HARRY H. EDWARDS AWARD WINNER

The PRESSS Five-Story PrecastConcrete Test Building

University of California at San DiegoLa Jolla, California

The five-story PRESSS (Precast Seismic Structural Systems) testbuilding, a large scale structure that includes four structuralprecast concrete frame systems and a post-tensioned jointedprecast concrete shear wall system, underwent severe seismictesting in late 1999 at the structural laboratory of the Universityof California at San Diego. The results of the five-story buildingtest have verified the design and analysis procedures developedin the earlier phases of the PRESSS research program, and havedemonstrated that properly designed precast concrete structuralsystems can perform extremely well when subjected to highseismic loading. This article presents an overview of the PRESSSresearch program, a description of the precast structural systemsused in the five-story test building, and the principalconclusions drawn from the testing program.

The PRESSS research program ispart of a joint U.S.-Japan effortwhose aim has been to demon

strate the viability of precast concretedesign for various seismic zones.’ Theprogram has been sponsored primarilyby the National Science Foundation(NSF) with industry support providedby the Precast/Prestressed ConcreteInstitute (PCI) and the Precast/Prestressed Concrete Manufacturers Association of California, Inc. (PCMAC).

An ongoing project since 1991, the

PRESSS research program has twoprimary objectives:

• To develop comprehensive and rational design recommendations neededfor a broader acceptance of precastconcrete construction in different seismic zones.

• To develop new materials, concepts, and technologies for precastconcrete construction in different seismic zones.

A further objective of the programis that the design and construction rec

ommendations resulting from thiscomprehensive program be incorporated into the appropriate buildingcodes.

The details of the PRESSS researchprogram are described by Priestley,’and a more thorough overview of thefive-story test building used in theprogram is presented by Nakaki, Stanton, and Sritharan.2

The test building is a five-storystructure that combines five differentseismic load resisting systems appro

20 PCI JOURNAL

priate for use in different seismiczones. It was intended to examine thesuitability of design concepts developed in the earlier phases of thePRESSS research program and in related research. One criterion used indetermining which systems would beincluded in the test building was thatthe concept should have been experimentally validated through componenttests. The testing of the completebuilding (see Fig. 1) addressed manydesign and constructibility issues thatcould not be accomplished by individual component testing.

The specific objectives for the five-story building test program were to:

• Validate a rational design procedure for precast seismic structural systems.

• Provide acceptance of prestressing/post-tensioning of precast seismicsystems.

• Furnish experimental proof ofoverall building performance undersevere seismic excitation.

• Establish a consistent set of designrecommendations for precast seismicstructural systems.

The test building was designedusing a direct displacement based design (DBD) procedure, rather than aforce based design method.2With theDBD approach, the resulting designbase shear is significantly lower thanwhat would be calculated by forcebased design. Moreover, it gives theengineer better control of the struc

Fig. 1. The seismic testing of the five-story building, which was carried out in thestructural laboratory of the University of California at San Diego, verified the designand analysis procedures developed in the earlier phases of the research program.

Fig. 2. The prestressed frame system (left) consisted of the hybrid frame connection and pretensioned frame connection. Theyielding frame system (right) comprised the TCY gap frame connection and the TCY frame connection. The flooring systems onthe first three floors were pretopped double tees while the upper two floors comprised topped hollow-core slabs.

September-October 2001 21

C0

4,UU

•0C

0

Fig. 3. Seismic loading and the concept of pseudodynamic testing.

Table 1. Structural components for the five-story PRESSS test building(2 bays x 2 bays).Item Quantity DimensionsPretopped double tees 12 8.0 x 30.0 ft6 in. thick hollow-core floor slabs 8 40 in. x 15.0 ft6 in. thick solid actuator connection floor slabs 12 Variable width x 15.0 ftGravity beams 20 8.5 x 16 in. x 15.0 ftGravity columns 2 18 x 18 in. x 37.5 ft

Hybrid frame 6Frame systems Pretensioned frame 2 Most are 14 x 23 in. x 15.0 ft

TCY gap frame 6TCY frame 4

Frame columns 9 -18x 18 in. x variable lengthSin, thick shear wall panels 4 9.0 x 18.75 ft

Note: 1 ft = 0.3048 m; 1 in. = 25.4 mm.

tural design by incorporating bothstrength and stiffness requirements.

