Hybrid System Using Precast Prestressed Frame with Corrugated Steel Panel Damper

Transcription

1 Journal of Advanced Concrete Technology Vol. 7, No. 3, , October 29 / Copyright 29 Japan Concrete Institute 297 Scientific paper Hybrid System Using Precast Prestressed Frame with Corrugated Steel Panel Damper Yukako Ichioka 1, Susumu Kono 2, Minehiro Nishiyama 3 and Fumio Watanabe 4 Received 31 May 29, accepted 27 July 29 Abstract This paper proposes an economical structural system that reduces seismic damage and needs little or no repair by combining precast prestressed concrete elements and corrugated steel panel dampers. Precast prestressed concrete structures show high self-centering characteristics with negligible damage. However, the lateral displacement response during earthquakes tends to be larger than ordinary reinforced concrete (RC) structures because of their lower hysteretic energy dissipation capability. Corrugated steel panels attached to a moment-resisting frame improve its seismic performance with high energy dissipating capability as a hysteretic damper. Five portal frames with corrugated steel panel dampers were tested to investigate the hysteretic characteristics of the proposed hybrid system. Experimental variables were the type of frame structure and the yield strength of the corrugated steel panels. All precast prestressed concrete frames showed a sufficient amount of energy dissipation, and much smaller residual deformations and damage than the monolithic RC frame. Superposition of the simulated hysteretic loops of the frames with that of the damper agreed well with the experimental results obtained by reversed-cyclic loading tests. Using a simple calculation method to estimate the equivalent viscous damping ratios and residual displacements, a design procedure seeking the optimization of the hybrid system is examined. 1. Introduction When major earthquakes have stricken large urban center over the last two decades, many reinforced concrete (RC) buildings have suffered severe damage and have not operated properly after the earthquakes, even if they did not collapse. The long delay for repair has resulted in serious social and economic losses. Although most current seismic design codes require little or no damage during moderate earthquakes and prevent the collapse of structures during major earthquakes, societal demands are shifting toward higher levels of performance. The general public has started to request structures that experience at most minor damage, hence necessitating no repair, and can be used immediately after an earthquake regardless of its intensity. Some methods such as base isolation have been developed as a solution, but their initial and posterior maintenance costs are too high for inclusion in ordinary buildings. Precast prestressed concrete structures need no or little repair after earthquakes because they exhibit nonlinear elastic behavior (Fig. 1(a)) with an effective restoring 1 Research Fellow, General Building Research Corporation of Japan, Japan. ichioka@gbrc.or.jp 2 Associate Professor, Dept. of Architecture and Architectural Engineering, Kyoto University, Japan. 3 Professor, Dept. of Architecture and Architectural Engineering, Kyoto University, Japan. 4 Professor Emeritus, Dept. of Architecture and Architectural Engineering, Kyoto University, Japan. force provided by prestressing tendons. Residual deformations remain very small if rocking at the member interfaces is allowed. However, the small energy dissipating capability and the large degradation of stiffness after gap openings may result in excessive seismic drift demands. In order to reduce seismic drift demands, the authors propose adding energy dissipating elements, which provide fat hysteresis loops as shown in Fig. 1(b). Priestley et al. (1991, 1996 and 1999) proposed some structural systems using precast prestressed concrete members with energy dissipating devices in the PRESSS (PREcast Seismic Structural Systems) research program. The initial concept of the self-centering structural system was attributed to the design of the rail bridge conceived by Cormack (1988). It was demonstrated in the PRESSS research program that those structural systems exhibited so-called flag shape hysteresis loops, as illustrated in Fig. 1(c), and excellent performance with limited or negligible seismic damage. Some of these structural systems were used for 39-story buildings in California as reported by Englekirk (22), and also have been applied to bridge piers (Pampanin et al. 26). The authors developed new energy dissipating devices for precast prestressed concrete frames using corrugated steel panels. When used as shear walls, corrugated steel panels have been shown to have high energy dissipation capabilities even after their peak load (Mo and Perng 2 and Chosa et al. 26). This paper introduces a hybrid structural system using precast prestressed frames with corrugated steel panel dampers. Cyclic loading tests on five portal frames with dampers were conducted. The corrugated steel panels

