You are on page 1of 14

2003 ECI Conference on Advanced Materials for Construction of Bridges, Buildings, and Other Structures III

% FILE:

/main/production/doc/data/journals/services.bepress.com/advanced materials/assets/location.inc Switzerland% FILE: (cont)

/main/production/doc/data/journals/services.bepress.com/eci/assets/proceedings/pdftitlepage.te

Editors: % FILE: /main/production/doc/data/journals/services.bepress.com/advanced materials/assets/editors.incV Mistry, P.E., Oce of Bridge Technology, Federal Highway Administration, USA Dr. Atorod Azizinamini, Ph.D., P.E., Civil Engineering Department, University of Nebraska, USA John M. Hooks, P.E., Oce of Infrastructure Research & Development, Federal Highway Administration, USA % FILE: /main/production/doc/data/journals/services.bepress.com/eci/assets/proceedings/denes.tmpl (cont)

Year

Paper

Seismic Performance of Precast Column-Foundation Connection Assembled by Post-Tensioning


Minehiro Nishiyama Fumio Watanabe

Seismic Performance of Precast Column-Foundation Connection Assembled by Post-Tensioning


Abstract
In order to develop design recommendations for column-foundation connection assembled by post-tensioning in seismic regions, cyclic loading tests were carried out on 14 test units simulating such kind of connections under earthquake loading. The tests were consisted of two series: Series A was mainly for comparison between precast reinforced and precast prestressed concrete column-foundation connections, and Series B for investigating differences between test units with grouted and ungrouted tendons. The main experimental parameter other than the above was an axial load level.

Kyoto University Kyoto University

http://services.bepress.com/eci/advanced materials/5 Copyright c 2005 by the authors.

Nishiyama and Watanabe: Seismic Performance of Precast Column-Foundation Connection Assem

SEISMIC PERFORMANCE OF PRECAST COLUMN-FOUNDATION CONNECTION ASSEMBLED BY POST-TENSIONING


Minehiro Nishiyama and Fumio Watanabe Built Environment Materials and Structural Systems Department of Urban and Environmental Engineering Kyoto University Kyoto 606-8501, JAPAN T & F: 81-75-753-5747; E: mn@archi.kyoto-u.ac.jp ABSTRACT In order to develop design recommendations for column-foundation connection assembled by post-tensioning in seismic regions, cyclic loading tests were carried out on 14 test units simulating such kind of connections under earthquake loading. The tests were consisted of two series: Series A was mainly for comparison between precast reinforced and precast prestressed concrete column-foundation connections, and Series B for investigating differences between test units with grouted and ungrouted tendons. The main experimental parameter other than the above was an axial load level. INTRODUCTION Post-tensioned precast construction has been getting popular in Japan because of the following advantages over conventional cast-in-situ construction: 1) Easier framing and less concrete casting at construction sites. 2) Shear transfer at the interface between members which are connected is easily achieved by friction due to prestressing force. 3) Full depth crack opening at the beam-column interface under cyclic loading at a large inelastic deformation, which may result in pinched hysteresis curves, is suppressed by prestress. 4) Permanent displacement after major earthquakes is smaller than that for ordinary reinforced concrete. One type of the post-tensioned connections used in practice is a column-foundation connection. Ordinary precast reinforced concrete system is also often used. In Japan non-prestressed precast columns are more popular than prestressed ones. However, from the viewpoint of construction and restriction of construction time, there is a case that precast prestressed concrete system may be a better solution. In Japan use of unbonded tendons for primary seismic resistant members like girders, columns and structural walls had been prohibited. This year the code has been revised and now unbonded tendons can be used for structural members if a kind of displacement-based design different from the currently used allowable stress based design is utilized, and some measures is taken against tendon fracture: protection for girders from falling down.

