Suzanne Dow Nakaki, S.E.Principal

Nakaki EngineeringSanta Ana, California

John F. Stanton, Ph.D., P.E.ProfessorDepartment of Civil EngineeringUniversity of WashingtonSeattle, Washington

At the culmination of the PRESSS (Precast SeismicS t r u c t u ral Systems) research program, a 60p e rcent scale f ive-story precast/prestressedconcrete building will be tested under simulatedseismic loading. This paper describes theprototype buildings used for design and thes t r u c t u ral features of the test building. Th ebuildings were designed using the directdisplacement based approach, wh i ch is able totake advantage of the unique properties ofprecast/prestressed concrete using dry jointedc o nst ru c t i on . The t e st building incorp orates f ourdifferent seismic frame systems in one direction,and a jointed shear wall system in the orthogonaldirection. Pretopped double tees are used on threefloors, while the other two floors are constructedusing topped hollow-core slabs. A major objectiveof the test program is to develop design guidelinesfor precast/prestressed concrete seismic systemsthat are appropriate for use in various seismiczones. These design guidelines can then beincorporated into the appropriate building codes.

The Precast Seismic Structural Systems (PRESSS)program has been in progress for ten years, with thefinal phase of the program well underway. PRESSS,sponsored by the National Science Foundation (NSF), Precast/Prestressed Concrete Institute (PCI) andPrecast/Prestressed Concrete Manufacturers Association ofCalifornia, Inc. (PCMAC), has coordinated the efforts ofover a dozen different research teams across the United

An Overview of the PRESSS Five-Story Precast Test Building

26 PCI JOURNAL

S. (Sri) Sritharan, Ph.D.Assistant Project Scientist

Department of Applied Mechanicsand Engineering Sciences

University of California, San DiegoLa Jolla, California

SPECIAL REPORT

SONYHighlight

March-April 1999 27

States to improve the seismic perfor-mance of precast/prestressed concretebuildings. In the context of this paper,buildings refer to low- and high-risebuildings such as office buildings,parking structures, hotels, hospitals,multi-family housing, and other spe-cial structures. However, bridges and

transportation structures are excluded.Since the very beginning of the

PRESSS program, all of the researchteams involved in the program havefocused their sights on two primaryobjectives: To develop comprehensive and ra-

tional design recommendations

needed for a broader acceptance ofprecast concrete construction in dif-ferent seismic zones.

To develop new materials, concepts,and technologies for precast con-crete construction in different seis-mic zones.The first and second phases of the

Fig. 1. Prototype building with pretopped double tees. Note: 1 ft = 0.3048 m.

Fig. 2. Prototype building with topped hollow-core slabs. Note: 1 ft = 0.3048 m; 1 in. = 25.4 mm.

8 in. hollow-core slabs(typ)

28 PCI JOURNAL

PRESSS program have been describedby Priestley in the PCI JOURNAL.1The third phase consists of the seismicdesign and analysis of a five-story pre-cast/prestressed concrete buildingusing dry jointed construction. A por-tion of this building will be built at 60percent scale and tested. The purposeof this paper is to present an overviewof the test building, describe the majorfeatures of the structural systems in-vestigated and offer some thoughts onthe practical implications of the testresults.

PRESSS III PROGRAMOBJECTIVES

Academic research is often focusedsolely on improving the performance ofexisting structural systems. While his-tory confirms that this is a worthy goal,the reality of the construction market-place is that improved performance of asystem will generally not be accepted

unless it also results in a lower cost.Thus, the PRESSS Phase III researchteam, comprising researchers and in-dustry advisory group members, haskept in mind that in addition to improv-ing performance, cost effectiveness ofthe resulting systems is crucial.

The PRESSS Phase III test programis based on the design of two proto-type five-story precast office build-ings, 100 x 200 ft (30.5 x 61 m) inplan, with 12 ft 6 in. (3.81 m) storyheights. Both buildings use frames toresist lateral loads in the longitudinaldirection and shear walls to resist lat-eral loads in the transverse direction.The first building, shown in Fig. 1,uses pretopped double tees to span be-tween a central gridline and theperimeter of the building. The secondprototype building, shown in Fig. 2, isbased on a topped hollow-core slabfloor system. For simplicity, the samefloor system was assumed at the roofas well as at each floor.

Fig. 3. Current code design choices for precast systems.