The test building consisted of fourdifferent ductile structural frame systems in one direction of loading and ajointed structural shear wall system inthe orthogonal direction. Table 1 provides information on the componentsused in the test building.

The structure was tested in two orthogonal directions — one in whichonly the frames would resist the simulated seismic loading, and the other inwhich only the shear wall would resistthe loading. The loads representedearthquakes of seismic input levels upto 50 percent greater than those required by the Uniform Building Codefor Seismic Zone 4. Two independently controlled actuators were usedat each floor level to prevent torsionfrom occurring during the test loadings.

Fig. 2 shows two sides of the testbuilding. The “Prestresssed Frame Elevation” comprises the hybrid frameand pretensioned frame systems,whereas the “Yielding Frame Elevation” incorporates the tension-compression yielding (TCY) gap frameand TCY frame systems. These fourframe systems will be elaborated uponin the next section.

The structure was tested under a series of simulated earthquake levels.The principal method of testing waspseudodynamic testing, using spectrum-compatible earthquake segments(see Fig. 3). Because the test buildingwas designed using the DBD approach, displacements at each level ofthe structure had to be applied delicately so that they could work in unison. Also, because of differential thermal displacements between night andday temperatures, much of the testingtook place during the night.

Fig. 4 is a plan view of the testbuilding showing the framing systemsof the lower three floors. The flooringsystem was selected based on the predominance of their use in today’s precast concrete construction. On thelower three levels, the system consistsof pretopped double tees that span between the seismic-resistant frames. Onthe upper two floors, topped hollowcore slabs spanned between the gravity frames and the precast concreteshear wall system.

Reactio,wall

22 PCI JOURNAL

FRAME CONNECTIONSYSTEMS

The four different types of ductileconnection systems built into the testbuilding were:

• Hybrid frame connection• Pretensioned frame connection• TCY(tension-compression

yielding) gap frame connection• TCY frame connectionEven though in practice these four

different frame system types are notintended to be used in a single building, the PRESSS research team andadvisors felt that these all should beincluded in the test structure. Testingall four frame systems would provideuseful data on the fundamentally different types of behavior that might beappropriate for various situations anddemonstrate the versatility of precastconcrete that other systems do notpossess.

Hybrid Frame Connection

The hybrid beam-to-column connection is a system of post-tensioningstrands that run through a duct in thecenter of the beam and through the column. Mild steel reinforcement isplaced in ducts on the top and bottom Fig. 4. Plan view of test building showing lower three floors.

Fig. 5. Condition of the hybrid frame connection (left) and pretensioned frame connection (right) after structure was subjected todrift levels more than twice the design level.4

15,-o” 15 -0”•1 1

Hybrid and Pretensioned Frames

_ _

rn

T I fT—-rr —r I Ti -

I III I III I I I III II I III I I I III I II II I ii

II

IH

V V II

I IIII VV

I lIEI CIII I z Iz

0 I 0 II I 1

0 I IV 1>

II(5 I l

S El 2I III I ii

TCY and TCY Gap Frames

: •q

Hybrid Frame Connection Pretensioned Frame Connection

September-October 2001 23

Fig. 6. Details of the hybrid frame connection. Fig. 7. Details of the pretensioned frame connection.

of the beam and through the column,and then grouted [see Figs. 5 (left) and

6]. The amount of mild steel reinforce

ment and post-tensioning steel are proportioned so that the frame recenters itself after a major seismic event.

Post-tensioning provides the shear

resistance for the beam (eliminatingthe need for corbels), and mild steelprovides ductility and energy dissipation in the connection region by yield

ing. The post-tensioning strands remain elastic throughout the seismicevent while the mild steel is yielding.

The post-tensioning helps the structure

return to its initial position with negli

gible residual displacement.

Pretensioned Frame Connection

The pretensioned frame is ideallysuited for construction where onestory columns are combined with

multi-span beams. The beams are fab

ricated in normal pretensioned castingbeds, with specified lengths of thestrand debonded. During erection, thebeams are set on the one-storycolumns, with the column reinforcingsteel extending through sleeves in thebeams. Reinforcing bar splices provide the continuity of the columnabove the beam [see Figs. 5 (right)and 7]. As the frame displaces laterally, the debonded strand remainselastic. The system recenters the structure after a seismic event. The overallbehavior of this system is similar tothat of the hybrid frame system.