Produced by The Berkeley Electronic Press, 2011

Advanced Materials for Construction of Bridges, Buildings, and Other Structures III [2003], Vol. P05, Article 5

In this paper, two series of loading tests are reported. One is Series A in which dual-phase composite prestressing steel bars are used, and reinforced concrete precast column-foundation connections are compared with prestressed ones in terms of seismic performance. The other is Series B for investigating differences in seismic performance between test units assembled by grouted and ungrouted tendons. EXPERIMENTAL WORK The experimental work is divided into two test series; in Series A dual-phase composite prestressing steel bars were used to assemble precast column-foundation connections. Ordinary precast reinforced concrete column-foundation connections were also constructed and tested. In addition, three test units were assembled using ordinary prestressing steel bars. Test series B was planned for investigating differences between precast prestressed column-foundation connections with grouted and ungrouted tendons. Series A Dual-Phase Composite Prestressing Steel Bars Unlike ordinary prestressing steel which was once heated to 900-1000C, ferrite with 0.48% carbon content was chauffaged to 850C, which turned it to dual-phase composite: ferrite and austenite. Water cooling changed the austenite into martensite. This process produced dual-phase composite of high-strength martensite and normal strength ferrite. Thus, it has a monotonic load-deformation relationship which is better modeled by a trilinear than a bilinear approximation. Fig. 1 schematically indicates the comparison of tensile force-strain relationship between ordinary prestressing steel and dual-phase composite prestressing steel
Ordinary prestressing steel bar Force A

200 150 Load (kN) 100 50 0

Yield Load 146kN

13mm Ordinary prestressing steel bar 15mm Dual-phase composite steel bar

C B

Yield Load 127kN Dia. (mm) Area (mm2) Yield strength (MPa) Tensile strength (MPa) Elongation (%) Ordinary 13 132.7 1100 1145 12 Dual-phase 15 176.7 718 966 17

Dual-phase composite steel bar D O

Strain

0.5

1.5 2 Strain (%)

2.5

3.5

Fig.1 Tensile force - strain relationship of dual-phase composite prestressing steel and ordinary prestressing steel bars

Fig.2 Measured tensile force - strain relations for ordinary and dual-phase tendons

http://services.bepress.com/eci/advanced_materials/5

Nishiyama and Watanabe: Seismic Performance of Precast Column-Foundation Connection Assem

250 50 150 50

250 25mm Thick Plate 50 Loading Point 4- 13 Prestressing Tendons D6@30 4- 9 Longitudinal Steel Bars 15mm Mortar Joint 500 400 100 100

D6 Ties 150 4- 9 Steel Bars 250

4- 13 Prestressing Tendons

Column section

50

15

(UNIT:mm)

800

25mm Thick Plate

Fig.3 Prestressed concrete test unit of Series A bar. Since dual-phase composite steel bar should have a larger sectional area than ordinary prestressing steel to have approximately the same yield tensile force, the sectional area and the elastic stiffness of the dual-phase composite steel are larger than the ordinary prestressing steel. In the dual-phase composite prestressing steel bar the second point of change in stiffness (Point A) occurs when the high-strength part (martensite) yields. The rst point of change in stiffness (Point B) is used to model the nonlinear behavior due to yielding of the ordinary strength part (ferrite). Initial prestress is expected to be introduced to a stress (Point C) between the two points. Hysteresis loops between the two points contribute to the hysteresis energy dissipation. Tensile force-strain curves obtained from the tensile tests of the prestressing steel bars are shown in Fig. 2. They have approximately the same yield tensile force. Therefore, the dual-phase bar has a larger sectional area, and larger elastic stiffness than the ordinary bar. Test Units Eight precast column-to-foundation connections were constructed. Three of them were conventionally reinforced by non-prestressed ordinary strength steel. The other units were post-tensioned by prestressing steel bars. Their specications are summarized in Table 1. A typical post-tensioned test unit is illustrated in Fig.3. The introduced prestress corresponded to the stress larger than the rst yield point of the dual-phase composite prestressing steel bar. Thus, the dual-phase composite prestressing steel bar was expected to be effective for hysteresis energy dissipation in the early stage of loading. Effective prestressing forces at the time of testing were 426.7kN, 411.4kN, 418.9kN, 348.1kN and 356.5kN for PC1, PC2, PC3, PC4 and