Detailing requirements

Column reinforcement to ensure weak beam/strong column

Column confinement reinforcement

Column shear reinforcement

Joint shear stress limitations

Beam shear reinforcement

Positive moment resistance in beam

Design base shear

Intermediate Moment Resisting Frame

Not required

Tight tie spacing is required on top and bottom of column

Required

No limit; however, joint shear reinforcement is required

Required

Required

150 to 160 percent of that required for Special Moment Resisting Frame

Special Moment Resisting Frame

Often a few additional longitudinal column bars are required

Same as Intermediate Moment Resisting Frameexcept where axial overload is possible

(normally at end bays of frames)

Required

Limited; this requires a larger column onlywhere beams are heavily reinforced

Required

Required

60 to 65 percent of that required for Intermedi-ate Moment Resisting Frame

Table 1. Detailing requirements of Special Moment Resisting Frame and Intermediate Moment Resisting Frame systems.

The size of the testing laboratorylimited the test building to 30 x 30 ft(9.14 x 9.14 m) in plan. Rather thandesigning the test building to resistjust its own inertial loads, the inertialloads of the prototype buildings werecalculated and then scaled down torepresent the scale of the test building.This gives a more accurate picture ofthe demand that a practical buildingconfiguration would be subjected to,without exceeding the space limita-tions of the laboratory. The test build-ing will be subjected to increasinglylarger seismic demands that representlow service level earthquakes, moder-ate (Zone 2 design level) earthquakesand design level earthquakes beyondthose required for Zone 4.

The ultimate objective of the re-search, however, is not the test itself,but the design recommendations thatwill result from the testing program.Because there are so many differentcombinations of systems included inthe test building, it does not representthe most economical way to implementthese new structural systems. The finaldesign recommendations are the key toobtaining improved performance of theproposed systems at a competetive costin practical applications.

EXISTING DESIGN CODESDuring the life of the PRESSS pro-

gram, there have been significant de-velopments in the model codes2 , 3 t h a tprovide some guidance to design engi-

SONYHighlight

SONYHighlight

March-April 1999 29

neers wanting to implement precastseismic systems in their buildings. Asshown in Fig. 3, current codes allowprecast seismic systems that either em-ulate monolithic concrete or rely onthe unique properties of precast con-crete (i.e., jointed, dry construction).

While jointed construction is allowedby the code, the focus of the prescrip-tive code provisions has been on emu-lation of monolithic concrete, largelybecause a consistent set of design rec-ommendations for jointed precast sys-tems have not been developed. Jointedsystems can only be used if they arejustified by test data on a case-by-casebasis. The PRESSS program goes astep further by focusing its efforts al-most exclusively on systems that relyon and take advantage of the uniqueproperties of precast concrete. The in-tention is then to develop a consistentset of design recommendations forjointed precast systems that can be usedto update existing code provisions.

Force Based Design

Seismic design in current codes isexclusively force based. That is, a de-signer uses elastic properties to deter-mine an elastic base shear, which isthen divided by a force-reduction fac-tor R to obtain the design base shear.The value of R depends largely on thenominal ductility capacity of the sys-tem chosen, which is somewhat arbi-trary and varies between codes. Whilemaximum structural displacementsmust satisfy certain limits, they are inmost cases based on elastic structuralproperties and are amplified by factorsintended to approximate the post-elas-tic response. This approach has somesignificant drawbacks, as discussed byP r i e s t l e y ,4 especially for precast con-crete. Despite these difficulties, it willcontinue to be the legal design proce-dure for at least the foreseeable future.

In Force Based Design, there are twomain ways that a designer can reducethe cost of a seismic system. Bothmethods depend on reducing the de-sign loads because for consistent de-tailing, a lower force results in a lowercost. In the first method, a larger R f a c-tor is used to reduce the design baseshear. For frames, the R value can bemaximized by detailing the structure asa Special Moment-Resisting Frame

Fig. 4. Test building Level 1 floor plan. Note: 1 ft = 0.3048 m.

Fig. 5. Test building Level 4 floor plan. Note: 1 ft = 0.3048 m.