TCY Gap Frame Connection

In this system, the beams are erectedbetween columns, leaving a small gapbetween the end of the beam and theface of the column (see Fig. 8). Onlythe bottom portion of the gap isgrouted to provide contact from beam

to column. At the center of this bot

tom-grouted region, post-tensioningbars clamp the frame together. At the

top of the beam, mild steel reinforce

ment is grouted into sleeves that ex

tend the length of the beam and

through the column.Careful debonding for a specified

length at the gap allows the reinforc

ing steel to yield alternately in tension

and compression without fracture.

Even as the connection yields, the

frame length does not change because

the gap opens on one side of the col

umn as it closes an equal amount on

the other side. The sleeved connectionprevents a premature failure.

TCY Frame Connection

Fig. 9 shows the TCY frame system.

This system is similar to the traditional tension-compression yielding

connection used in cast-in-place con-

I

24 PC) JOURNAL

Fig. 8. Details of the TCY gap frame connection. Fig. 9. Details of the TCY frame connection.

struction except that, rather than theyielding being distributed over a finiteplastic hinge length, the yielding isconcentrated at the connection. Thebeam reinforcement that provides moment strength and energy dissipationis debonded over a short length at thebeam-to-column interface so that thereinforcement will not fracture prematurely at this yielding location.

PRECAST POST-TENSIONEDSHEAR WALL SYSTEM

Fig. 10 illustrates a jointed precastpost-tensioned shear wall system inthe test building. Vertical unbondedpost-tensioning is used to resist thelateral loads that could not be resistedby the inherent gravity load of the system. The special feature of this systemis the U-shaped flexure plate (UFP).These plates serve as vertical joint

connection devices where damping isachieved by means of flexural yielding of the plates (see Fig. 11).

The unbonded post-tensioning is designed to recenter the wall systemafter the seismic event has occurred.As a result, there will be negligibleresidual drift. Recentering is ensuredby relating the elastic capacity of thepost-tensioning system to the yieldstrength of the panel-to-panel connections.

TEST RESULTSThe major results of the testing pro

gram are as follows:

• The test results were very satisfactory, with the structural response exceeding the building code requirements.

• The PRESSS structural systemsprovide a level of seismic perfor

mance equal to or greater than that ofother structural systems.

• These systems are suitable for low,moderate, and high seismic regions.

• The systems typically sustain lessseismic damage than those of conventional cast-in-place systems.

• The design of these systems issimple and straightforward.

• Response to seismic loading, particularly drift, can be predicted accurately.

• All hardware (precast concreteproducts, prestressing steel, mild reinforcing steel, steel plates, and othermaterials) used in the various structural systems is conventional andwidely available.

The highlights of the testing event atthe University of California, SanDiego, are presented in Reference 3,and a comprehensive discussion of theresults and conclusions drawn from the

September-October 2001 25

program is presented in Reference 4.Among the five systems examined,

results from the hybrid frame. pretensioned frame, and the post-tensionedshear wall were particularly good. Notonly did these three systems experience minimal damage, but they alsoexhibited a self-centering characteristic that does not exist in other seismicsystems. This self-centering characteristic allows immediate re-occupancyof the building after a major seismicevent and makes precast concrete anattractive choice in seismic areaswhere previously precast constructionwould not have been an option.

The PRESSS research program hasproved that all-precast concrete systems can provide the highest qualitysolution for major commercial, institutional, and industrial buildings in allseismic zones. The research under thisprogram is now entering the code-ap

proval process. The PCI JOURNALwill present further test analysis, design and construction details, designimplications of the testing, and the direct displacement based design procedure in future issues.

JURY COMMENTS“This work will advance the precast,

prestressed concrete industry by opening new markets in moderate and highseismic areas. A key point is that theperformance standards not only metbut also exceeded all expectations.The benefits of this successful test areonly beginning to be seen, with moreundoubtedly to come. Showing theworld that precast concrete behaveswell in a high seismic area will createmore widespread use of the productand better, more cost-effective structures overall.”

REFERENCES1. Priestley, M.J.N., “The PRESSS Pro

gram — Current Status and ProposedPlans for Phase Ifi,” PCI JOURNAL,V. 41, No. 2, March-April 1996, pp.22-40.

2. Nakaki, S.D., Stanton, J.F., and Sritharan, S., “An Overview of the PRFSSSFive-Story Precast Test Building,” PCIJOURNAL, V. 44, No.2, March-April1999, pp. 26-39.