Produced by The Berkeley Electronic Press, 2011

Advanced Materials for Construction of Bridges, Buildings, and Other Structures III [2003], Vol. P05, Article 5

PC5, respectively. Immediately after introduction of prestress grout was injected into the sheath. W/C ratio of the grout was 45%. The compressive strength of the grout attained 38.0MPa. The compressive strength of concrete used for the columns and the foundations of the test units were 35.7MPa and 38.8MPa, respectively. The joint mortar at the interface of the foundation and the column had a compressive strength of 56.9MPa. Table 1 Test units in Series A Longitudinal rebars Column axial load (kN) 550 (0.25)* 8-D13 (SD395) 980 (0.45)* -224 (-0.46)** 550 (0.25)* Ordinary prestressing steel bars 980 (0.45)* 4-13 (SBPR930/1080) -224 (-0.38)** 550 (0.25)* Dual-phase composite prestressing -224 (-0.44)** steel bars 4-15

Test unit RC1 RC2 RC3 PC1 PC2 PC3 PC4 PC5

*(): N/f'cAg, N: axial load, f'c: concrete compressive strength, Ag: column sectional area **(): N/fyAs, fy: yield strength of longitudinal rebar, As: total area of rebars

Testing Methods After the speci ed axial load was applied, horizontal load was quasi-statically applied to the top of the column. The rst loading cycle was up to the rst cracking load which was detected by a observer. Then, the load was reversed to the negative direction to as large displacement as the positive loading. This loading cycle was followed by a series of deection controlled cycles comprising two full cycles to each of the column rotation angles of 1/200, 1/100, 1/50, 1/33 and 1/25. General Behavior of Test Units Figure 4 shows the horizontal deection at the top of the column plotted against the corresponding load of the column for each unit. The horizontal load plotted in Fig. 4 includes horizontal component of the axial load. All test units was able to be loaded to the last loading cycles to the column rotation angle of 1/25. Prestressed Concrete vs. Reinforced Concrete The prestressed concrete test unit PC1 showed narrower hysteresis loops than the reinforced concrete unit RC1. They were subjected to 550kN (0.25fcAg). Prestressing force was equivalent to the axial compressive load of 0.19fcAg if loss of prestress due to column shortening was not considered. Equivalent viscous damping of each specimen is calculated and shown in Fig. 5. PC1 dissipated 1.58 times larger

http://services.bepress.com/eci/advanced_materials/5

Nishiyama and Watanabe: Seismic Performance of Precast Column-Foundation Connection Assem

hysteresis energy than RC1 at the column rotation angle of 1/100, but equivalent damping factor of RC1 is larger than that of PC1 at the larger displacement: 48% larger at 1/50 and 71% at 1/25 in the column rotation angle. For the test units under the axial load of 980kN, PC2 and RC2, both units were able to be loaded to a column rotation angle of 1/25. The hysteresis loops obtained are stable without pinching and large capacity reduction. RC2 dissipated larger hyseresis energy than PC2. Comparison of equivalent damping factor shows 21% larger equivalent damping at 1/100 and 31% at 1/25 of RC2 than PC2. For the test units subjected to tensile axial load, RC3 and PC3, less pinching of PC3 had been expected because of prestressing force which connected the column and foundation tightly. However, the actual hysteresis loops of PC3 are pinched and narrow. This is not because of slip at the interface between the column and the foundation or at the joint mortar. Displacements measured at the interface indicate larger transverse displacement in RC3 than in PC3. One of the possible reasons may be a stress-strain relationship of prestressing steel. Stiffness reduction due
1/100 1/50 1/100 1/50

200
Load, P (kN)

RC2
Load, P (kN)

200 100

PC2
0.45f'cAg 0.25f'cAg

RC2 RC1 RC3


RC

PC2 PC1 PC3


Ordinary

100 0 -100 -200 -30 -20


Axial load = 980kN (0.45f'cAg) Ordinary RC

PC4 PC5
Dualphase

0 -100 -200
Axial load = 980kN (0.45f'cAg) Ordinary ps steel bars

-0.46-0.38 fyAs

-10 0 10 Deflection,d (mm)