SONYHighlight

30 PCI JOURNAL

(UBC R = 8.5, NEHRP R = 8) ratherthan an Intermediate Moment-Resist-ing Frame (UBC R = 5.5, NEHRP R =5). The second method consists ofusing a longer period to reduce the de-sign base shear. This method forms thebasis of recommendations proposed bythe PCI Ad Hoc Committee Report onPrecast Walls.5

Frame Systems

Ordinary Moment-Resisting Frames(OMRF) are not permitted in moderateand high seismic zones (UBC Zones 2,3, and 4) because of their fundamentallack of ductile behavior. For seismicdesign using frames in moderate seis-mic zones, a designer has a choice be-tween using an Intermediate Moment-Resisting Frame (IMRF) or a SpecialMoment-Resisting Frame (SMRF).Table 1 compares the detailing require-ments of the two frame types. In highseismic zones, only SMRF frames arep e r m i t t e d .

The appearance of a choice is de-ceptive because the SMRF is almostinvariably the most cost effectiveframe solution. This is so because thedesign loads on an SMRF are 35 to 40percent lower, primarily due to thehigher R factor. Also, the period of anSMRF system is slightly longer thanthat of an IMRF system for the samebuilding, due to the lower frame stiff-ness. This, too, means that the SMRFdesign load is lower. These benefitseasily outweigh the extra costs of theslightly more stringent detailing re-quirements for the SMRF.

In summary, therefore, it is fromthis perspective of the need for ductileperformance and cost effective designthat only SMRF systems were chosenfor the PRESSS III test building.These systems are appropriate, andcost effective, in all seismic zones.

Wall System

Wall systems designed under cur-rent codes are described as either load-bearing or non-loadbearing walls.Since non-loadbearing walls are usu-ally more ductile than loadbearingwalls, the UBC R factor for them is 18percent more than that for loadbearingwalls. This results in an 18 percent de-crease in the design base shear and aconcomitant reduction in the cost for

Fig. 6. Prestressed frame elevation. Note: 1 ft = 0.3048 m; 1 in. = 25.4 mm.

Fig. 7. Tension-Compression Yielding (TCY) frame elevation. Note: 1 ft = 0.3048 m;1 in. = 25.4 mm.

SONYHighlight

SONYHighlight

March-April 1999 31

non-loadbearing wall systems that areotherwise identical to their loadbear-ing wall counterparts.

It is fai rly straightforward tolengthen the building period in a pre-cast shear wall system by providingvertical joints between the panels thatmake up a wall (see PCI Ad HocCommittee Report on Precast Walls5).Thus, by providing a jointed shearwall, the design forces are reduced, re-sulting in a reduced building cost.

Results of Force Based Design

It should be noted that, although im-proving ductility and lengthening thesystem period reward buildings withlower design loads, the magnitude ofthe reduction reflects only poorly thetrue advantages that well-designedprecast systems offer. For example,the design base shear for the prototypebuilding using force based design inaccordance with the 1997 UBC (Zone4) is as follows:Frame direction (Tn = 0.67 seconds)Design base shear = 2248 kips (10000 kN)

Fig. 8. Hybrid frame interior joint (transverse reinforcement not shown for clarity). Note: 1 ft = 0.3048 m; 1 in. = 25.4 mm.

Fig. 9. Hybrid framehysteresis loop(from Ref. 6).

Wall direction (Tn = 0.48 seconds)Design base shear = 4889 kips (21746 kN)

These values reflect the advantagesof a ductile system (i.e., R = 8.5 forframes) and a lengthened period forthe shear wall building which is com-

prised of jointed wall panels. The de-sign base shear for an equivalent cast-in-place frame system would be iden-tical, since the elastic stiffnesses of aprecast frame and a cast-in-placeframe are similar. However, the elastic

SONYHighlight

SONYHighlight

32 PCI JOURNAL

period for an equivalent, non-jointed,cast-in-place wall would be substan-tially shorter than the jointed wall pe-riod. Except in cases where the maxi-mum base shear governs, a shorterperiod would result in a higher baseshear.

While the systems included in thetest building are expected to be costeffective even using force based de-sign, the PRESSS III test buildingadopts an alternative design procedurethat more efficiently incorporates theadvantages of well-designed precast

Fig. 10. Pretensioned frame interior joint (transverse reinforcement not shown for clarity). Note: 1 ft = 0.3048 m; 1 in. = 25.4 mm.

Fig. 11. Pretensioned frame hysteresis loop (from Ref. 7).

systems. As will be discussed below, afurther reduction to design base shearis achieved, providing substantial costsavings for precast buildings in allseismic zones.

