer using SAP-90 computer model was developed; whereas, a three dimensional model of the second building was developed using ETABS computer program. The first building, located in northern California, was analyzed for the motions recorded during the 1984 Morgan Hill (Ms = 6.2) earthquake. The second building, located in southern California was analyzed for the motions recorded during the 1987 Whittier (Ms = 5.9) earthquake. It was observed that when soil flexibility and the effect of initial concrete cracking were ignored in the analysis, the periods of vibration obtained from the computer model were smaller than those measured from recorded response. The correlation between the measured and computed time-history response was also poor, Figure 2. Therefore, the computer model was modified to incorporate these effects. Foundation flexibility was modeled using the relatively simple soil springs approach outlined

F-B4

in the ATC-3-06 document. As shown in Figure 3, good correlation between the recorded motions and the time-history analyses results was observed when the effects of foundation flexibility and cracked section properties were included. Comparison were then made with the response of shear wall buildings during the March 1985 Chilean earthquake (Ms = 7.8) which had a peak acceleration of 0.36 g and duration of strong shaking of more than 60 seconds. Even though the Chilean buildings are designed to similar force levels as California, their ductile detailing and quality control requirements are quite lax by US standards. However, these buildings have performed remarkably well during past earthquakes. To understand the reasons for this good performance, displacement ductility demand spectra were developed from 5 percent damped elastic spectra using the method recommended by Newmark and Hall. These are shown in Figure 4. The calculated base shear strengths and the periods measured from ambient vibrations were then plotted in this figure. It was observed that these data points correspond to a ductility demand of about 3 which corresponds to appreciable local damage. However, when the building strength was plotted as a function of the building period considering soil flexibility, it was noted that several buildings fell on or above the elastic spectra and the largest ductility demand was 2. It was, therefore, concluded that foundation

APpendix F, supplemental Information on Foundation Effects

f d

o a \1

t~

z

All


flexibility was one of the reasons for the reduced displacement ductility demands in these buildings. The authors also studied the available ductility of the Chilean shear walls. They found that, the available curvature ductility of lightly reinforced walls was about 10 to 12 and was greater than the typical ductility demand of about 3 to 5.

3, nd

...... .,..

Therefore, it was concluded that for rectangular, symmetrically reinforced walls, concrete confinement in transverse reinforcement is not necessary for good seismic performance and that the US bearing wall buildings should also perform well during a major earthquake.

1---------,. ,..... ---------ol

se k

(&) Building 1 - Floorl 2·5

· iii LIJ-+'

~

-t-'

y t Jds

l·'r l

las

1.1£'"

• -=tv.

(b) Building 2 - Floors 2·10

[t

Figure 1. Building Plans

!ctl


F·BS


i

1.0

;. , i

I_--~~~~~~~~~~~~~~~~~~~~

0.0 ...

1u1l4f"'9 1 . CIIoMoiI 7-Roof !AWl (NS) -1.04---~--~--~--~--~--~--~--~--~--__~--~~

i '.0

I .i

0.0

0.0

5.0

10.0

15.0

20.0

-

25.0

30.0

llEASURED

• - - - COWM'ED

-+---~~~~~'.lJMoI,A.I.+J.l....p..;.l,~~_..........

""toOo.._a__'

Build"', 2

lloof UveI (EW)

-1.04-----~----~--~----~----~----~----~--~ 0.0 5.0 10.0 15.0

20.0

TIME (SEC) FIgure 2. Measured and Computed RDDI Dlsplacement-cross,sectlon Model

~ 1.0

~....

IoIWUIIED • - - - COWPIITED

0.0 -l---"""'AA-A-AN~\.A+\A-I+A-AA~~~p,..j:~tH.w:\-JloL\-H,II:H~+\'f1A1WIn

~III

Buildin9 1 CIlO.....1 7-Roofl..VlI (NS)

is - 1.0 ~-....--....--....--...---....--..--..---..---..---,....,-....,....--l

0.0

5.0

10.0

15.0

20.0

25.0

~o.o

IoIEASURED • - - - COIolPUTtD

20.0

FIgure 6. Measure and Computed RooF Dlsplacement-Cracked·Sectlon and SSI Model

F·8a



T-;:;;;;:;;::=:=:---==:::-=-::~~;----I lastic Vino del Mar - S20W

0.75

o o

Measured Periods Computed PeriodS -

Flexible Bo~e Plaza del Mar Acapulco Festival Miramar Torres del Sol EI Foro Almendral Villa Real

1

2 J 4

5 6

7

6

l-....---,.-....,....-:-::::=:;~~~~~~~;;~ -r----------=::--:-:-:::'--:----------,

0.00

0.75

Elastic

-

o

o •

~

ATC Soil Type 2 Fixed Base Flexible Bose California Buildings

1

Plaza del Mar

!0.50

2 - Acapulco J restival

ai

5

Miramar Torres del Sol 6 EI Foro 7 - Almendral 8 - Villa Real 4

Lo.I

VI

~ 0.25

rn o.oo+-~....---,.-....,....-~-...._~--~-~-....---,.-....,....~

0.00

1.00

2.00

J.OO

PERIOD (SEC)

Figure 4. Comparison oF Ductility Demands

as

Appendix F. Supplemental Information an Faundation Effects

F-87


Resource Summary 18 Yim, S. C-S and Chopra, A. K. Simplified Earthquake Analysis of Multistory Structures with Foundation Uplift American Society of Civil Engineers, Journal of Structural Engineering, Vol. 111, No. 12, December 1985. pp. 2708-2731 This investigation is an extension of the approach for the analysis of single degree of freedom systems (Chopra and Yim) to the analysis of IlUIltistory structures with foundation flexibility and uplift. Figure 1 shows the system considered in this investigation. Although, the foundation soil can be represented by the two spring -dashpot model or the distributed Winkler spring and dashpot mode, only the results of analyses with two spring and dashpot system were presented in this paper. Equations of motions were developed for this multistory structure foundation system. It was noted that although the governing equations of motions were nonlinear, the dynamic response of the system under consideration could be obtained as sequential response of three linear systems corresponding to three contact conditions for the foundation mat. The time history response of an idealized 10story structure was analyzed for the north-south component of the 1940 EI Centro earthquake. The idealized structure had uniform stiffness, mass, and inter-story height. The damping was 5 percent in all modes of vibration and the slenderness ratio for the first mode of vibration was 10. The base shear and overturning moment spectra obtained from this analysis are shown in Figures 2 and 3. The beneficial effect of allowing uplift were observed in terms of lowered base shear and overturning moments for periods of vibration less than about 1 sec. The authors also observed that foundation flexibility and uplift have little impact on the higher

