Engineering Structures 27

o

Han

foeThe objective of this study was to investigate the seismic behaviors of columns in Ordinary Moment Resisting Concrete Frames (OMRCF)and Intermediate Moment Resisting Concrete Frames (IMRCF). For this purpose, two three-story OMRCF and IMRCF were designedaccording to the minimum design and reinforcement detailing requirements specified in ACI 318-02. This study assumed that the buildingwas located in seismic zone 1, as classified by UBC. According to ACI 318-02 the reinforcement detailing requirements for OMRCF areless stringent than those for either IMRCF or SMRCF (Special Moment Resisting Concrete Frames). Tests were carried out to evaluate theseismic behaviors of OMRCF and IMRCF columns using 2/3 scale model columns. Each column was considered as consisting of an upperpart and lower part divided at the point of inflection. Quasi-static reversed cyclic loading was applied to the specimens with either constantor varying axial forces. The test variables of this experimental study were the type of axial force (constant and varying, and low and high),the existence of lap splices (with or without lap splice) and type of moment resisting concrete frame (OMRCF or IMRCF). It was observedthat all OMRCF and IMRCF column specimens had strength larger than that required by ACI 318, and they had drift capacities greater than3.0% and 4.5%, respectively. However, the drift capacity of specimens varied with respect to the existence of lap splices and the spacing oflateral reinforcement at column ends. 2005 Elsevier Ltd. All rights reserved.Keywords: Seismic behavior; Columns; Reinforcement details; Seismic design

1. Introduction

Moment frames have been widely used for seismicresisting systems due to their superior deformation andenergy dissipation capacities. A moment frame consistsof beams and columns, which are rigidly connected. Thecomponents of a moment frame should resist both gravityand lateral load. Lateral forces are distributed according tothe flexural rigidity of each component. ACI 318-02 [1]states the design and reinforcement detailing requirementsfor each type of moment frame and each earthquakerisk level. The type of moment frame should be selected

moderate and high according to seismic zones specified inUBC [2]. Seismic design category is specified in NHERP[3] and IBC [8].

ACI 318-02 classifies concrete moment frames intothree types: Ordinary Moment Resisting Concrete Frame(OMRCF); Intermediate Moment Resisting Concrete Frame(IMRCF); and Special Moment Resisting Concrete Frame(SMRCF). Fig. 1 shows the minimum column reinforcementdetails for the columns of OMRCF, IMRCF, and SMRCFspecified in ACI 318 and notes [1,4].

OMRCF and IMRCF are the most popular types ofSeismic behaviors of columns inresisting con

Sang WhanDepartment of Architectural Engineering, Hany

Received 30 March 2004; received in revisedAvailable onlin

Abstractaccording to levels of seismic risk or seismic designcategory. Seismic risk levels can be classified into low,

Corresponding author. Tel.: +82 2 2290 1715; fax: +82 2 2291 1716.E-mail address: swhan@hanyang.ac.kr (S.W. Han).

0141-0296/$ - see front matter 2005 Elsevier Ltd. All rights reserved.doi:10.1016/j.engstruct.2005.01.012(2005) 951962www.elsevier.com/locate/engstruct

rdinary and intermediate momentcrete framesan, N.Y. Jeeg University, Seoul 133-791, Republic of Korea

rm 31 January 2005; accepted 31 January 200510 March 2005moment frames in low to moderate seismic zones. Thedesign and reinforcement requirements for OMRCF andIMRCF are less stringent than those for SMRCF. Whena large earthquake occurs, SMRCF is expected to havesuperior ductility and provide superior energy dissipationcapacity. In current seismic design procedures, design base

in

m

Notation

Ag grossf c specify speci

reinfMmax maxiMACI mom

proceVACI nomi

99max maximax maxi

shear force shaldivided by R faoverstrength, restructural systemhighest R factoOMRCF becausshould be noteddecreasing R fac

This study focin moment resisOMRCF and IMto ACI 318-02.story of those bseismic behavior

2 2 = 8)f the exterior andMRCF (2) weresplices whereasus longitudinal

ence in OMRCFf the transversesi-static reversedwith either fixed

pe of axial forceexistence of lap

type of momentCF).