DESIGN OF PRESSS III TEST BUILDING

The PRESSS Phase III test buildingis not intended to create new designconcepts, but rather to examine thesuitability of design concepts createdin earlier phases of the PRESSS pro-gram or other precast concrete re-search. One criterion used in deter-mining which systems would beincluded in the test building was thatthe concept had to have been experi-mentally validated through componenttests.

The complete building test is impor-tant because it addresses many ques-tions of design and constructability,which do not arise in component tests.Also, the behavior of a complete, stati-cally indeterminate system involvesmany features, including verification

March-April 1999 33

The building will be tested in boththe frame and wall directions indepen-dently under simulated seismic loadsthat represent earthquakes up to 50percent stronger than Zone 4 designlevel earthquakes recognized in codes.During the loading in each direction,

two independently controlled actuatorsat each floor level will prevent torsion.

Frame Connection Systems

Four different types of ductile connec-tion systems are used in the PRESSS IIItest building frames. They are:

Fig. 12. TCY gap frame interior joint (transverse reinforcement not shown for clarity). Note: 1 ft = 0.3048 m; 1 in. = 25.4 mm.

Fig. 13. TCY gap frame hysteresis loop (from Ref. 7).

of seismic design methods that do notoccur in statically determinate compo-nent tests.

The specific objectives of the testare to: Validate a rational design procedure

for precast seismic structural sys-tems.

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

Provide experimental proof of over-all building performance under seis-mic excitation.

Establish a consistent set of designrecommendations for precast seis-mic structural systems.The PRESSS III test building con-

sists of frames in one direction and ashear wall in the other, as shown inFigs. 4 and 5. The floor system used inthe first three levels is pretopped dou-ble tees, and the top two levels consistof topped hollow-core slabs. Thosechoices were made in order to includethe two major structural framing sys-tems commonly used in precast con-struction today.

SONYHighlight

SONYHighlight

34 PCI JOURNAL

Tension-Compression Yielding(TCY) gap connection

TCY connection Hybrid connection Pretensioned connection

The first three types of connectionsconsist of multistory columns and sin-

gle-bay beams, and are appropriatefor floor-by-floor construction. Thepretensioned connection uses multi-bay beams and single-story columnsand is appropriate for up-and-outc o n s t r u c t i o n .

The hybrid connection and preten-

Fig. 14. TCY frame interior joint (transverse reinforcement not shown for clarity). Note: 1 ft = 0.3048 m; 1 in. = 25.4 mm.

Fig. 15. TCY frame hysteresis loop (from Ref. 7).

sioned connection are used in oneseismic frame, referred to as the Pre-Tensioned Frame, and the remainingtwo connections are adopted in theother seismic frame, known as theTension-Compression Yielding Frame.These two frame elevations are shownin Figs. 6 and 7, respectively. Theamounts of energy dissipation andresidual displacement vary among thefour connections, allowing a designerto control seismic behavior of thestructure with an appropriate choice ofconnection system.

Hybrid Frame

The hybrid connection was devel-oped during the last phase of a multi-year project at the National Institute ofStandards and Technology (NIST).6The hybrid frame interior joint isshown in Fig. 8. The beams are con-nected to multistory columns by un-bonded post-tensioning strands thatrun through a duct in the center of thebeam and through the columns. Mildsteel reinforcement is placed in ductsat the top and bottom of the beam,

SONYHighlight

SONYHighlight

SONYHighlight

SONYHighlight

SONYHighlight

Table 2. Frame system alternatives.

March-April 1999 35

through the column, and is grouted. Ityields alternately in tension and com-pression and provides energy dissipa-tion (see Fig. 9). The amount of mildsteel reinforcement and post-tension-ing steel are balanced so that the framere-centers after a major seismic event.

The exterior joint of the HybridFrame uses a stub beam that con-tains the multistrand anchor. This isonly required due to the scale of thetest building. Research8 indicates thatanchors located within the joint mayactually improve joint performance.

PreTensioned Frame

The PreTensioned frame, named soas to differentiate it from just anyframe constructed with pretensionedmembers, is intended to be used forconstruction where the most economi-cal method consists of using one-storycolumns with multi-span beams.Long, multi-span beams are cast innormal pretensioned casting beds,with specified lengths of the preten-sioning strand debonded.