F-88

modes of response. Therefore, they concluded that the analysis of a multistory building with foundation flexibility and uplift can be simplified by calculating only the fundamental mode of response considering these effects; the contribution from higher modes can be computed using standard procedures by disregarding the effects of foundation flexibility and uplift. Excellent correlation between the exact solutions and the solutions obtained using this approach was observed, see Figure 4. A simplified approach for estimating the response of uplifting multistory structures was then presented. In this approach, the maximum earthquake base shear for the system is obtained as the SRSS of the modal maxima of individual modes of vibrations. An approximate expression for the maximum base shear in the first mode of vibration considering foundation flexibility and uplift was developed. This equation is analogous to the one developed for the single degree of freedom systems and does not require time history analyses. The modal maxima for the other mod~s of vibration are obtained using the standard procedures for the structure on rigid foundation without uplift. The base shear response spectra using this simplified approach as presented in Figure 5 show reasonable degree of accuracy. Therefore, the simplified approach can be used for practical design or for performing parametric studies on the influence of foundation flexibility and uplift on response of multistory structures.


A

SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUlLo::IIGS

I I I I I I I I I

1 1 1 1 1

r

...

I.,. ",

1,/

at

1=11

II F=:::::-':=::::::d

II

"

on ,f

"

• -ilt Figure 1. Multistory structure supported on Two Spring-Damper Element Foundation

as

to

,m s

or

:ts


F-8.

SE~::r;",IC

EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

2

. ...\......

I

o.a

~

0.1

..,',

\

CIA

• ~ '),.0

.. Z

-

III

"

t

II:

0.04

8

. ~

•~

0.2

0.1 0.08 O.Ge

u

,

ClIITlCAL lASE SHEAI!

CQUI'ICIENT V.

C

., III % III

~

0.02

CD

0.01 0.008 0.001 -·· ..-ItIOlD SUHO«TJJeG

QOO4

"-U'I"E

S~ItTING

.,n'''. ""-In "lIIrvlNTlD

ITSTtM

-

.....'.C.T6C'I'.U.... ''' ... h'.TlD

-

____ DIDe.'..cT." ....I"

.'''''''D

0.002

0.1

0.2

O.~

0.eO.8 I

2

4

6

8 10

20

FUNDAMENTAL VIBRATION PERIOD TI ' SEC.

FIgure 2. Base Shear Response spectra for MultIstory structures Subjected to EI centro Ground MotIon for Three SupportIng CondItIons

F·80


FI

AI


2r----------------------------------, • Z

I 0.1 0.1

~.

0.4

iu

0.2

it11.1 o

U

0.1 0.08

~

0.06

....

I

OIIERTUMING _ENT ~-----------=~~::~::~==~~~~~~'T~~~A~L~.~A~S~E~~~ \ CO£FFICIEIIT MC \

\, \,,

0.04

, ~,

CIJ

z

~

"" 0.02

...

i

t

\

~ ODI 00.001

=Co.ooa ~

..............0 ",'IIOtIITINQ STIlIM, Ufl\.1'T . .IY'NTE.D

0.004

nIJUI"" ~TING snTIM

UP",,, "1'41""0

-

IC*OID CowrAC',

-

u.a.DIDCOtI'..CT.~I"pl_I"U

0.002

0.1

0.2

0.4 0.60.' I

•

2

4

I

I 10

20

FUNDAMENTAL. VIBRATION PERIOD T, • SEC.

·r

Figure $. Base Overturning Moment Response Spectra for Multistory structures Subjected to EI centro Cround MotIon for Three support conditions

AppenCllx F, Supplemental Information on Foundation Effects

F·91


2~--------------------------~----~ I

o.a

o.

.. .

~

.... E

0.2

:::..

z

-it '"

0.1 O.oa

~

Q.06

U

II: C

CI.O.f

'":l

o.oz

'"en:I:

CRITICAL lASE SHEAR

COUFICIENT 1',

III

0.01

0.001 0.001

_ D CCIIT&CT.IW\.I'T ,.,ytNTED I -

IlAe, lO,-ut,aM

I ___ ....... I. .".~UtlO.

0.004

UOO_DlD CDIIT&CT.UPL"r PPII'TT'D I -

,lAC' "''''"w

.. - - - - . . . . . . 111&" IOUllIO.

0.1

0.2

0.4 0.60.8 I

2

4

6

8 10

20

FUNDAMENTAL VIBRATION PERIOD TI • SEC.

Figure 4. comparison Of Base Shear Response spectra computed by Exact and Approximate Analysis procedures lor Multistory Buildings

F·92



2~-----------------------------------' I 0.8

0 .•

.

0.4

a

E

-:,.0

0.2

... Z

w C3

ifw

CRITICAL USE SHEAII

0.1 0.08

CO£FFICIENT

V.

0.06

8

a: 0.04 c w %

(II

0.02 w (II c III

0.01 0.008

IONO!D co.. rACT. UP,",' r PREVENTED

0.006

I

-

2 ---

0.004

'UCT SOWTIO., 'JIOII'~I'IIDMOO"&. ..... I.'"I'.oelOU".

UNIONOED CONTACT .UPLIFT '1"M.TT£D

S •

cueT SOLUTIO .. - - - - "",I*,,-I"'D MOOM. ..... 1.'111 P'ftOC&DUItI

0.002

0.1

0.2

0.4 0.60.8 I

2

4

6

8 10

20

FUNDAMENTAL VIBRATION PERIOD TI • SEC.