2 298 Y. Ichioka, S. Kono, M. Nishiyama and F. Watanabe / Journal of Advanced Concrete Technology Vol. 7, No. 3, , 29 P P P δ δ δ Flag Shape Hysteresis Loop (a) Nonlinear elastic behavior (b) Plastic behavior (c) Flag shape behavior Fig. 1 Hysteresis loops of precast prestressed concrete structures and energy dissipating elements. Concrete slab Inner cables Corrugated steel plate web Deviator Corrugated steel plate web External cables (a) As webs of a bridge Concrete slab (b) As a shear wall in a school building Fig. 2 Practical applications of corrugated steel shear panels. yielded at the expected drift angle and dissipated a sufficient amount of hysteretic energy. Good self-centering behavior was observed when the lateral load contribution of the damper was adequate. A calculation method to determine the adequate contribution of the damper was derived from a simple M-θ model. 2. Corrugated steel shear panel damper Corrugated steel shear panels are mainly used as webs of box girder bridges as shown in Fig. 2(a) because they weigh less and decrease prestressing loss due to their negligible axial stiffness. Mo and Perng (2) suggested the use of corrugated steel shear panels instead of reinforced concrete shear walls as the main lateral load resisting components in building structures. Although their experimental results showed poor seismic performance because of insufficient connection rigidity between the shear panel and the peripheral frame, Chosa et al. (26) confirmed that the shear capacity and stiffness of corrugated shear panels can be fully utilized with anchorage satisfying the Japanese design guideline, Design Guidelines for Composite Structures (1985). Corrugated steel shear panels have already been used as shear walls in Japan (Fig. 2(b)). 3. Experiments Experimental studies were conducted at Kyoto University to investigate the seismic behavior of precast prestressed frames built incorporating corrugated steel panel dampers. It was observed that corrugated steel panel dampers yielded at the expected drift angle and improved energy dissipating capabilities with small residual deformations. Ductility of frames and dampers and damage conditions were also checked. 3.1 Specimens Specimens consisted of portal frames and corrugated steel panel dampers. Experimental variables were the yield strength of corrugated steel panel dampers (3 MPa, 225 MPa and 1 MPa) and types of frame structures (precast post-tensioned frames with or without grouting and a monolithic conventional reinforced concrete frame). The specimens are summarized in Table 1. Prestressing forces, F i, corresponding to 85% of the yield strength of tendons were initially introduced in the four precast post-tensioned frames. Constant axial load of 9 kn was applied to each column. Also listed in Table 1 are initial axial force ratios, P i /A g f c (P i : initial axial load including initial prestressing force and constant axial load, A g : gross cross-sectional area, f c : measured concrete compressive strength), for beams and columns. 3.2 Corrugated steel panel dampers The configurations of the dampers of the five specimens were identical, as shown in Fig. 3. The flat steel plates at the top and bottom of the corrugated plate were thick enough to concentrate shear deformation in the corrugated plate. Triangular panels extruding from both sides of the flat plates were used to prevent flexural deformation of the flat plate. The height of the corrugated steel

3 Y. Ichioka, S. Kono, M. Nishiyama and F. Watanabe / Journal of Advanced Concrete Technology Vol. 7, No. 3, , Table 1 Main characteristics of test specimens. Specimen Frame Tendon F i (kn) P i /A g f c Damper Column.26 Extremely low PCbS yield strength Beam.8 with steel (LY1*) grouting Column.26 Low strength Post- Indented PCbM 277 steel tensioned 9. mm* Beam.8 (LY225*) precast Column.26 Extremely low members PCuS yield strength without Beam.8 steel (LY1*) grouting Indented Column.3 PCuL 434 Mild strength 11.2 mm* Beam.12 steel Reinforced concrete Column.2 RC - - (SS4*) (cast monolithically) Beam F i : Initial prestressing force per a tendon, Axial force ratio: Axial stress to concrete compressive strength, *LY1: f y =15 MPa, LY225: f y =235 MPa, SS4: f y =37 MPa, Indented tendon 9. mm* and 11.2 mm*: Diameter of tendon is 9. mm and 11.2 mm Fig. 3 Configuration and dimensions of corrugated steel panel damper. panel, h D (= 19 mm), was the minimum height to accommodate one corrugation to ensure the corrugated steel panel yielded at a small drift. The peripheral flange plates at the sides of the corrugated plate were so strong and stiff that they remained elastic when the corrugated panel carried 1.5 times the shear force at shear yielding. The flange plates were welded to the corrugated steel panel, flat plates or triangular plates with full penetration welds. The corrugated steel panel and the flat plates were connected by high-strength bolts. The damper and the portal frame were connected through mortar by high-strength bolts. The mechanical properties of the plates are listed in Table 2. The expected drift angle and lateral load at shear yielding of the shear panel are summarized in Table 3. The expected lateral load at damper yielding was computed by multiplying sectional area of the corrugated steel panel by shear yield strength (= f y / 3, f y : yield strength of the corrugated steel panel). The expected drift angle was calculated on the assumption that the lower half part of the damper was an elastic cantilever beam with H-section. 3.3 Frames Frames were designed at 4% scale and modeled as an internal span of the first story of a mid or low-rise building. The dimensions, reinforcing arrangement and cross-section configurations are illustrated in Figs 4 and 5. In the precast specimens (PCbS, PCbM, PCuS and PCuL), columns, beams and stubs were cast separately