These beams are then set on one-story columns with the column rein-forcing steel extending throughsleeves in the beams. Reinforcing barsplices ensure the continuity of thecolumn above the beam, as shown inFig. 10. As the frame displaces later-ally, the debonded strand remainselastic. While the system dissipatesrelatively less energy than other sys-tems7,9,10 (see Fig. 11), it re-centers thestructure after a major seismic event.TCY Gap Frame

The TCY gap frame addresses theproblem of frame beam elongation inan innovative way. The beams areerected between columns leaving asmall gap between the end of the beamand the face of the column. Only thebottom portion of this gap is groutedto provide contact between the beamand column (see Fig. 12). Centered onthis bottom grout region, post-tension-ing bars clamp the frame together. Atthe top of the beam, mild steel rein-forcement is grouted into sleeves thatextend the length of the beam andthrough the column.

The reinforcing steel is carefullydebonded for a specified length at thegap so that it can yield alternately in

Cast-in-place concrete

Special Moment Resist-ing Frame per code

Masonry

Frame per code

Structural steel

Dog bone

Cover plates

Meyers NelsonHoughton connection

Others

Precast concrete

Hybrid

Pretensioned

Tension/compressionyielding gap

Tension/compressionyielding

Others

tension and compression without frac-ture. Since the gap opens on one sideof the column as it closes on the otherside by an equal amount, the length ofthe frame does not change, even as theconnection yields. The TCY gap con-nection tested in a PRESSS Phase IIresearch program7 used a coupler tosplice the reinforcing steel through thecolumn, rather than the sleeve throughthe column shown in Fig. 12.

The hysteresis loop obtained inPRESSS Phase II shows that this sys-tem7 was performing as expected, anddissipated significant energy, untilpremature failure of the reinforcingbar couplers at the top of the beamfailed the connection (see Fig. 13).The possibility of a premature failureof this type is eliminated by thesleeved connection.

TCY Frame

The TCY frame connection at-tempts to model a traditional tension/compression yielding connection,similar to what is used in cast-in-placeconstruction. However, rather thandistributed yielding over a finite plas-tic hinge length, yielding is concen-trated at the connection. To ensurethat the beam reinforcement that pro-vides moment strength and energydissipation does not fracture prema-turely at this concentrated yielding lo-cation, it is debonded over a shortlength at the beam-to-column inter-face (see Fig. 14).

This type of connection was alsotested in PRESSS Phase II researchp r o g r a m ,7 , 9 , 1 0 where it showed slightlypinched hysteretic behavior due tovertical slip at the beam-to-column in-terface (see Fig. 15). Although thistype of behavior may also occur in the

PRESSS III test building, the connec-tion has been included since it is con-ceptually very similar to traditionalmethods of construction. If verticalslip starts to occur at the ends of thesebeams, steel corbels will be installedduring the test so that slip does not ad-versely affect the overall test results.

Frame Columns

The frame columns used for all sys-tems contain both mild steel reinforce-ment and post-tensioning bars (seeFigs. 8, 10, 12 and 14). The post-ten-sioning bars are intended to representthe equivalent dead loads based on theprototype structure, but their inclusionin the test will also validate that thismethod of adding vertical load to aprecast column is an effective way toinfluence system performance.

In addition, the columns in the pre-stressed concrete frame are preten-sioned up to the fourth level of thebuilding. This bonded prestressingeconomically adds strength to thecolumns, which are prevented fromyielding using capacity based design.These details will validate the perfor-mance of pretensioned frame columns.

Building Frame Choices

While it was never intended thatmultiple connection types would beused on different floors or in differentframes of the same building in practice,the PRESSS research team and indus-try advisors felt strongly that severaldifferent frame systems should be in-cluded in the test building. The objec-tive was to provide designers with sev-eral alternatives using precast concrete;not just different ways of building con-ceptually similar systems (e.g., struc-tural steel in Table 2), but systems with

SONYHighlight

SONYHighlight

SONYHighlight

SONYHighlight

SONYHighlight

SONYHighlight

SONYHighlight

36 PCI JOURNAL

fundamentally different types of behav-ior that might be appropriate for differ-ent situations. This, as shown in Table2, will provide versatility using precastconcrete that is not currently availableusing any other building material.

In addition, by validating severaldifferent frame types, it is hoped thatfuture innovations can fit into theframework developed by the PRESSSresearch program, through componenttesting rather than requiring additionallarge-scale building tests.