Figure S. Comparison oF Base Shear Response Spectra compUted by EXact and Simplified Modal Analysis procedures for Multistory Buildings

:ts



AppendixG

Applied Technology Council proJects and Report Information One of the primary purposes of Applied Technology Council is to develop resource documents that translate and summarize useful infonnation to practicing engineers. This includes the development of guidelines and manuals, as well as the development of research recommendations for specific areas determined by the profession. ATC is not a code development organization, although several of the ATC project reports serve as resource documents for the development of codes, standards and specifications. Applied Technology Council conducts projects that meet the following criteria: 1. The primary audience or benefactor is the design practitioner in structural engineering. 2. A cross section or consensus of engineering opinion is required to be obtained and presented by a neutral source. 3. The project fosters the advancement of structural engineering practice. A brief description of several major completed projects and reports is given in the following section. Funding for projects is obtained from government agencies and tax-deductible contributions from the private sector. ATC-I: This project resulted in five papers that were published as part of Building Practices for

Disaster Mitigation, Building Science Series 46, proceedings of a workshop sponsored by the National Science Foundation (NSF) and the National Bureau of Standards (NBS). Available through the National Technical Information Service (NTIS), 5285 Port Royal Road, Springfield, VA 22151, as NTIS report No. COM-73-50188. ATC-2: The report, An Evaluation of a Response Spectrum Approach to Seismic Design of Buildings, was funded by NSF and NBS and was conducted as part of the Cooperative Federal Program in Building Practices for Disaster Mitigation. Available through the ATC office. (Published 1974,270 Pages)

ABSTRACT: This study evaluated the applicability and cost of the response spectrum approach to seismic analysis and design that was proposed by various segments of the engineering profession. Specific building designs, design procedures and parameter values were evaluated for future application. Eleven existing buildings of varying dimensions were redesigned according to the procedures. A TC-3: The report, Tentative Provisions for the Development of Seismic Regulations for Buildings (ATC-3-06), was funded by NSF and NBS. The second printing of this report, which includes proposed amendments, is available through the ATC office. (Published 1978, amended 1982, 505 pages plus proposed amendments)

Appendix C, APplied Technology Council projects and Report Information

G-1


ABSTRACT: The tentative provisions in this document represent the results of a concerted effort by a multi-disciplinary team of 85 nationally recognized experts in earthquake engineering. The provisions serve as the basis for the seismic provisions of the 1988 Uniform Building Code and the 1988 and subsequent issues of the NEHRP Recommended Provisions for the Development of Seismic Regulation for New Buildings. The second printing of this document contains proposed amendments prepared by a joint committee of the Building Seismic Safety Council (BSSC) and the NBS.

ATC-3-2: The project, Comparative Test Designs of Buildings Using ATC-3-06 Tentative Provisions, was funded by NSF. The project consisted of a study to develop and plan a program for making comparative test designs of the ATC-3-06 Tentative Provisions. The project report was written to be used by the Building Seismic Safety Council in its refinement of the ATC-3-06 Tentative Provisions. ATC-3-4: The report, Redesign of Three Multistory Buildings: A Comparison Using ATC-3-06 and 1982 Uniform Building Code Design Provisions, was published under a grant from NSF. Available through the ATC office. (Published 1984, 112 pages) ABSTRACT: This report evaluates the cost and

technical impact of using the 1978 ATC-3-06 report, Tentative Provisions for the Development of Seismic Regulationsfor Buildings, as amended by a joint committee of the Building Seismic Safety Council and the National Bureau of Standards in 1982. The evaluations are based on studies of three existing California buildings redesigned in accordance with the ATC-3-06 Tentative Provisions and the 1982 Uniform Building Code. Included in the report are recommendations to code implementing bodies. ATC-3-5: This project, Assistance for First Phase of ATC-3-06 Trial Design Program Being Conducted by the Building Seismic Safety Council, was funded by the Building Seismic Safety Council

G·2

to provide the services of the ATC Senior Consultant and other ATC personnel to assist the BSSC in the conduct of the first phase of its Trial Design Program. The first phase provided for trial designs conducted for buildings in Los Angeles, Seattle, Phoenix, and Memphis. A TC-3-6: This project, Assistance for Second Phase of ATC-3-06 Trial Design Program Being Conducted by the Building Seismic Safety Council, was funded by the Building Seismic Safety Council to provide the services of"the ATC Senior Consultant and other ATC personnel to assist the BSSC in the conduct of the second phase of its Trial Design Program. The second phase provided for trial designs conducted for buildings in New York, Chicago, St. Louis, Charleston, and Fort Worth. ATC-4: The report, A Methodology for Seismic Design and Construction of Single-Family Dwellings, was published under a contract with the Department of Housing and Urban Development (HUD). Available through the ATC office. (Published 1976,576 pages) ABSTRACT: This report presents the results of an in-depth effort to develop design and construction details for single-family residences that minimize the potential economic loss and life-loss risk associated with earthquakes. The report: (1) discusses the ways structures behave when subjected to seismic forces, (2) sets forth suggested design criteria for conventional layouts of dwellings constructed with conventional materials, (3) presents construction details that do not require the designer to perform analytical calculations, (4) suggests procedures for efficient plan-checking, and (5) presents recommendations including details and schedules for use in the field by construction personnel and building inspectors.

ATC-4-1: The report, The Home Builders Guide for Earthquake Design, was published under a contract with HUD. Available through the ATe office. (Published 1980, 57 pages)

Appendix C, Applied Technology Council projects and Report Information

J

,

1

c c

A

}j

w

A 2


ABS1RACT: This report is a 57-page abridged

version of the ATC-4 report. The concise, easily understood text of the Guide is supplemented with illustrations and 46 construction details. The details are provided to ensure that houses contain structural features that are properly positioned, dimensioned and constructed to resist earthquake forces. A brief description is included on how earthquake forces impact on houses and some precautionary constraints are given with respect to site selection and architectural designs.

I, il

A TC·5: The report, Guidelines for Seismic Design and Construction of Single-Story Masonry Dwellings in Seismic Zone 2, was developed under a contract with HUD. Available through the ATC office. (Published 1986,38 pages)

al

:s

m

d

n

ABS1RACT: The report offers a concise methodology for the earthquake design and construction of single-story masonry dwellings in Seismic Zone 2 of the United States, as defined by the 1973 Uniform Building Code. The Guidelines are based in part on shaking table tests of masonry construction conducted at the University of California at Berkeley Earthquake Engineering Research Center. The report is written in simple language and includes basic house plans, wall evaluations, detail drawings, and material specifications.