ic performanceaximum lateral

mpared with thecodes. Moreover,evaluation of theum method.ic performance

ames with non-re frame and aarea of columnfied compressive strength of concretefied yield strength of nonprestressedorcementmum moment resistance of the specimenent capacity calculated using ACI 318-99duresnal shear strength according to ACI 318-

mum displacementmum drift angle

l be calculated by elastic strength demandctor (1). R factor accounts for ductility,dundancy, and damping inherent in a. Current design provisions assigned the

r to SMRCF, and the lowest R factor toe of its less stringent design requirement. Itthat design base shear becomes larger withtor.used on the behaviors of first story columnsting frames. For this purpose, three-storyRCF buildings were designed according

Experimental test of columns in the firstuildings was carried out to investigate the

Eight 2/3 scale column specimens (2representing the lower and upper part (2) ointerior columns (2) in the OMRCF and Imade. Lower column specimens had lapupper column specimens had continuoreinforcement. The most significant differand IMRCF columns is the spacing oreinforcement at the end of the column. Quacyclic loading was applied to the specimensor varying axial forces.

The test variables in this study were ty(constant and varying, and low and high),splices (with or without lap splice) andresisting concrete frames (OMRCF or IMR

2. Previous studies

Han et al. [6,7] have evaluated the seismof a three-story OMRCF. In their studies mforce was measured during the test and codesign base shear force specified in currentthis study carried out seismic performancethree-story OMRCF using a capacity spectr

Lee and Woo [9] investigated the seismof low-rise reinforced concrete (RC) frseismic detailing. They considered a ba952 S.W. Han, N.Y. Jee / Engineer

Fig. 1. Minimum reinforceof the columns. This study considered thatg Structures 27 (2005) 951962

ent details for columns.

a column consisted of an upper part and lower part dividedat the point of inflection.masonry infilled frame. An experiment was conducted

StS.W. Han, N.Y. Jee / Engineering

using 1/5 scaled specimens having two bays and threestories. They found that RC moment frames had a certainseismic resistance even though they were designed withoutconsidering seismic loads. Furthermore, masonry infill wasbeneficial to the seismic performance of the frame.

Aycardi et al. [10] investigated the behavior of gravityload-designed reinforced concrete column members undersimulated seismic loading. They tested four columnspecimens governed by flexure under reversed cyclic loadingat increasing drift amplitudes. It was observed that allcolumn specimens were capable of sustaining at least twocycles of loading at a 4% drift angle.

Lynn et al. [11] tested eight reinforced concrete columnspecimens having typical details of those built before themid-1970s. This study observed that column specimens thatwere governed by shear experienced gravity load failuresoon after loss of lateral force resistance. When flexurewas dominant in specimens, gravity load capacity wasmaintained to large displacement.

3. Experimental program

3.1. Design of building frames

Typical three-story office buildings were designed.Seismic resistance for these buildings was provided byOMRCF and IMRCF. The dimension and design loads wereadopted from Han [6]. Fig. 2 shows the dimensions of thebuilding.

Specified compressive strength ( f c) of concrete wasassumed to be 23.5 MPa. Longitudinal reinforcement andreinforcement for hoop and stirrup were assumed to have ayield strength ( fy) of 392.3 MPa. Design loads for a typicaloffice building were used (5.2 kPa for dead load and 2.4 kPafor live load). The section of all columns and beams in themodel frames were assumed to be 330 mm 330 mm and250 mm 500 mm, respectively.

The buildings were assumed to be located in seismiczone 1 as classified in UBC [2]. Member forces wereobtained using SAP 2000 (2000). Gravity loads governed themember design. Since the seismic load was small, OMRCFand IMRCF had the same member sizes as well as thesame amount of reinforcement except for the transversereinforcement in columns and beams (see Fig. 3). Thisdesign was desirable for this study since this study attemptedto investigate the effect of different reinforcement detailsof OMRCF and IMRCF columns on seismic behaviors ofthose columns without interference of other factors such asamount of longitudinal reinforcement and different membersizes.

3.2. Column specimens

As shown in Fig. 2, only the columns in the first storywere evaluated since these columns should resist the largest

axial and lateral forces during an earthquake. The exteriorructures 27 (2005) 951962 953

(a) Plan view.

(b) Elevation.

Fig. 2. Building layouts.

columns in the first story of the two model frames weredesigned for an axial force of 644.3 kN and a bendingmoment of 37 kN m, and the interior columns were designedfor an axial load of 1234.7 kN and a bending moment of47.1 kN m. It is noted that those design forces contain loadfactors.