Wall System

For the past several years, the PCIAd Hoc Committee on Precast Wallshas been promoting precast shearwalls as seismic resisting systems forall seismic zones.5 This work has fo-

cused on tuning jointed walls tolengthen the structural period and re-duce the design base shear forces. Thefocus was on evaluating elastic stiff-ness, without explicit consideration ofductility. Elastic forces were dis-tributed so that sufficient resistance tooverturning was provided by the grav-ity loads on the system.

The PRESSS test building takes thisconcept one step further by consider-ing the behavior of the jointed shearwall system when the wall lifts off androcks, together with its effect on de-sign forces. An appropriate level ofhysteretic damping is added to thewall system through the connectiondevices located at the vertical joint be-tween the wall panels.

Due to limitations on the buildingsize, imposed by the dimensions of the

Fig. 16. Elevation of jointed shear wall system. N o t e : 1 ft = 0.3048 m; 1 in.= 25.4 mm.

testing laboratory, only one jointedwall system is incorporated in the testbuilding. Instead of limiting the lateralloads to those that could be resisted bythe inherent gravity loads in the sys-tem, vertical unbonded post-tension-ing is used to resist overturning in thiswall system.

U-shaped flexure plates (UFP), astested in PRESSS Phase II,1 1 are usedfor vertical joint connection deviceswhere damping is achieved by meansof flexural yielding of the plates. Theunbonded post-tensioning is designedto re-center the wall system when theload is removed so there will be noresidual drift after a design-level earth-quake. Re-centering is ensured by re-lating the elastic capacity of the post-tensioning system to the yield strengthof the panel-to-panel connections.1 2

Fig. 16 shows the shear wall eleva-tion, with unbonded post-tensioninglocated at the center of each panel.The shear wall is expected to displacelaterally to approximately 2 percentstory drift under a design-level earth-quake. This is consistent with the driftlimits specified in both the UBC2 andNEHRP provisions.3

This lateral displacement requires avertical panel-to-panel displacementof about 2 in. (51 mm) for the 9 ft(2.74 m) panel. Thus, the UFP connec-tion shown in Fig. 17 was chosen forits ability to retain its force capacitythrough this large displacement. Thepost-tensioning was designed to bejust at the point of yielding at 2 per-cent drift. Should the designer desire asmaller design story drift, or less en-ergy dissipation, simpler panel con-nections could be used.

DIRECT DISPLACEMENTBASED DESIGN

As noted previously, Force BasedDesign represents the behavior ofjointed precast systems poorly. Themethod relies on an initial elastic pe-riod, which is not only difficult tocompute in a system whose flexibilityresides largely in the connections, butalso has little influence on the post-elastic behavior of the structure. The Rfactors included in design codes arealso not intended to be applied to sys-tems, such as some of those used here,

SONYHighlight

March-April 1999 37

which do not emulate monolithic con-crete structures. Thus, the results ob-tained by representing the seismic per-formance of precast systems using a Force Based Design approach are questionable.

For this reason, the test building wasdesigned using a more consistent Di-rect Displacement Based Design(DBD) procedure,4 in which the designis based directly on an inelastic targetdisplacement and effective stiffness.The target structural displacement isdetermined from an allowable inter-story drift permitted by design codeswhile the effective stiffness is approxi-mated to the secant stiffness of thebuilding corresponding to its expectedfundamental mode of response. Use ofboth the elastic stiffness for determin-ing inelastic structural displacementsand arbitrary reduction factors, as inForce Based Design, are completelyeliminated in this design approach.

Direct Displacement Based Design Procedure

Direct Displacement Based Design(DBD) is a process that is intended toensure that the structure reaches, butdoes not exceed, a target displacementselected by the designer, in response toa given ground motion. In this methodthe true hysteretic behavior is replacedby a linear system in which the stiff-ness is equal to the true secant stiffnessand the viscous damping provides thesame energy dissipation per cycle.

The DBD design procedure, asadopted in the test building, is illus-trated in Fig. 18. Once the target driftis chosen, the damping is estimated forthe building using prior componenttest results. Representing the buildingwith a SDOF system, the fundamentalperiod corresponding to the target dis-placement is found from the displace-ment spectrum. The effective stiffnessis computed from the known mass andthe estimated period.