1/100 1/50

20

30

-30

-20

-10 0 10 Deflection,d (mm)


1/100 1/50

20

30
1/100 1/50

200
Load, P (kN)

RC1
Load, P (kN)

200 100 0 -100 -200 -30

PC1
Load, P (kN)

200 100 0 -100 -200


30

PC4

100 0 -100 -200 -30 -20


Axial load = 550kN (0.25f'cAg) Ordinary RC

Axial load = 550kN (0.25f'cAg) Ordinary ps steel bars

Axial load = 550kN (0.25f'cAg) Graded composite ps steel bars

-10 0 10 Deflection,d (mm)


1/100 1/50

20

30

-20

-10 0 10 Deflection,d (mm)


1/100 1/50

20

-30

-20

-10 0 10 Deflection, d (mm)


1/100 1/50

20

30

150 100
Load, P (kN)

150

150

RC3
Load, P (kN)

100 50 0 -50 -100 -150 -30 30

PC3
Load, P (kN)

100 50 0 -50 -100


30

PC5

50 0 -50 -100 -150 -30 -20


Axial load = -224kN (-0.46fyAs) Ordinary RC

Axial load = -224kN (-0.38fyAs) Ordinary ps steel bars

-10 0 10 Deflection, d (mm)

20

-20

-10 0 10 Deflection, d (mm)

20

-150 -30

Axial load = -224kN (-0.44fyAs) Graded composite ps steel bars

-20

-10 0 10 Deflection,d (mm)

20

30

Fig.4 Measured horizontal load - deection curves

Produced by The Berkeley Electronic Press, 2011

Advanced Materials for Construction of Bridges, Buildings, and Other Structures III [2003], Vol. P05, Article 5

to Baushinger effect may have occurred in the early stage of an unloading path to negative (compressive) direction when the load was reversed in the post-yield range. This sort of stiffness reduction in ordinary strength steel occurs on an unloading path in the negative stress region. In high-strength steel like prestressing steel this may occur in the earlier stage of unloading path even in the positive stress region.
RC1 PC1 PC4 RC2 PC2 RC3 PC3 PC5

Equivalent damping factor

0.3

0.2

0.1

0 0 0.01 0.02 0.03 0.04 Rotation angle (rad)

Fig.5 Equivalent damping factors for Series A test units Effect of Dual-Phase Composite Prestressing Steel Bar
RC2 PC2

PC4 with dual-phase composite prestressing bars 1 P C 1 leastCequivalent damping R C has the P 4 factor in the displacement range of 1/100 to 1/25 among the test units subject to the RC3 PC3 PC compressive axial load. At 1/200 its factor is slightly larger than5 that of PC1. PC5 subjected to tensile axial load indicates pinched hysteresis similar to PC3. Equivalent damping factor of PC5 is as large as that of PC3 and 25% of RC3 at a column rotation angle of 1/25, which were also subjected to tensile axial load. Therefore, dual-phase composite prestressing steel bar was not effective for improving hysteresis energy dissipation. However, hysteresis loops in the less displacement range than 1/100 in a column rotation angle should have been examined in detail during loading. Series B Test Units Six test units were constructed. Three of them were assembled by post-tensioning using grouted prestressing steel bars. The other three were post-tensioned by ungrouted tendons. The test units in Series B are summarized in Table 2. A typical test unit is illustrated in Fig.6. The column section is slightly smaller than that of Series A. This is because of the loading setup used. The concrete compression strength measured was 39.2MPa. The grout which

http://services.bepress.com/eci/advanced_materials/5

Nishiyama and Watanabe: Seismic Performance of Precast Column-Foundation Connection Assem

30 25 75 75 50 50 50 50 50 50 50 50 10 15 600

250 30 45 100 45 30

25
17(SBPR930/1080) D10(SD295)

30

Column section
30 190 30 30

400

D19(SD345) D10(SD295)