All columns had a section of 330 mm 330 mmand contained four longitudinal reinforcements (D19(19.1 mm diameter), fy = 392.3 MPa). Columnlongitudinal reinforcement ratio () was 1.01%, whichslightly exceeded a minimum longitudinal reinforcing steelof 1.0% (Section 10.9.1 in ACI 318-02).

Maximum shear force in the first story columns inducedby factored gravity loads was 31.4 kN. According to theformula in Section 11.3.1.2 of the ACI 318-02, the concreteshear strength of the first story columns (Vc) was calculatedas 73.5 kN. Minimum tie reinforcement (D10 with adiameter of 9.53 mm) with spacing of 300 mm was placedthroughout the column in the OMRCF (Section 7.10.5.2in ACI 318-02) and the first tie reinforcement was placed

150 mm from the slab or footing surface according

rin

mn

spacing in the OMRCF columns. The yield strength of D10was 3for tespliceFigs.

Itpart othe lomadeof copointearthq

Figof thethe cwere

specimand esplicethe orcalcuD13 (usedrepreframe

acedacks)).

fora, a

gatetests

inare

umn

D6tests

t themn)of a92.3 MPa. A lap splice length of 525 mm was requirednsion according to Section 12.15 in ACI 318-02. Laps in a column were placed just above the slabs (see1 and 3).is worthwhile to note that as shown in Fig. 1 the upperf columns has continuous longitudinal bars whereaswer part of columns has lap splices. Thus this studyspecimens representing the upper part and lower part

lumns (see Fig. 2). It was assumed that the inflectionwas located at the mid-height of the column duringuakes.. 3 shows the reinforcement details and dimensions2/3 scale column test specimens. Table 1 describes

haracteristics of the specimens. All reinforcementsscaled by 2/3 for test specimens. A total of eightens were made (upper and lower part (2) of interior

xterior columns in OMRCF and IMRCF (2)). Laplength (350 mm) was also simply scaled by 2/3 of

iginal length (525 mm). No attempt was made to re-late lap splice length for scaled reinforcement. Rebar12.7 mm diameter) and D6 (6.35 mm diameter) werefor longitudinal and transverse tie reinforcement tosent 2/3 of D19 and D10 in the columns of model

using reinforcement of D13 (13 mm diameter) were plto resist the forces transferred from the specimen. No crwere detected in the column base after test (see Fig. 4(a

3.3. Material tests

Based on concrete trial mixes from various recipesattaining the 28-day target strength ( f c) of 23.5 MPdesign mix was determined. The maximum size of aggrefor 2/3 scale model specimens was 25 mm. Cylinderwere performed. Each concrete cylinder was 200 mmheight and 100 mm in diameter. Cylinder test resultsgiven in Table 2.

Longitudinal and transverse reinforcement in the colspecimens (2/3 scale) were D13 (13 mm diameter) and(6 mm diameter), respectively. The results of materialare given in Table 3.

3.4. Loading and measurements

All eight-column specimens were laterally loaded alevel of 1000 mm above the base (mid-height of a colu(see Fig. 4). It was assumed that the inflection point954 S.W. Han, N.Y. Jee / Enginee

Fig. 3. Colu

to the requirement of the first tie location specified inSection 7.10.5.4 in ACI 318-02.

Fig. 3 shows that the first tie was placed 100 mm from theslab in 2/3 scale specimens. The longitudinal reinforcementdetails in the IMRCF columns and the OMRCF columnswere exactly the same. Tie spacing in the plastic hingingregion of the IMRCF columns, however, was half of the ties.g Structures 27 (2005) 951962

specimens.

The dimension of column base is shown in Fig. 3.The longitudinal reinforcement of column specimens wasanchored to the column base satisfying development lengthrequired by ACI 318-02 (chapter 12). In order to fix thespecimen to the strong floor, four high strength bolts havinga diameter of 70 mm were installed (see Fig. 4(a)). All boltsare fully tightened to provide specimens with fixed baseconditions. In column base, sufficient closed type stirrupscolumn is located at mid-height of the column.

Str

TableChara

lice

OM

O(1) (

Qtrolleing afluctusigniing w

pec-mns

singloadFig. 4. Test setting.