A TC·6: The report, Seismic Design Guidelines for Highway Bridges, was published under a contract with the Federal Highway Administration (FHWA). Available through the ATC office. (Published 1981, 210 pages) ABS1RACT: The Guidelines are the

recommendations of a team of sixteen nationally recognized experts that included consulting engineers, academics, state and federal agency representatives from throughout the United States. The Guidelines embody several new concepts that were significant departures from then existing design provisions. Included in the Guidelines are an extensive commentary, an

example demonstrating the use of the Guidelines, and summary reports on 21 bridges redesigned in accordance with the Guidelines. The guidelines have been adopted by the American Association of Highway and Transportation Officials as a guide specification. ATC·6·1: The report, Proceedings of a Workshop on Earthquake Resistance of Highway Bridges, was published under a grant from NSF. Available through the ATC office. (Published 1979,625 pages) ABS1RACT: The report includes 23 state-of-the-

art and state-of-practice papers on earthquake resistance of highway bridges. Seven of the twenty-three papers were authored by participants from Japan, New Zealand and Portugal. The Proceedings also contain recommendations for future research that were developed by the 45 workshop participants. A TC·6·2: The report, Seismic Retrofitting Guidelines for Highway Bridges, was published under a contract with FHWA. Available through the ATC office. (Published 1983,220 pages) ABS1RACT: The Guidelines are the

recommendations of a team of thirteen nationally recognized experts that included consulting engineers, academics, state highway engineers, and federal agency representatives. The Guidelines, applicable for use in all parts of the U.S., include a preliminary screening procedure, methods for evaluating an existing bridge in detail, and potential retrofitting measures for the most common seismic deficiencies. Also included are special design requirements for various retrofitting measures. ATC·7: The report, Guidelines for the Design of Horizontal Wood Diaphragms, was published under a grant from NSF. Available through the ATC office. (Published 1981, 190 pages) ABS1RACT: Guidelines are presented for

designing roof and floor systems so these can

Appendix G, Applied Technology Council proJects and Report Information

GoJ


function as horizontal diaphragms in a lateral . force resisting system. Analytical procedures, connection details and design examples are included in the Guidelines. ATC-7-1: The report, Proceedings of a Workslwp of Design of Horizontal Wood Diaphragms, was published under a grant from NSF. Available through the ATC office. (Published 1980, 302 pages)

ABSTRACT: The report includes seven papers on state-of-the-practice and two papers on recent research. Also included are recommendations for future research that were developed by the 35 workshop participants. ATC-8: This report, Proceedings of a Workshop on the Design of Prefabricated Concrete Buildings for Earthquake Loads, was funded by NSF. Available through the ATC office. (Published 1981, 400 pages)

ABSTRACT: The report includes eighteen stateof-the-art papers and six summary papers. Also included are recommendations for future research that were developed by the 43 workshop partiCipants. ATC-9: The report, An Evaluation of the Imperial County Services Building Earthquake Response and Associated Damage, was published under a grant from NSF. Available through the ATC office. (Published 1984, 231 pages)

ABSTRACT: The report presents the results of an in-depth evaluation of the Imperial County Services Building, a 6-story reinforced concrete frame and shear wall building severely damaged by the October 15, 1979 Imperial Valley, California, earthquake. The report contains a review and evaluation of earthquake damage to the building; a review and evaluation of the seismic design; a comparison of the requirements of various building codes as they relate to the building; and conclusions and recommendations pertaining to future building code provisions and future research needs.

G-4

ATC-IO: This report, An Investigation of the Correlation Between Earthquake Ground Motion and Building Peiformonce, was funded by the U.S. Geological Survey (USGS). Available through the ATC office. (Published 1982, 114 pages)

ABSTRACT: The report contains an in-depth analytical evaluation of the ultimate or limit capacity of selected representative building framing types, a discussion of the factors affecting the seismic performance of buildings, and a summary and comparison of seismic design and seismic risk parameters currently in widespread use.

A

a /;

l' 1

ATC-IO-I: This report, Critical Aspects of Earthquake Ground Motion and Building Damage Potential, was co-funded by the USGS and the NSF. Available through the ATC office. (Published 1984, 259 pages)

ABSTRACT: This document contains 19 stateof-the-art papers on ground motion, structural response, and structural design issues presented by prominent engineers and earth scientists in an ATC seminar. The main theme of the papers is to identify the critical aspects of ground motion and building performance that currently are not being considered in building design. The report also contains conclusions and recommendations of working groups convened after the Seminar.

A J,

R u

A

ATC-ll: The report, Seismic Resistance of Reinforced Concrete Shear Walls and Frame Joints: Implications of Recent Research for Design Engineers, was published under a grant from NSF. Available through the ATC office. (Published 1983, 184 pages)

ABSTRACT: This document presents the results of an in-depth review and synthesis of research reports pertaining to cyclic loading of reinforced concrete shear walls and cyclic loading of joint reinforced concrete frames. More than 125 research reports published since 1971 are reviewed and evaluated in this report. The

Appendix C, Applied Technology Council projects and Report Information

A


preparation of the report included a consensus process involving numerous experienced design professionals from throughout the United States. The report contains reviews of current and past design practices, summaries of research developments, and in-depth discussions of design implications of recent research results. A TC-12: This report, Comparison of United States and New Zealand Seismic Design Practices for Highway Bridges, was published under a grant from NSF. Available through the ATC office. (Published 1982,270 pages) ABSTRACT: The report contains summaries of

all aspects and innovative design procedures used in New Zealand as well as comparison of United States and New Zealand design practice. Also included are research recommendations developed at a 3-day workshop in New Zealand attended by 16 U.S. and 35 New Zealand bridge design engineers and researchers.