Plastic pipe 800

26(SBPR930/1080)

Foundation section

Fig.6 Test unit of Series B was injected into the sheath of the grouted test units had a compressive strength of 49.1MPa. The high-strength mortar used at the interface of the column and the foundation attained a compressive strength of 59.8MPa and 51.0MPa for the grouted units and the ungrouted units, respectively. D10 rebar used as longitudinal and shear reinforcements in the columns had a yield strength of 363MPa. The 17mm diameter prestressing bars (SBPR930/1080) had a yield strength of 1060MPa. Table 2 Test units in Series B Test unit PCa-B1 PCa-B2 PCa-B3 PCa-U1 PCa-U2 PCa-U3 Axial load (kN) 330 (0.153)** 660 (0.306) variable* 330 (0.153) 660 (0.306) variable* Prestressing force (kN) 618.5 (0.287)** 610.5 (0.283) 628.7 (0.292) 603.4 (0.280) 609.7 (0.283) 606.2 (0.281) grouted or ungrouted grouted

ungrouted

*The axial load N varied linearly with the moment M from (M, N)=(-68kNm, 0) to (68kNm, 660kN) **(): Axial load or prestressing force/Agf'c, Ag: gross area of column, f'c: compressive strength of concrete The introduced prestress corresponded to the tendon stress 0.8fy, where fy is the nominal 0.2% offset yield stress of the prestressing steel. W/C ratio of the grout was

Produced by The Berkeley Electronic Press, 2011

30
170 170 30

25

220

110

Advanced Materials for Construction of Bridges, Buildings, and Other Structures III [2003], Vol. P05, Article 5

45%. The testing methods are similar to the ones for Series A. A series of deection controlled cycles comprising two full cycles to each of the column rotation angle of 1/400, 1/200, 1/100, 1/50, 1/33, 1/25, 1/20 and 1/13 was imposed. Load-Displacement Relationships The test units under the lower axial load failed in exural compression. Moment at the column base - column rotation angle relationships are shown in Fig.7. Additional moment due to P- effect is included. Load carrying capacity calculated based on ACI318 is indicated by the horizontal straight line in the gures. For the ungrouted units the tendon stress increament, was estimated by =0.75e +0.25y (e: effective prestress, y: yield strength), which is proposed in AIJ design and construction recommendations for partially prestressed concrete structures for members subject to vertical loading. Prestressing steel bars did not yield for all test units regardless of grouted or ungrouted. Black circles indicate yielding of longitudinal mild steel reinforcement in compression, and black triangles indicate yielding of shear reinforcement. PCa-B1 and PCa-U1 subjected to the lower constant axial load had stable hysteresis loops with a slight reduction in load capacity beyond load cycles to 1/20. However, PCa-B2 and PCa-U2 subjected to the larger constant axial load indicated the large reduction in load capacity after they attained the maximum load at the column rotation angle of 1/100. These units with the larger axial load were loaded up to the column rotation angle of 1/20. Because during cycles to 1/20 the columns became unstable and seemed not to sustain the axial load, the loading was stopped at these cycles. Ultimate Deformation In this study ultimate deformation is dened as the deformation where load carrying capacity reduces to 80% of the maximum load. Table 3 summarizes the ultimate deformations for all test units in both positive and negative loadings. The test units with ungrouted tendons have 11-36% smaller ultimate deformation than the units with grouted tendons. This is because the ungrouted tendons in the compression region of the column cross-section did not work well as compression reinforcement and concrete was subjected to larger compression load than that in the units with grouted tendons. Equivalent viscous damping factor Equivalent viscous damping factors for the test units were calculated based on their load-deformation curves and illustrated in Fig.8. Larger damping factors are obtained from the units subjected to the larger axial load, PCa-B2 and PCa-U2. The other test units show almost the same values. The larger damping factors for PCa-B2

http://services.bepress.com/eci/advanced_materials/5

Nishiyama and Watanabe: Seismic Performance of Precast Column-Foundation Connection Assem

100

longitudinal rebar yielded in compression

100

+Mcal

longitudinal rebar yielded in compression

+Mcal

Moment (kNm)

Moment (kNm)

50

50

PCa-B1
0

PCa-U1
0

-50 -Mcal -100 -8 -6 -4 -2 0

-50 -Mcal
longitudinal rebar yielded in compression

-100

longitudinal rebar yielded in compression

-8

-6

-4

-2

Rotation Angle (%)


100

Rotation Angle (%)


100
longitudinal rebar yielded in compression

longitudinal rebar yielded in compression

Yielding of shear reinf.