1cteristics of the column specimens

Location Specimen name Loading plan Lap sp

RCF (IMRCF)

Interior Lower part OIL (IIL) Constant column axial loadUpper part OIN (IIN)

(P = 0.28)

Exterior Lower part OEL (IEL) Varying column axial loadUpper part OEN (IEN)

(P = 1.83V + 17.1)

I L (1): OMRCF (O), IMRCF (I) : having lap splices2) (3) (2): Interior (I), Exterior (E) : not having lap splices(3): with lap splice (L), without lap splice (N)

uasi-static reversed cyclic loading (displacement con-d) was applied. Two cycles were applied for each load-mplitude. The loading history is shown in Fig. 5. Sinceation of axial loads in the exterior columns was more

ficant than in the interior columns, varying axial load-

constant axial load was applied to the interior column simens. Larger axial loads were applied to interior coludue to a larger tributary area.

Axial loads were calculated by frame analysis uSAP 2000 commercial software [5] without consideringS.W. Han, N.Y. Jee / Engineeringas applied to the exterior column specimens, whereas auctures 27 (2005) 951962 955factors. Since specimens were scaled by 2/3, calculated

rin

adin

also obtained using SAP 2000 under first mode lateral forceprofile. The range of varyspecimens is 0.07 0.20

As shown in Fig. 4, olateral forces and twoaxial forces. Three pairsat the column faces toFig. 4(a)). However, curvthis paper. One LVDT wdisplacement of the stuwas observed that slip bnegligible during the testspecimens was not modifipassing through stubs ofnot measured since it wasa rigid base condition. Foto measure the lateral dis

4.2. Hysteretic performance

CF and IMRCF columnsctively. Fig. 6(a), (b) and

for the interior specimenscolumn specimens OEL), (b) and (c), (d) showns IIL and IIN, and forand IEN of the IMRCF.

ost elastically until a drift

, strength of specimensective of the existence

n OIN exhibited greaterIL specimen having laping axial loads for exterior columnAg f c .

ne actuator was installed to controlactuators were used to controlof linear transducers were placedcapture column curvatures (see

ature results were not included inas also placed to measure a lateralb of column specimens. Since itetween the stub and strong floor is, recorded displacement of columned. This was due to tightened boltsspecimens. Curvature of stubs wasassumed that tightened bolts made

ur linear transducers were installed

Hysteretic curves for the OMRare presented in Figs. 6 and 7, respe(c), (d) shows the hysteretic curvesOIL and OIN, and for the exteriorand OEN of the OMRCF. Fig. 7(athe curves for the interior specimethe exterior column specimens IELAll column specimens behaved almratio of 0.3%.

According to Fig. 6(a) and (b)OIN and OIL was similar irrespof lap splices. However, specimedeformation capacity than the O956 S.W. Han, N.Y. Jee / Enginee

Fig. 5. Lo

Table 2Concrete properties of the specimens

Design strength 28 day strength Strain at ultimate Youngs modulus(MPa) (MPa) strength, co (MPa)23.5 24.6 0.003 23,438

Table 3Reinforcing steel properties

Bar Yielding strength Yielding strain Ultimate strength Youngs modulus(MPa) (106) (MPa) (MPa)

D6 374 2206 598 176,520D13 397 2035 594 194,956

force for original frame members should be scaled by 4/9for the scaled model. Interior column specimens (OIL, OIN,IIL, IIN) were tested with a constant axial load of 333.4 kN(0.28 Ag f c ; Ag is the gross sectional area of a column). Forexterior column specimens, varying axial loads (P (axialforce) = 1.83V (lateral force) + Pg) were applied. Pg isgravity axial load, which is 167.6 kN, and V is lateralshear force acting on the column. This relationship wasplacements of specimens.g Structures 27 (2005) 951962

g history.

4. Test results and observation

4.1. Observations

Within a drift ratio of 1%, the first flexural crack wasobserved in all specimens. The lateral force causing the firstcrack in all specimens was about 25 kN. At a drift ratio of3%, the concrete cover started spalling at the column ends.It is noted that drift ratio was defined as a measured lateraldisplacement divided by the height of specimens.