F. ~,

ATC-12-1: This report, Proceedings of Second Joint U.S.-New Zealand Workshop on Seismic Resistance of Highway Bridges, was published under a grant from NSF. Available through the ATC office. (Published 1986,272 pages)

d rs y

ABSTRACT: This report contains written

versions of the papers presented at this 1985 Workshop as well as a list and prioritization of workshop recommendations. Included are summaries of research projects being coriducted in both countries as well as state-of-the-practice papers on various aspects of design practice. Topics discussed include bridge design philosophy and loadings; design of columns, footings, piles, abutments and retaining structures; geotechnical aspects of foundation design; seismic analysis techniques; seismic retrofitting; case studies using base isolation; strong-motion data acquisition and interpretation; and testing of bridge components and bridge systems.

i

ts:

3, ts h :ed 1t

on

I

ATC-13: The report, Eanhquake Damage Evaluation Datafor California, was developed under a contract with the Federal Emergency Management Agency (FEMA). Available through the ATC office. (Published 1985,492 pages) ABSTRACT: This report presents expert-opinion

earthquake damage and loss estimates for industrial, commercial, residential, utility and transportation facilities in California. Included are damage probability matrices for 78 classes of structures and estimates of time required to restore damaged facilities to pre-earthquake usability. The report also describes the inventory information essential for estimating economic losses and the methodology used to develop loss estimates on a regional basis. ATC-14: The report, Evaluating the Seismic Resistance of Existing Buildings, was developed under a grant from the NSF. Available through the ATC office. (Published 1987, 370 pages) ABSTRACT: This report, written for practicing

structural engineers, describes a methodology for performing preliminary and detailed building seismic evaluations. The report contains a state-of-practice review; seismic loading criteria; data collection procedures; a detailed description of the building classification system; preliminary and detailed analysis procedures; and example case studies, including nonstructural considerations. ATC-IS: The report, Comparison of Seismic Design Practices in the United States and Japan, was published under a grant from NSF. Available through the ATC office. (Published 1984, 317 pages) ABSTRACT: The report contains detailed technical papers describing design practices in the United States and Japan as well as recommendations emanating from a joint U.S.Japan workshop held in Hawaii in March, 1984. Included are detailed descriptions of new seismic design methods for buildings in Japan

Appendix G, APplied Technology Council projects and Report Information

c-s


and case studies of the design of specific buildings (in both countries). The report also contains an overview of the history and objectives of the Japan Structural Consultants Association. ATC·15·1: The report, Proceedings of Second U.S.Japan Workshop on Improvement of Building Seismic Design and Construction Practices, was published under a grant from NSF. Available through the ATC office. (Published 1987,412 pages) ABS1RACT: This report contains 23 technical

papers presented at this San Francisco workshop in August, 1986, by practitioners and researchers from the U.S. and Japan. Included are state-of-the-practice papers and case studies of actual building designs and information on regulatory, contractual, and licensing issues. ATC·lS·2: The report, Proceedings of Third U.S.Japan Workshop on Improvement of Building Structural Design and Construction Practices, was published jointly by ATC and the Japan Structural Consultants Association. Available through the ATC office. (Published 1989,358 pages) ABS1RACT: This report contains 21 technical papers presented at this Tokyo, Japan, workshop in July, 1988, by practitioners and researchers from the U.S., Japan, China, and New Zealand. Included are state-of-the-practice papers on various topics, including braced steel frame buildings, beam-column joints in reinforced concrete buildings, summaries of comparative U. S. and Japanese design, and base isolation and passive energy dissipation devices.

ATC·lS·3: The report, Proceedings of Fourth U.S.Japan Workshop on Improvement of Building Structural Design and Construction Practices, was published jointly by ATC and the Japan Structural Consultants Association. Available through the ATC office. (Published 1992,484 pages) ABS1RACT: This report contains 22 technical

papers presented at this Kailua-Kona, Hawaii,

G-G

workshop in August, 1990, by practitioners and researchers from the United States, Japan, and Peru. Included are papers on postearthquake building damage assessment; acceptable earthquake damage; repair and retrofit of earthquake damaged buildings; base-isolated buildings, including Architectural Institute of Japan recommendations for design; active damping systems; wind-resistant design; and summaries of working group conclusions and recommendations. ATC·15·4: The report, Proceedings of Fifth U.S.Japan Workshop on Improvement of Building Structural Design and Construction Practices, was published jointly by ATC and the Japan Structural Consultants Association. Available through the ATC office. (Published 1994, 360 pages) ABs1RACT: This report contains 20

technical papers presented at this San Diego, California workshop in September, 1992. Included are papers on performance goals/acceptable damage in seismic design; seismic design procedures and case studies; construction influences on design; seismic isolation and passive energy dissipation; design of irregular structures; seismic evaluation, repair and upgrading; quality control for design and construction; and summaries of working group discussions and recommendations. ATC·16: This project, Development of a 5-Year Plan for Reducing the Earthquake Hazards Posed by Existing Nonfederal Buildings, was funded by FEMA and was conducted by a joint venture of ATC, the Building Seismic Safety Council and the Earthquake Engineering Research Institute. The project involved a workshop in Phoenix, Arizona, where approximately 50 earthquake specialists met to identify the major tasks and goals for reducing the earthquake hazards posed by existing nonfederal buildings nationwide. The plan was developed on the basis of nine issue papers presented at the workshop and workshop working group discussions. The Workshop Proceedings and Five-Year Plan are available through the Federal Emergency

Appendix C, Applied Technology Council ProJects and Report Information

o

a fi (I

A M N

(I


,d

Management Agency, 500 "C" Street, S.W., Washington, DC 20472.

:e

ATC-17: This report, Proceedings ofa Seminar and Workshop on Base Isolation and Passive Energy Dissipation, was published under a grant from NSF. Available through the ATe office. (Published 1986,478 pages)

s

.s I

ABSTRACT: The report contains 42 papers describing the state-of-the-art and state-of-thepractice in base-isolation and passive energydissipation technology. Included are papers describing case studies in the United States, applications and developments worldwide, recent innovations in technology development, and structural and ground motion issues. Also included is a proposed 5-year research agenda that addresses the following specific issues: (1) strong ground motion; (2) design criteria; (3) materials, quality control, and long-term reliability; (4) life cycle cost methodology; and (5) system response. ATC-17-1: This report, Proceedings ofa Seminar on Seismic Isolation, Passive Energy Dissipation and Active Control, was published under a grant from NSF. Available through the ATe office. (Published 1993, 841 pages)

p

by

e

et :ral n

ms. a:e

Ion

ABSTRACT: The 2-volume report documents 70 technical papers presented during a two-day seminar in San Francisco in early 1993. Included are invited theme papers and competitively selected papers on issues related to seismic isolation systems, passive energy dissipation systems, active control systems and hybrid systems. ATC-19: The report, Structural Response Modification Factors was funded by NSF and NCEER. Available through the ATe office. (Published 1995,70 pages) ABSTRACT: This report addresses structural response modification factors (R factors), which are used to reduce the seismic forces associated with elastic response to obtain design forces.