Yielding of shear reinf.

Moment (kNm)

+Mcal 50

Moment (kNm)

50

+Mcal

PCa-B2
0

PCa-U2
0

-50 -Mcal -100 -8 -6 -4 -2 0 2 4 6 8


longitudinal rebar yielded in compression

-50
-Mcal longitudinal rebar yielded in compression

-100 -8 -6 -4 -2 0

Rotation Angle (%)

Rotation Angle (%)

100

longitudinal rebar yielded in compression

100
longitudinal rebar yielded in compression Yielding of shear reinf.

Moment (kNm)

50

Moment (kNm)

50

PCa-B3
0

PCa-U3
0

-50

-50

-100 -8 -6 -4 -2 0 2 4 6 8

-100 -8 -6 -4 -2 0

longitudinal rebar yielded in compression

Rotation Angle (%)

Rotation Angle (%)

Fig.7 Moment at the base of the column - column rotation angle relations and PCa-U2 can be attributed to crushing of the compressed concrete. Non-linear elastic hysteresis loops, which are typical for prestressed concrete members, are not observed in these precast post-tensioned columns.

Produced by The Berkeley Electronic Press, 2011

10

Advanced Materials for Construction of Bridges, Buildings, and Other Structures III [2003], Vol. P05, Article 5

0.4

Equivalent damping factor

PCa-B1 (grouted, 0.15f'cAg) PCa-B2 (grouted, 0.31f'cAg) PCa-B3 (grouted, variable axial load) PCa-U1 (ungrouted, 0.15f'cAg) PCa-U2 (ungrouted 0.31f'cAg) PCa-U3 (ungrouted, variable axial load)

0.3

0.2

0.1

0 0 1 2 3 4 5 6

Rotation Angle (%)

Fig.8 Equivalent damping factors for Series B test units Table 3 Ultimate deformation of Series B test units
Test unit Loading Ultimate (%) PCa-B1 PCa-B2 PCa-B3 PCa-U1 PCa-U2 PCa-U3 +ve -ve +ve -ve +ve -ve +ve -ve +ve -ve +ve -ve 4.48 4.66 2.10 2.74 3.36 4.34 3.80 1.84 1.89 2.16 4.57 2.42 3.36 4.07 1.87 2.16 Average (%) direction rotation angle

Axial strain at the centroid Axial strain at the center of the column section - column rotation angle relationships are shown in Fig.9. The axial strain was obtained from the measurements in the column hinge region whose length corresponded to the column hight, 250mm. The moment-column rotation envelope curves are also plotted in the gures. In PCa-B1 subjected to the lower axial load, the axial strain measured was almost in tension. In PCa-U1 under the lower axial load and with ungrouted tendons, the axial compressive strain increased rapidly after the loading cycles to 4%. This corresponds to the ultimate deformation which is dened as the deformation where the load carrying capacity reduced to 80% of the maximum load. For the test units subjected to the larger axial load, PCa-B2 and PCa-U2, the axial compressive strain started to increase at the loading cycles to 1.5-2.0%. Compared with the

http://services.bepress.com/eci/advanced_materials/5

Nishiyama and Watanabe: Seismic Performance of Precast Column-Foundation Connection Assem

11

0.02 0.01 0
Axial strain

120 80 40 0
Moment envelope (+ve) Moment envelope (-ve)

0.02

120

PCa-U1
0.01
Axial strain

80

Moment (kNm)

40

Moment (kNm)