In the exterior column specimens having lap splices(OEL, IEL), spalling and cracks (vertical and horizontal)were more prominent when applied lateral loading was inthe positive direction (the direction in which axial loadincreases). As mentioned earlier, varying axial loads (P =1.83V + 167.6 (kN)) were applied to exterior columnspecimens. Moreover, many vertical cracks in the region oflap splices were observed. Flexural cracks were relativelyuniformly distributed in the plastic hinging region of thespecimens.

In the final stage of the test, all column specimens faileddue to buckling of the longitudinal bars and crushing of theconcrete.splices. In the IMRCF interior column specimens, IIN and

g StS.W. Han, N.Y. Jee / Engineerin

(a) OIL.

(b) OIN.

(c) OEL.

(d) OEN.Fig. 6. Hysteretic curves of the OMRCF columns.ructures 27 (2005) 951962 957

(a) IIL.

(b) IIN.

(c) IEL.

(d) IEN.

Fig. 7. Hysteretic curves of the IMRCF columns.

in958 S.W. Han, N.Y. Jee / Engineer

IIL, (Fig. 7(a) and (b)), no discernible difference could befound in either strength or deformation capacity.

The strength of all the OMRCF and IMRCF interiorcolumn specimens (OIN, OIL, IIN, IIL) was similar,whereas OMRCF interior columns had inferior deformationcapacities to the IMRCF interior columns. Moreover,pinching was detected more prominently in OMRCFspecimens. Strength drops in the second cycle were,however, similar between the OMRCF interior columnspecimens and their corresponding IMRCF specimens.

In interior column specimens, IMRCF has a more stableand fuller hysteretic curve than OMRCF. In particular, theimprovement was more prominent when columns had lapsplices (compare Figs. 6(a) and 7(a)). The better hystereticbehavior in interior columns of IMRCF was attributed tonarrower spacing of transverse reinforcement in the plastichinging region of the column.

Fig. 8(a) shows the envelope curves extracted fromthe hysteretic curves of the interior column specimens.According to this figure, the IMRCF specimens had fullerenvelope curves than the OMRCF specimens. Thus, itis expected that interior column specimens of IMRCFexhibited better seismic behavior than those of OMRCF.Since larger axial force was applied to the interior specimensdue to a larger tributary area of gravity loads, narrowerspacing of transverse reinforcement in the plastic hingeregion improved the seismic behavior of the columnssignificantly.

Figs. 6(c) and (d), and 7(c) and (d) show hystereticbehaviors of the exterior column specimens of the OMRCF(OEL, OEN) and IMRCF (IEL, IEN), respectively. In thesefigures, the positive loading direction was the direction inwhich the axial load increases.

The hysteretic behaviors of the exterior OMRCFand IMRCF column specimens were relatively similarbetween the OMRCF specimens and corresponding IMRCFspecimens (OEN versus IEN, and OEL versus IEL).However, the strength drop of specimen OEL in thesecond cycle was greater than that of specimen IEL. Thespecimens having lap splices showed inferior deformationcapacities compared to the specimens without lap splices(OEL versus OEN, and IEL versus IEN). Unlike interiorcolumns, hysteretic behavior was not significantly improvedby narrower spacing of lateral reinforcement (OEL versusIEL, and OEN versus IEN). Fig. 8(b) shows the envelopecurves. From this figure, the specimens having lap splicesalso had narrower envelope curve. Also, this figure showsthat the envelope curves of the exterior column specimensof OMRCF and IMRCF are quite similar (OEN versusIEN, and OEL versus IEL). However, more strength drop inspecimen OEL was observed than that in IEL in the negativeloading direction. In exterior column specimens, lateralreinforcement spacing does not affect seismic behavior asmuch as in the interior column specimens. However, it wasobserved that seismic behavior in exterior column specimens

became poor when they had lap splices.g Structures 27 (2005) 951962

4.3. Maximum strength

Table 4 shows the ratio of actual maximum strength Mmaxobtained from experimental tests to the calculated strength(MACI). Actual maximum strength Mmax is calculated bymaximum shear force times the height of the columnspecimen (Mmax = Vmax 1m(h)). Fig. 9 shows PMinteraction curves of column specimens. This figure alsoincludes the actual moment strength (Mmax) of columnspecimens. All nominal strength (MACI) in this study wascalculated using material strength obtained from materialtests. According to Fig. 9 and Table 4(7), all OMRCFand IMRCF column specimens have strength larger thancalculated nominal strength.