The design base shear is then ob-tained from the effective stiffness andtarget displacement. Member sizes andreinforcement are chosen to resist thisbase shear. The true physical proper-ties of the members are used to gener-ate a more refined, hysteretic, force-displacement curve. The effectivedamping is calculated from the hys-

Fig. 17. Detail of U-shaped flexure plate. Note: 1 ft =0.3048 m; 1 in. = 25.4 mm.

Fig. 18. Flowchart showing Direct-Displacement Based Design method.

SONYHighlight

38 PCI JOURNAL

teresis loop area and is checkedagainst the assumed value. If they dif-fer significantly, the process is re-peated with a new value of assumeddamping. This final step is only neces-sary because of the lack of informa-tion on global hesteretic damping forthe systems used in the test building.

Results of Direct DisplacementBased Design

For the PRESSS III prototype build-ing, Direct Displacement Based De-sign resulted in a design base shearnoticeably lower than would be usedfor force based design. For the proto-type building, the design base shearsare as follows:Frame direction:Design base shear = 1467 kips (6525 kN)Wall direction:Design base shear = 2223 kips (9888 kN)

In the frame direction, this is 65 per-cent of the equivalent Force Based De-sign base shear, resulting in a substan-tial cost savings. In the wall direction,the savings are similar, even if thelengthened period is used in ForceBased Design. The wall direction DBDbase shear is just 45 percent of theequivalent Force Based Design value(see Fig. 19). Clearly, the improvedperformance of these systems can alsoresult in substantial cost savings overtraditional structural systems.

TESTING SCHEDULEThe PRESSS III test building is

under construction in the Charles LeePowell Structural Laboratory of the

University of California at San Diego,as of the publication date of thispaper. Following the completion ofthe building in April 1999, testing isscheduled to begin in May. Testing isexpected to be complete by July 1999,with analysis and reports to follow.The report on design recommenda-tions is scheduled for completion byAugust 2000.

CONCLUDING REMARKSThe PRESSS Phase III test building

and Design Recommendations will val-idate the seismic performance of fivedifferent ductile precast concrete sys-tems. These systems are economicaleven using Force Based Design, butwill be even more advantageous oncetheir beneficial attributes can be di-rectly taken into account using DirectDisplacement Based Design (DBD).

As is clear in the design of the testbuilding, the benefits of the DBD ap-proach to precast concrete buildingsare substantial. Following validationof this design method by the PRESSSIII test building, a coordinated effortcan hasten the development of designrecommendations. Once the designrecommendations are published, theprecast industry should be well posi-tioned to implement the DBD ap-proach and facilitate its acceptanceinto building codes.

Recently, several codes have in-cluded sections on precast concreteseismic systems, but they apply pri-marily to emulative systems. Thesesections should be expanded to coverjointed systems and to incorporate theresults of the PRESSS research pro-

gram if its benefits are to be fully uti-lized. Then, precast concrete will trulybe the solution of choice in all seis-mic regions of the world.

ACKNOWLEDGMENTSThe PRESSS research program is

funded by grants from the NationalScience Foundation, the Precast/Pre-stressed Concrete Institute, the Pre-cast/Prestressed Concrete Manufactur-ers Association of California, Inc., andby various precasters and suppliers.Their contributions are gratefully ac-knowledged.

In addition to the design and testingcomponents of the PRESSS III project(NSF Grant Numbers CMS 97-10735(UW) and CMS 97-00125 (UCSD),analysis of the test structure is beingperformed by Lehigh University (NSFGrant Number CMS 97-08627). Theiranalysis is based on the methods de-veloped in Refs. 13, 14 and 15.

The technical input provided bymany PCI members, especially byPCIs Ad Hoc Committee on ATLSSand PRESSS under the chairmanshipof Mario Bertolini, has played a sig-nificant role in the planning, develop-ment, and execution of this program.Also, M. J. Nigel Priestley, the princi-pal coordinator of the PRESSS re-search program, played a key role inproviding leadership through allphases of this research.

During the design of the test build-ing, the following members of thePRESSS Phase III research team andIndustry Advisory Group provided in-valuable guidance and their contribu-tions are greatly appreciated:

Fig. 19. Comparison of

design base shears.