Axial Strain

-0.01 -0.02 -0.03 -0.04

Axial Strain

PCa-B1

-0.01

-40 -80

-0.02

-40

-0.03
Moment envelope (+ve) Moment envelope (-ve)

-80

-8

-6

-4

-2

-120

-0.04 -8 -6 -4 -2 0 2 4 6 8

-120

Rotation (%)
0.02 120
0.02

Rotation (%)
120

PCa-B2
0.01 0
Axial strain

PCa-U2
80 40 0 -40 -80 -120
0.01
Axial strain

80 40 0 -40 -80 -120

Moment (kNm)

Axial Strain

Moment (kNm)

-0.01 -0.02 -0.03 -0.04

Axial Strain

-0.01 -0.02 -0.03 -0.04

Moment envelope (+ve) Moment envelope (-ve)

Moment envelope (+ve) Moment envelope (-ve)

-8

-6

-4

-2

-8

-6

-4

-2

Rotation (%)
0.02 120

Rotation (%)
0.02 0.01 0 120

PCa-B3
0.01 0
Axial strain

80 40 0 -40 -80
Moment envelope (+ve) Moment envelope (-ve)

PCa-U3

Moment envelope (+ve) Moment envelope (-ve)

80 40 0 -40 -80 -120

Moment (kNm)

Moment (kNm)

Axial Strain

-0.01 -0.02 -0.03 -0.04

Axial Strain

-0.01 -0.02 -0.03 -0.04

Axial strain

-8

-6

-4

-2

-120

-8

-6

-4

-2

Rotation (%)

Rotation (%)

Fig.9 Axial strain in columns - column rotation angle relations with moment envelope grouted units, the ungrouted units, PCa-U1, PCa-U2 and PCa-U3 showed large axial compressive strains. This revealed that the ungrouted tendons did not work as compression reinforcement and the compressed concrete had to share larger part of the axial load than the units with the grouted tendons. DESIGN RECOMMENDATIONS The ungrouted test units showed smaller ductility than the grouted units as illustrated in Fig.10 because the ungrouted tendons did not function as compressive

Produced by The Berkeley Electronic Press, 2011

12

Advanced Materials for Construction of Bridges, Buildings, and Other Structures III [2003], Vol. P05, Article 5

reinforcements and the larger part of the compressive force on the column had to be borne by the compressed concrete. Generally, the higher the axial load is, the more signicant the reduction in ductility is. Equivalent damping factors of the test units with larger axial load are larger than those of the units with lower axial load as shown in Figs.5 and 8. This energy dissipation is mainly attributed to concrete crushing in the plastic hinge region. Even the prestressed concrete test units under large axial load showed as large a energy disspation as the ordinary reinforced concrete test units. However, It should be discussed that energy dissipation due to concrete crushing can be justied or not.
5

Maximum rotation angle (%)

PCa-B1

4 PCa-U1 3 2 1 0
grouted ungrouted

PCa-B2

PCa-U2

0.1

0.2

0.3

0.4

Axial load level (N/f'cAg)

Fig.10 Maximum rotation angle - axial load level relations CONCLUSIONS 1. The dual-phase composite prestressing steel bars were not effective for improing hysteresis loops of the precast prestressed concrete column-foundation assemblies. 2. Hysteresis energy dissipation of the prestressed units under lower axial load was smaller than that of the reinforced units. However, under larger axial load the equivalent damping factors increased due to concrete crushing. 3. The ungrouted test units showed smaller ductility than the grouted units because the ungrouted tendons did not work as compressive reinforcements. ACKNOWLEDGMENTS The authors would like to thank Netsuren, Co., Ltd. which provided the prestressing steel bars used in the experiment. REFERENCES 1. M Nishiyama, F Watanabe and H Sato, "Precast Connections Post-tensioned by Graded Composite Steel", Concrete 95 Toward Better Concrete Structures, Conference Papers Vol.1, Brisbane, Australia, 4-7 September 1995, pp.579-588.

http://services.bepress.com/eci/advanced_materials/5