In exterior column specimens, moment capacities (Mmax)in the positive loading direction were greater than those inthe negative loading direction. Since axial loads in the testare not large, strength of column specimens is below thebalanced failure point in PM interaction curve. Thus, inexterior column specimens, moment strength increases asaxial force is larger (positive lateral loading direction).

In the PM interaction curve (Fig. 9), calculated momentstrength (MACI) of exterior column specimens in the positiveloading direction is larger than that in the negative loadingdirection as well. In those specimens with relatively lowaxial loads, the ratio of maximum strength (Mmax) tocalculated strength (MACI) was almost the same as that ofthe interior column specimens.

It is worthwhile to note that all column specimenswere designed according to the minimum design and detailrequirements in ACI 318-02. Since all specimens in thisstudy were governed by flexure rather than shear (seeTable 4(10)), different results can be obtained for columnsgoverned by shear.

4.4. Deformation capacity

In this study the maximum drift ratio (max = max/h)was used to determine the drift capacities, where max isthe maximum drift and h is the height of column specimens.The maximum drift was obtained when the lateral strengthof the specimen was reduced by 20% of maximum strength.Table 4 and Fig. 10 show the maximum drift ratios of thespecimens.

Table 4 shows that the drift capacities (max) ofOMRCF and IMRCF exceeded 3.0% and 4.5%, respectively.Specimen IEN had the largest drift capacity (5.98%) in thenegative loading direction, whereas specimen OEL had theleast drift capacity (3.00%) in the negative loading direction.

Among the interior column specimens that resisted alarger constant axial load throughout the test, the IMRCFcolumns had greater drift capacity. This is due to thenarrower spacing of transverse reinforcement in the plastichinging region. It should be noted that the drift capacity ofthe specimen IIL having lap splices was even greater than

that of OIN without lap splices. Moreover, the difference

S.W. Han, N.Y. Jee / Engineering Structures 27 (2005) 951962 959(a) Interior columns.

(b) Exterior columns.

Fig. 8. Envelope curves.Fig. 9. Maximum strength.

in

4).2.4.4.3

.3

.0

.5

.0

.0

.5

IIN .1 47.6 50.5 6.01 5.05 38.8 1.22 9.83 3.88 0.39IMRCFIEL + 0.2 45.5 45.2 5.82 4.52 33.0 1.55 9.36 3.30 0.35 0.07 33.2 47.0 5.90 4.70 26.2 1.10 8.72 2.62 0.30IEN + 0.2 49.4 48.0 5.9 4.80 33.0 1.37 9.36 3.30 0.35 0.07 34.2 59.8 6.27 5.98 26.2 1.31 8.72 2.62 0.30

(2) = maximum shear force (kN), (3) = maximum displacement (mm), (4) = displacement ductility (= max/y), where y is yield displacemet, (5)= maximum drift angle (%), (6) = moment capacity calculated using ACI 318-02 procedures (kN m), (7) = ratio of the maximum moment capacity,Mmax(Vmax h) to MACI (h: height of specimen (m)), (8) = nominal shear strength according to ACI 318-02 (kN), (9) = shear force corresponding toMACI or 2MACI/ l, where l = the column clear height (kN), (10) = ratio of VP to VACI.

Fig. 10. Drift capacity (%).

between drift capacities of specimens OIL and IIL wasgreater than that of specimens OIN and IIN. Thus, spacing oftransverse reinforcement in the plastic hinging region affectsthe drift capacity of the interior columns significantly. Thiswas more prominent when the specimen had lap splices.It is noted that larger gravity load (0.28Ag f c) was appliedto interior column specimens. Also, in interior columnspecimens of OMRCF, the existence of lap splices in thecolumn specimen prominently influences the drift capacity.In contrast, the existence of lap splices did not affect driftcapacities of IMRCF column specimens.

In the exterior column specimens that resisted varyingaxial loads during the test, the existence of lap splices alsoaffected drift capacities. Specimens IEL and OEL having lap

drift capacities of the exterior columns of the OMRCF andIMRCF was small compared to that between the interiorcolumns of the OMRCF and IMRCF. Thus, in the exteriorcolumns, the spacing of transverse reinforcement was notinfluential. It is noted that this study considered a three-storybuilding, in which axial forces in columns are smallcompared with those in high-rise buildings. When axialforces in columns become large, the spacing of transversereinforcement may be influential even on the behavior ofexterior columns.