March-April 1999 39

Industry Advisory Group

Mario J. Bertolini, ChairmanRobert ClarkNed M. ClelandThomas J. DArcyRobert E. EnglekirkS. K. GhoshPaul JohalRobert KonoskeH. S. LewRobert F. MastDoug MooradianJohn G. NannaDavid C. SeagrenEdward A. Wopschall

PRESSS Researchers

M. J. Nigel Priestley, Principal Coordinator

Catherine FrenchNeil M. HawkinsRebecca HixMichael E. KregerLe-Wu LuSuzanne D. NakakiStephen P. PessikiRichard SauseFrieder SeibleS. (Sri) Sritharan John F. Stanton

In addition, various producers andsuppliers have donated significantproducts and effort to allow the fivedifferent systems to be tested. Theyare:

A. T. Curd Structures, Inc.Charles Pankow Builders, Ltd.Clark Pacific Coreslab Structures, L.A.Dywidag Systems International ERICO Fontana SteelHeaded Reinforcement CorporationNMB Splice Sleeve

Pomeroy Corporation Spancrete of CaliforniaSumiden Wire

Their support and donations aregratefully acknowledged.

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

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

2. Uniform Building Code, I n t e r n a t i o n a lConference of Building Officials,Whittier, CA, 1997.

3. Building Seismic Safety Council,NEHRP Recommended Provisions forSeismic Regulations for New Buildingsand Other Structures, National Earth-quake Hazard Reduction Program,Washington, D.C., 1997.

4. Priestley, M. J. N., Displacement-Based Approaches to Rational LimitStates Design of New Structures, Pre-sented at the Eleventh European Con-ference on Earthquake Engineering,Paris, September 1998.

5. PCI Ad Hoc Committee on PrecastWalls, Design for Lateral Force Re-sistance with Precast Concrete ShearWalls, PCI JOURNAL, V. 42, No. 5,September-October 1997, pp. 44-65.

6. Stanton, J. F., Stone, W. C., andCheok, G. S., A Hybrid ReinforcedPrecast Frame for Seismic Regions,PCI JOURNAL, V. 42, No. 2, March-April 1997, pp. 20-32.

7. Palmieri, L., Sagan, E., French, C.,and Kreger, M., Ductile Connectionsfor Precast Concrete Frame Systems,Paper No. SP162-13, Mete A. SozenSymposium, ACI SP 162, AmericanConcrete Institute, Farmington Hills,MI, 1997, pp. 313-355.

8. Stanton, J. F., MacRae, G., Sugata, M.,and Day, S., Preliminary Test Reportof a Hybrid Frame Exterior Beam-Col-umn Specimen Test, University of

Washington, Seattle, WA, March1999.

9 . Palmieri, L., and French, C., DuctileMoment-Resisting Connections forPrecast Frames in Seismic Regions:Experimental Study, Manuscriptsubmitted to PCI JOURNAL for p u b l i c a t i o n .

1 0 . Palmieri, L., and French, C., DuctileMoment-Resisting Connections forPrecast Frames in Seismic Regions:Numerical Simulation, Manuscriptsubmitted to PCI JOURNAL for p u b l i c a t i o n .

11. Schultz, A. E., and Magana, R. A.,Seismic Behavior of Connections inPrecast Concrete Walls, Paper No.SP162-12, Mete A. Sozen Sympo-sium, ACI SP 162, American ConcreteInstitute, Farmington Hills, MI, 1996,pp. 273-311.

12. Kurama, Y., Pessiki, S., Sause, R., andLu, L.-W., Seismic Behavior and De-sign of Unbonded Post-Tensioned Pre-cast Concrete Walls, PCI JOURNAL,V. 44, No. 3, May-June 1999.

13. Perez, F., Pessiki, S., and Sause, R.,Lateral Load Behavior and Design ofUnbonded Post-Tensioned PrecastConcrete Walls with Ductile VerticalJoint Connectors, ATLSS Report No.99-01, Center for Advanced Technol-ogy for Large Structural Systems,Lehigh University, Bethlehem, PA,January 1999.

14. El-Sheikh, M., Sause, R., Pessiki, S.,Lu, L.-W., and Kurama, Y., SeismicAnalysis, Behavior, and Design of Un-bonded Post-Tensioned Precast Con-crete Frames, PRESSS Report No.98/04 (also Report No. EQ-97-02),Department of Civil and Environmen-tal Engineering, Lehigh University,Bethlehem, PA, November 1997.

15. El-Sheikh, M., Sause, R., Pessiki, S.,and Lu, L.-W., Seismic Behavior ofUnbonded Post-Tensioned PrecastConcrete Frames, PCI JOURNAL, V.44, No. 3, May-June 1999.