3. Johal, L.S., and Nasser, G.D., “Successful Testing of PRESSS Five-StoryPrecast Building Leads to InnovativeSeismic Solutions,” PCI JOURNAL,V. 44, No. 5, September-October1999,pp. 120-123.

4. Pnestley, M.J.N., Sritharan, S., Conley, J.R., and Pampanin, S., “Preliminary Results and Conclusions from thePRESSS Five-Story Precast ConcreteTest Building,” PCI JOURNAL, V.44, No. 6, November-December 1999,

pp. 42-67.

Fig. 10. Details of the post-tensioned shear wall system. Fig. 11. Details of the U-shaped flexure plate (UFP) for theshear wall system.

26 PCI JOURNAL

CREDITS

The PRESSS research program hasbeen funded by the National Science

Foundation, the Precast/PrestressedConcrete Institute, the Precast/Prestressed Concrete Manufacturers Association of California, Inc., and by

various precasters and suppliers.In addition to the design and testing

components of the PRESSS test build

ing [NSF Grant Numbers CMS 97-

10735 (UW) and CMS 97-00125(UCSD)], analysis of the structure wasperformed by Lehigh University (NSFGrant Number CMS 97-08627).

The technical input provided by

many PCI members, especially by

PCI, s Ad Hoc Committee on ATLSSand PRESSS under the chairmanshipof Mario Bertolini, has played a significant role in the planning, development, and execution of this program.In particular, M.J. Nigel Priestley, theprincipal coordinator of the PRESSSresearch program, provided key leadership through all phases of this research.

During the design of the test building, the members of the PRESSS research team and Industry AdvisoryGroup provided invaluable guidance.

Project Sponsors

National Science Foundation, Ailing-ton, Virginia

Precast/Prestressed Concrete Institute(PCI), Chicago, Illinois

Precast/Prestressed Concrete Manufacturers Association of California(PCMAC), Glendale, California

Engineers of Record

John F. Stanton, Civil Engineering Department, University of Washington,Seattle

Suzanne Dow Nakaki, The NakakiBashaw Group, Inc., Irvine, California

General Contractor

School of Engineering, University ofCalifornia at San Diego, La Jolla,California

Precast Concrete Manufacturers

A. T. Curd Structures Inc., Rialto, Cal

iforniaClark Pacific, West Sacramento, Cali

forniaCoreslab Structures (LA) Inc., Perris,

CaliforniaHanson Spancrete Pacific Inc., Irwin-

dale, CaliforniaMid-State Precast, L.P., Corcoran,

CaliforniaPomeroy Corporation, Penis, Califor

nia

Industry Advisory Group

Mario J. Bertolini, Chairman,

ATLSS/PRESSS CommitteeThomas J. D’Arcy, Chairman,

R & D CommitteeRobert E. ClarkNed M. ClelandRobert E. EnglelcirkS. K. GhoshR. Jon GraftonJames K. IversonL.S. (Paul) JohalRobert H. KonoskeH. S. LewDouglas L. LorahPaul D. MackRobert F. MastWilliam D. MichaleryaDouglas M. MooradianJohn G. NannaNorman L. ScottDavid C. SeagrenA. Fattah ShaikhEdward A. Wopschall

PRESSS Researchers

M. J. Nigel Priestley, PrincipalCoordinator

Daniel P. AbramsCatherine E. FrenchNeil M. HawkinsRebecca HixMichael E. KregerLe-Wu LuRafael A. MaganaAntoine E. NaamanSuzanne D. NakakiMichael G. OlivaStephen P. PessikiJose PincheiraRichard SauseArturo B. SchultzFrieder SeibleS. (Sri) SritharanJohn F. StantonMaher K. TadrosSharon L. Wood

Materials and Services Providers

BauMesh Co.BauTech, Inc.California Field Iron WorkersAdministrative Trust

Dayton SuperiorDywidag Systems InternationalErico, Inc.Fontana SteelGillies TruckingHeaded Reinforcement Corporation

Horizon High ReachInsteel Wire ProductsJVI, Inc.LG DesignMTS Systems CorporationMaster Builders, Inc.The Nakaki Bashaw Group, Inc.Nielsen Dillingham BuildersNMB Splice SleevePrecision ImagerySumiden WireUCSD_TVWhite Cap

September-October 200127