This study also calculated the displacement ductilityof each specimen, which is shown in Table 4. Yield drift(y or y) was obtained from a bilinear representation offorce (V ) versus drift ratio ( or ). Secant stiffness was960 S.W. Han, N.Y. Jee / Engineer

Table 4Test result of specimens

Specimen PuAg f c Vmax max (1) (2) (3) (

OMRCF

OIL +0.28

49.1 34.9 4 49.5 35.3 4

OIN + 46.9 44.5 4 48.8 44.5 4

OEL + 0.2 36.9 42.6 4 0.07 37.2 28.5 3OEN + 0.2 48.5 45.3 4 0.07 28.3 58.1 6

IIL +0.28

48.7 53.8 6 45.9 53.5 5+ 46.0 49.5 6splices had smaller drift capacities. The difference betweeng Structures 27 (2005) 951962

u MACIMmaxMACI

VACI VPVP

VACI(5) (6) (7) (8) (9) (10)

6 3.49 36.8 1.33 74.3 37.9 0.516 3.53 36.8 1.33 74.3 37.9 0.511 4.45 36.8 1.28 74.3 37.9 0.519 4.35 36.8 1.33 74.3 37.9 0.514 4.26 33.3 1.25 69.6 32.6 0.472 3.01 25.5 1.25 63.4 26.0 0.419 4.53 33.4 1.45 69.6 32.6 0.475 5.81 25.5 1.11 63.4 26.0 0.41

2 5.38 38.8 1.28 9.83 3.88 0.397 5.35 38.8 1.21 9.83 3.88 0.390 4.95 38.8 1.34 9.83 3.88 0.39used to connect the origin to the point of 0.75Vmax. The

St

y

dissipated a similar amount of energy until cycle 6 (1%drift ratio). Beyondifferent amount o

Specimens IIN,of energy until faicapacity than otheIEL dissipated sim(5%). Since specim(cycle 11), the amothan specimens Othe least amount ospecimens.

5. Conclusions

This study invein OMRCF and IMwere modeled anand reinforced in

(3) In the exterior column specimens, no discerniblers were foundspecimens. Theerior hystereticns without lapapplied to the

t of transverseinfluential thanat the effect ofes important torticularly in the

ore stable andnterior columnmn specimens,pe of hystereticinterior columnlices was notd that cycle, every specimen dissipated af energy.IIL, and IEN dissipated similar amounts

lure. They have larger energy dissipationr specimens. Specimens OEN, OIN, andilar amounts of energy until cycle 13en OIL failed in an earlier loading stageunt of energy of specimen OIL is smallerEN, OIN and IEL. Specimen OEL hasf energy dissipation capacity among the

stigates the seismic behaviors of columnsRCF. For this purpose eight specimens

d tested. The specimens were designed

differences in the hysteretic behaviobetween OMRCF and IMRCF columnspecimens having lap splices showed infbehavior to the corresponding specimesplices. Since smaller axial force wasexterior column specimens, the effecreinforcement spacing seems to be lessinterior columns. However, it is noted thtransverse reinforcement spacing becomexterior specimen having lap splices, panegative loading direction.

(4) IMRCF interior column specimens had mfuller hysteretic curves than OMRCF ispecimens. In OMRCF interior coluexistence of lap splices changed the shacurve significantly. However, in IMRCFspecimens, the change due to lap spS.W. Han, N.Y. Jee / Engineering

Fig. 11. Energ

maximum drift (max, max) was approximated as the driftratio corresponding to the strength deteriorated by 20% ofVmax (0.8 times Vmax). As shown in Table 4, all specimenshave a ductility capacity exceeding 3.0. The IMRCF exteriorcolumn specimen without lap splices (IEN) has the largestductility capacity (6.27) in the negative loading directionwhich has the least axial loading condition ( PuAg f c = 0.07).In contrast, the OMRCF exterior column specimen with lapsplices (OEL) has the smallest ductility capacity (3.02) inthe negative loading direction. Specimen IEL has a ductilitycapacity of 5.90 in the negative loading direction. Thus theeffect of lateral reinforcement spacing on displacement andductility capacity becomes important to all interior columnspecimens, and exterior specimens having lap splices.