The report documents the basis for current R values, how R factors are used for seismic design in other countries, a rational means for decomposing R into key components, a framework (and methods) for evaluating the key components of R, and the research necessary to improve the reliability of engineered construction designed using R factors. A TC-20: The report, Procedures for Postearthquake Safety Evaluation ofBuildings, was developed under a contract from the Califomia Office 'of Emergency Services (OES), Califomia Office of Statewide Health Planning and Development (OSHPD) and FEMA. Available through the ATC office (Published 1989, 152 pages) ABSTRACT: This report provides procedures and guidelines for making on-the-spot evaluations and decisions regarding continued use and occupancy of earthquake damaged buildings. Written specifically for volunteer structural engineers and building inspectors, the report includes rapid and detailed evaluation procedures for inspecting buildings and posting them as "inspected" (apparently safe), "limited entry" or "unsafe". Also included are special procedures for evaluation of essential buildings (e.g., hospitals), and evaluation procedures for nonstructural elements, and geotechnical hazards. ATC-20-1: The report, Field Manual: Postearthquake Safety Evaluation of Buildings, was developed under a contract from OES and OSHPD. Available through the ATC office (Published 1989, 114 pages) ABSTRACT: This report, a companion Field Manual for the ATC-20 report, summarizes the postearthquake safety evaluation procedures in brief concise format designed for ease of use in the field. ATC-20-2: The report, Addendum to the ATC-20 Postearthquake Building Safety Procedures was published under a grant from the National Science

Appendix C, Applied Technology council projects and Report Information

G·'


Foundation and funded by the USGS. Available through the ATC office. (Published 1995.94 pages) ABSTRACT: This report provides updated assessment forms. placards. and procedures that are based on an in-depth review and evaluation of the widespread application of the ATC-20 procedures following five earthquakes occurring since the initial release of the ATC-20 report in 1989.

ATC-20-3: The report. Case Studies in Rapid Postearthquake Safety Evaluation of Buildings, was funded by ATC and R. P. Gallagher Associates. Available through the ATC office. (Published 1996, 295 pages) ABSTRACT: This report contains 53 case studies

using the ATC-20 Rapid Evaluation procedure. Each case study is illustrated with photos and describes how a building was inspected and evaluated for life safety, and includes a completed safety assessment form and placard. The report is intended to be used as a training and reference manual for building officials. building inspectors, civil and structural engineers. architects, disaster workers, and others who may be asked to perform safety evaluations after an earthquake. ATC-20-T: The report. Postearthquake Safety Evaluation of Buildings Training Manual was developed under a contract with FEMA. Available through the ATC office. (Published 1993. 177 pages; 160 slides) ABSTRACT: This training manual is intended to facilitate the presentation of the contents of the ATC-20 and ATC-20-1. The training materials consist of 160 slides of photographs. schematic drawings and textual information and a companion training presentation narrative coordinated with the slides. Topics covered include: posting system; evaluation procedures; structural basics; wood frame, masonry. concrete, and steel frame structures;

nonstructural elements; geotechnical hazards; hazardous materials; and field safety. ATC-21: The report. Rapid Visual Screening of Buildings for Potential Seismic Hazards: A Handbook. was developed under a contract from FEMA. Available through the ATC office. (Published 1988. 185 pages) ABSTRACT: This report describes a rapid visual

screening procedure for identifying those buildings that might pose serious risk of loss of life and injury. or of severe curtailment of community services. in case of a damaging earthquake. The screening procedure utilizes a methodology based on a "sidewalk survey" approach that involves identification of the primary structural load resisting system and building materials. and assignment of a basic structural hazards score and performance modification factors based on observed building characteristics. Application of the methodology identifies those buildings that are potentially hazardous and should be analyzed in more detail by a professional engineer experienced in seismic design. ATC-21-1: The report. Rapid Visual Screening of Buildings for Potential Seismic Hazards: Supporting Documentation. was developed under a contract from FEMA. Available through the ATC office. (Published 1988. 137 pages) ABSTRACT: Included in this report are (1) a

review and evaluation of existing procedures; (2) a listing of attributes considered ideal for a rapid visual screening procedures; and (3) a technical discussion of the recommended rapid visual screening procedure that is documented in the ATC-21 report.

J

I (

)

J. d

tl 1 l: Jj

ATC-21-2: The report. Earthquake Damaged Buildings: An Overview of Heavy Debris and Victim Extrication, was developed under a contract from FEMA. (Published 1988.95 pages) ABSTRACT: Included in this report, a

companion volume to the ATC-21 and

G-a

Appendix C, APplied Technology Council projects and Report Information

AI


al

ATC-21-1 reports, is state-of-the-art information on (1) the identification of those buildings that might collapse and trap victims in debris or generate debris of such a size that its handling would require special or heavy lifting equipment; (2) guidance in identifying these types of buildings, on the basis of their major exterior features, and (3) the types and life capacities of equipment required to remove the heavy portion of the debris that might result from the collapse of such buildings.

ATC-21-T: The report, Rapid Visual Screening of Buildings for Potential Seismic Hazards Training Manual was developed under a contract with FEMA. Available through the ATC office. (Published 1996, 135 pages; 120 slides)

a

ail

, a

i

:t

)n

ABSTRACT: This training manual is intended to facilitate the presentation of the contents of the ATC-21 report. The training materials consist of 120 slides and a companion training presentation narrative coordinated with the slides. Topics covered include: description of procedure, building behavior, building types, building scores, occupancy and falling hazards, and implementation.

ATC-22: The report, A Handbookfor Seismic Evaluation of Existing Buildings (Preliminary), was developed under a contract from FEMA. Available through the ATC office. (Originally published in 1989; revised by BSSC and published as the NEHRP Handbookfor Seismic Evaluation of Existing Buildings in 1992, 211 pages) ABSTRACT: This handbook provides a methodology for seismic evaluation of existing buildings of different types and occupancies in areas of different seismicity throughout the United States. The methodology, which has been field tested in several programs nationwide, utilizes the information and procedures developed for and documented in the ATC-14 report. The handbook includes checklists, diagrams, and sketches designed to assist the user.