4.5. Energy dissipation

Fig. 11 shows the amount of energy dissipated in eachloading cycle. According to this figure, all specimenscompliance with the ACI 318-02 forructures 27 (2005) 951962 961

dissipation.

OMRCF and IMRCF. It should be noted that all specimenswere governed by flexure rather than shear. Thus, differentresults may be obtained for columns governed by shear. Theconclusions of this study are as follows:

(1) The strength of all OMRCF and IMRCF columnspecimens exceeded the strength calculated with thecode formula (ACI 318-02). Thus, the OMRCF andIMRCF columns had satisfactory strength irrespectiveof the existence of the lap splices and the transversereinforcement spacing in the plastic hinging region.

(2) The IMRCF interior column specimens had superiordrift capacities compared to the OMRCF columnspecimens. Existence of lap splices influenced driftcapacities of the OMRCF columns. However, the effectof lap splice in the IMRCF columns was not as large asthat in the OMRCF specimens. This is attributed to thenarrower transverse reinforcement spacing in the plastichinging region of the IMRCF.noticeable.

962 S.W. Han, N.Y. Jee / Engineering Structures 27 (2005) 951962

(5) In exterior column specimens of OMRCF and IMRCF,hysteretic curves became narrower when they had lapsplices. However, the hysteretic curves of OMRCFexterior columns are similar to those of correspondingIMRCF exterior columns.

(6) According to the test results, the OMRCF and IMRCFcolumn specimens had drift capacities greater than3.0% and 4.5%, respectively. Ductility capacity of thosespecimens exceeded 3.01 and 4.53, respectively.

Acknowledgement

The support of Hanyang University is greatly acknowl-edged.

References

[1] American Concrete Institute. Building code requirements forreinforced concrete, ACI 318-95, 99, 02, Detroit, Michigan, 1995,1999, 2002.

[2] Uniform Building Code (UBC). International conference on buildingofficials. 1997.

[3] Building Seismic Safety Council. NEHRP recommended provisionsfor the development of seismic regulation for new buildings, Part 1and 2, provisions and commentary, Washington, DC: FEMA; 1994,1997.

[4] Portland Cement Association. Notes on ACI 318-99, 02 building coderequirements for structural concrete, Skokie, Illinois, 1999, 2002.

[5] Computers and Structures Inc. SAP2000, 1997.[6] Han SW, Kwon OS, Lee L-H, Foutch DA. Seismic performance

evaluation of ordinary moment resisting concrete frames. SRS No.633, Urbana, Illinois; 2002 November.

[7] Han SW, Kwon OS, Lee L-H. Evaluation of the seismic performanceof a three-story ordinary moment resisting concrete frame. Earthquakeengineering and structural dynamics, vol. 33. Wiely and Sons; 2004.p. 66985.

[8] BOCA, ICBO, SBCCI. International Building Code. 2000.[9] Lee H-S, Woo S-W. Effect of masonry infills on seismic performance

of a 3-storey RC frame with non-seismic detailing. Earthquakeengineering and structural dynamics, vol. 31. Wiely and Sons; 2002.p. 35378.

[10] Aycardi LE, Mander JB, Reinhorn AM. Seismic resistance ofreinforced concrete frame structures designed only for gravity loads:experimental performance of subassemblages. Structural JournalAmerican Concrete Institute 1994;91:55263.

[11] Lynn A, Moehle J, Mahin SA, Holmes WT. Seismic evaluationof existing RC building columns. Earthquake Spectra 1996;12(4):71539.

Sang Whan Han. Associate Professor, Dept. of Architectural Eng.,Hanyang Univ., Seoul 133-791, Korea. Ph.D. (1994), University of Illinoisat Urbana-Champaign, Professional Engineer (US, Korea).

N.Y. Jee. Professor, Dept. of Architectural Eng., Hanyang Univ., Seoul 133-791, Korea. Ph.D. (1992), Tokyo National University.

Seismic behaviors of columns in ordinary and intermediate moment resisting concrete framesIntroductionPrevious studiesExperimental programDesign of building framesColumn specimensMaterial testsLoading and measurements

Test results and observationObservationsHysteretic performanceMaximum strengthDeformation capacityEnergy dissipation

ConclusionsAcknowledgementReferences