ATC-22-1: The report, Seismic Evaluation of Existing Buildings: Supporting Documentation, was developed under a contract from FEMA. Available through the ATC office. (Published 1989, 160 pages) ABSTRACT: Included in this report, a companion volume to the ATC-22 report, are (I) a review and evaluation of existing buildings seismic evaluation methodologies; (2) results from field tests of the ATC-14 methodology; and (3) summaries of evaluations of ATC-14 conducted by the National Center for Earthquake Engineering Research (State University of New York at Buffalo) and the City of San Francisco.

ATC-23A: The report, General Acute Care Hospital Earthquake Survivability Inventory for California, Part A: Survey Description, Summary of Results, Data Analysis and Interpretation, was developed under a contract from the Office of Statewide Health Planning and Development (OSHPD), State of California. Available through the ATC office. (Published 1991,58 pages) ABSTRACT: This report, completed in 1991, summarizes results from a seismic survey of 490 California acute care hospitals. Included are a description of the survey procedures and data collected, a summary of the data, and an illustrative discussion of data analysis and interpretation that has been provided to demonstrate potential applications of the ATC23 database.

ATC-23B: The report, General Acute Care Hospital Earthquake Survivability Inventory for California, Part B: Raw Data, is a companion document to the ATC-23A Report and was developed under the same contract from OSHPD. Available through the ATC office. (Published 1991, 377 pages) ABSTRACT: Included in this report, completed in 1991, are tabulations of raw general site and

Appendix C, APplied TeChnology council proJects and Report Information

G·g


building data for 490 acute care hospitals in California. ATC·24: The report. Guidelines for Seismic Testing of Components of Steel Structures. was jointly funded by the American Iron and Steel Institute (AIS!). American Institute of Steel Construction (AISC). National Center for Earthquake Engineering Research (NCEER). and NSF. Available through the ATC office. (Published 1992. 57 pages)

ABSTRACT: This report. completed in 1992. provides guidance for most cyclic experiments on components of steel structures for the purpose of consistency in experimental procedures. The report contains recommendations and companion commentary pertaining to loading histories. presentation of test results. and other aspects of experimentation. The recommendations are written specifically for experiments with slow cyclic load application. ATC·2S: The report. Seismic Vulnerability and Impact of Disruption of Lifelines in the Conterminous United States. was developed under a contract from FEMA. Available through the ATC office. (Published 1991. 440 pages)

ABSTRACT: Documented in this report is a national overview of lifeline seismic vulnerability and impact of disruption. Lifelines considered include electric systems. water systems. transportation systems. gas and liquid fuel supply systems. and emergency service facilities (hospitals. fire and police stations). Vulnerability estimates and impacts developed are presented in terms of estimated first approximation direct damage losses and indirect economic losses. ATC·2S·1: The report. A Model Methodology for Assessment of Seismic Vulnerability and Impact of Disruption of Water Supply Systems. was developed under a contract from FEMA. Available through the ATC office. (Published 1992. 147 pages)

G-10

ABSTRACT: This report contains a practical methodology for the detailed assessment of seismic vulnerability and impact of disruption of water supply systems. The methodology has been designed for use by water system operators. Application of the methodology enables the user to develop estimates of direct damage to system components and the time required to restore damaged facilities to preearthquake usability. Suggested measures for mitigation of seismic hazards are also provided.

J

I

1

r 1

A TC·28: The report. Development of Recommended Guidelines for Seismic Strengthening of Existing Buildings. Phase I: Issues Identification and Resolution. was developed under a contract with FEMA. Available through the ATC office. (Published 1992. 150 pages)

ABSTRACT: This report identifies and provides resolutions for issues that will affect the development of guidelines for the seismic strengthening of existing buildings. Issues addressed include: implementation and format. coordination with otber efforts. legal and political. social. economic. historic buildings. research and technology. seismicity and mapping. engineering philosophy and goals. issues related to the development of specific provisions. and nonstructural element issues. ATC·29: The report. Proceedings of a Seminar and Workshop on Seismic Design and Performance of Equipment and Nonstructural Elements in Buildings and Industrial Structures. was developed under a grant from NCEER and NSF. Available through the A TC office. (Published 1992. 470 pages)

ABSTRACT: These Proceedings contain 35 papers describing state-of-the-art technical information pertaining to the seismic design and performance of equipment and nonstructural elements in buildings and industrial structures. The papers were presented at a seminar in Irvine. California in 1990. Included are papers describing current practice. codes and regulations; earthquake performance; analytical

Appendix G. Applied Technology CounCil Projects and Report Information

A f \1

IJ

f, S (I

A (

R [

A


and experimental investigations; development of new seismic qualification methods; and research, practice, and code development needs for specific elements and systems. The report also includes a summary of a proposed 5-year research agenda for NCEER. ATC-3D: The report, Proceedings of Workshop for Utilization ofResearch on Engineering and Socioeconomic Aspects of 1985 Chile and Mexico Earthquakes, was developed under a grant from NSF. Available through the ATC office. (Published 1991,113 pages)

g

ABSTRACT: This report documents the fmdings

1

of a 1990 technology transfer workshop in San Diego, California, co-sponsored by ATC and the Earthquake Engineering Research Institute. Included in the report are invited papers and working group recommendations on geotechnical issues, structural response issues, architectural and urban design considerations, emergency response planning, search and rescue, and reconstruction policy issues.

ri s

e

i

ATC-31: The report, Evaluation of the Peiformance of Seismically Retrofitted Buildings, was developed under a contract from the National Institute of Standards and Technology (NIST, formerly NBS) and funded by the U. S. Geological Survey. Available through the ATC office. (Published 1992, 75 pages) ABSTRACT: This report summarizes the results from an investigation of the effectiveness of 229 seismically retrofitted buildings, primarily unreinforced masonry and concrete tilt-up buildings. All buildings were located in the areas affected by the 1987 Whittier Narrows, California. and 1989 Loma Prieta, California, earthquakes.

ATC-32: The report,lmproved Seismic Design Criteria for California Bridges: Provisional Recommendations, was funded by the California Department of Transportation (Caltrans). Available

through the ATC office. (Published 1996,215 pages) ABSTRACT: This report provides recommended revisions to the current Caltrans Bridge Design Specifications (BDS) pertaining to seismic loading, structural response analysis, and component design. Special attention is given to design issues related to reinforced concrete components, steel components, foundations, and conventional bearings. The recommendations are based on recent research in the field of bridge seismic design and the p.erformance of Caltrans-
ATC-3S: This report, Enhancing the Transfer of U.S. Geological Survey Research Results into Engineering Practice was developed under a contract with the USGS. Available through the ATC office. (Published 1996, 120 pages) ABSTRACT: The report provides a program of

recommended "technology transfer" activities for the USGS; included are recommendations pertaining to management actions, communications with practicing engineers, and research activities to enhance development and transfer of information that is vital to engineering practice. ATC-3S-1: The report, Proceedings of Seminar on New Developments in Earthquake Ground Motion Estimation and Implications for Engineering Design Practice, was developed under a cooperative agreement with USGS. Available through the ATC office. (Published 1994, 478 pages) ABSTRACT: These Proceedings contain 22

technical papers describing state-of-the-art information on regional earthquake risk (focused on five specific regions-California, Pacific Northwest. Central United States, and northeastein North America); new techniques for estimating strong ground motions as a function of earthquake source, travel path, and site parameters; and' new developments

Appendix C, APplied Technology Council !'roJects and Report Information

C-'I1

I SEISMIC EVALUATION AND RETROFIT OF CONCRETE BUILDINGS

specifically applicable to geotechnical engineering and the seismic design of buildings and bridges. ATe-R-1: The report, Cyclic Testing o/Narrow Plywood Shear Walls, was developed with funding from the Henry J. Degenkolb Endowment Fund of the Applied Technology Council. Available through the ATC office (Published 1995,64 pages)

panels having the standard 3.5-to-l height-towidth ratio and anchored to the sill plate using typical bolted, 9-inch, 5000-lb. capacity holddown devices. The report provides a description of the testing program and a summary of results, including comparisons of drift ratios found during testing with those specified in the seismic provisions of the 1991 Uniform Building Code.

ABSTRACT: This report documents ATC's first

self-directed research program: a series of static and dynamic tests of narrow plywood wall

C-'I2

Appendix G, APplied Technology Council projects and Report Information

AI


ATC BOARD OF DIRECTORS (1973-Present)

n

s,

(1979-85) Milton A. Abel (1978-81) James C. Anderson (1988-94) Thomas G. Atkinson* (1976-77) Albert J. Blaylock (1984-88) Robert K. Burkett (1989-90) H. Patrick Campbell Arthur N. L. Chiu (1996-99 Anil Chopra (1973-74) (1976-80) Richard Christopherson* (1973) Lee H. Cliff (1986-87, 1991-97) John M. Coil* (1985-86) Eugene E. Cole (1996-99) Edwin T. Dean (1996-98) Robert G. Dean (1978-81) Edward F. Diekmann (1973-74) Burke A. Draheim (1973) John E. Droeger (1989-96) Nicholas F. Forell * (1993-97) Douglas A. Foutch (1991-92) Paul Fratessa (1986-89) Sigmund A. Freeman (1986-89) Barry J. Goodno (1984-87) Mark R. Gorman Gerald H. Haines (1981-82, 1984-85) (1985-86) William J. Hall (1975-78) Gary C. Hart Lyman Henry (1973) James A. Hill (1992-95) Ernest C. Hillman, Jr. (1973-74) (1983-84) Ephraim G. Hirsch (1983-87) William T. Holmes" Warner Howe (1977-80) (1990-97) Edwin T. Huston" (1973-75) Paul C. Jennings (1974-76) Carl B . Johnson (1988-89) Edwin H. Johnson Stephen E. Johnston" (1973-75, 1979-80) Joseph Kallaby* (1973-75) Donald R. Kay (1989-92) (1984-88) T. Robert Kealey" (1975-76) H. S. (pete) Kellam (1979-82) Helmut Krawinkler (1982-85) James S. Lai (1973-74) Gerald D. Lehmer (1992-98) James R. Libby (1989-92) Charles Lindbergh (1983-86) R. Bruce Lindermann (1987-90) L. W. Lu (1975-78) Walter B. Lum (1991-98) Kenneth A. Luttrell (1979-82) Melvyn H. Mark (1978-82) John A. Martin

John F. Meehan" (1973-78) Andrew T. Merovich (1996-99) David L. Messinger (1980-83) Stephen McReavy (1973) Bijan M ohraz (1991-97) William W. Moore* (1973-76) Gary Morrison (1973) Robert Morrison (1981-84) Ronald F. Nelson (1994-95) Joseph P. Nicoletti" (1975-79) Bruce C. Olsen" (1978-82) Gerard Pardoen (1987-91) Norman D. Perkins (1973-76) Maryann T. Phipps (1995-96) Sherrill Pitkin (1984-87) Edward V. Podlack (1973) Chris D. Poland (1984-87) (1976-79) Egor P. Popov (1987-93) Robert F. Preece" (1985-91) Lawrence D. Reaveley" (1986-89) Philip J. Richter" John M. Roberts (1973) Arthur E. Ross" (1985-91, 1993-94) C. Mark Saunders (1993-97) (1975-79) Walter D. Saunders" (1981-84) Lawrence G. Selna (1990-91) Wilbur C. Schoeller (1980-84) Samuel Schultz" (1977-81) Daniel Shapiro" (1996-99) Jonathan G. Shipp (1980-84) Howard Simpson" (1990-93) Mete Sozen (1982-83) Donald R. Strand (1975-79) James L. Stratta (1996-97) Scott Stedman (1976-79) Edward J. Teal (1973) W. Martin Tellegen (1991-98) John C. Theiss" (1992-99) Charles H. Thornton (1973) James L. Tipton (1975-77) Ivan Viest (1977-80, 1981-85) Ajit S. Virdee* (1987-90) J. John Walsh (1990-91) Robert S. White (1980-81, 1982-86) James A. Willis' (1974-77) Thomas D. Wosser (1987-88) Loring A. Wyllie (1981-84) Edwin G. Zacher (1982-85) Theodore C. Zsutty • President

ATC EXECUTIVE DIRECTORS (1973-Present) Ronald Mayes Christopher Rojahn Ion

(1979-81) (1981-present)

Roland L. Sharpe

Appendix C, Applied Technology Council proJects and Report Information

(1973-79)

G·IS

ATC 40 Vol. 2

Recommend Documents