4 3 Y. Ichioka, S. Kono, M. Nishiyama and F. Watanabe / Journal of Advanced Concrete Technology Vol. 7, No. 3, , 29 Table 2 Mechanical properties of steel plates. Specimen Plate type Yield Tensile Thickness strength strength (mm) (MPa) (MPa) PCbS, PCuS LY * 247* Corrugated PCbM LY * 324* steel panel PCuL, RC SS * 4* Side flange plate Flat plate, All specimens Top and bottom SS4 flange plate, Triangular plate * In direction parallel to corrugations Table 3 Expected drift angle and lateral load carried by damper at shear yielding. Damper Specimen Expected drift angle at Expected lateral load at damper yielding (%) damper yielding (kn) LY1 PCbS, PCuS LY225 PCbM SS4 PCuL, RC D16 Mortar joint 2mm 2-D16 D1@ 8 Corrugated steel panel damper 2-D16 Anchor plate 3mm D1@ 7 4-D22 D13@ 1 5-D22 5-D (a) PCbS, PCbM, PCuS and PCuL D19 D1@ 7 2-D22 D1@ 8 Corrugated steel panel damper 2-D22 4-D22 D13@ 1 6-D (b) RC Fig. 4 Arrangement of reinforcements and dimensions (unit: mm). 143 D Cover D D D1 D D1 D Column Beam (a) PCbS, PCbM, PCuS and PCuL Column (b) RC Beam Fig. 5 Cross-section dimensions of columns and beams (unit: mm). and connected through mortar joints by post-tensioning. Specimen RC was cast monolithically. The material properties of the reinforcement, concrete and mortar are listed in Tables 4 to Loading arrangement The loading system is shown in Fig. 6. Axial load of 9 kn was applied to each column (the axial load ratio was.2) and kept constant during testing. Equal magnitude of lateral load was applied to both ends of the beam by two 1 kn hydraulic jacks. The first cycle of loading was applied until cracking was observed in either beam or columns. This was followed by a series of loading cycles, which consisted of two full cycles to a story drift

5 Y. Ichioka, S. Kono, M. Nishiyama and F. Watanabe / Journal of Advanced Concrete Technology Vol. 7, No. 3, , angle, R, of.1%,.2%,.4%,.6%,.8%, 1.%, 2.%, 3.%, and 4.%. Story drift angle was defined as the average of lateral displacements of both column tops at the central height of the beam divided by story height (= 1 mm). Additional cycles to a story drift angle larger than 4% were applied to some of the specimens. 4. Experimental results 4.1 Hysteresis restoring force characteristics Figure 7 shows the lateral load carried by the frame and damper-story drift relations up to R = 4%. The lateral load and drift angle at yielding of the corrugated panel, and the peak load are summarized in Table 7. The yield Table 4 Measured mechanical properties of reinforcement. Bar type Area Yield strength Tensile strength Young s Modulus (mm 2 ) (MPa) (MPa) (GPa) D D D D D Indented tendon 9. mm * 15* 2* Indented tendon 11.2 mm 1 141* 15* 2* * Data in the Inspection Certificate Table 5 Measured mechanical properties of concrete. Specimen Member Compressive strength Tensile strength Young s Modulus (MPa) (MPa) (GPa) PCbS Column, Beam Stub PCbM Column, Beam Stub PCuS Column, Beam Stub PCuL Column, Beam Stub RC All members Table 6 Mechanical properties of joint mortar*. Specimen Compressive strength Tensile strength Young s Modulus (MPa) (MPa) (GPa) Damper-Frame Beam-Column PCbS Column-Footing Grouting in Beam Grouting in Column Damper-Frame Beam-Column PCbM Column-Footing Grouting in Beam Grouting in Column Damper-Frame PCuS Beam-Column Column-Footing Damper-Frame PCuL Beam-Column Column-Footing RC Damper-Frame * φ 5 mm x 1 mm cylinders were used

6 32 Y. Ichioka, S. Kono, M. Nishiyama and F. Watanabe / Journal of Advanced Concrete Technology Vol. 7, No. 3, , 29 1kN hydraulic jack 12kN hydraulic jack 1kN hydraulic jack Reaction wall 5kN load cell 5kN load cell Reaction floor Reaction block Fig. 6 Loading setup (unit: mm) Lateral Load (kn) Total Damper Damper yielding Lateral Load (kn) Total Damper Damper yielding Total Damper Damper yielding (a) PCbS (b) PCbM (c) PCuS Total Damper Damper yielding (d) PCuL Total Damper Damper yielding (e) RC Fig. 7 Lateral load-story drift angle relations. Table 7 Lateral load and drift angle at damper yielding and peak load. At damper yielding At peak load Specimen Drift angle (%) Lateral load (kn) Drift angle (%) Lateral load (kn) Positive Negative Positive Negative Positive Negative Positive Negative PCbS.37 (-.9) 19. (-186.1) PCbM.42 (-.41) (-255.5) PCuS (-.2).38 (111.6) PCuL.97 (-.82) (-322.4) RC.86 (-.73) (-272.7) points were determined by strain gauge readings with the Von Mises yield criterion. Shown in parentheses are the second yield points, which may not be justified because of residual strains after the first yielding. Each specimen dissipated a large amount of hysteretic energy. Residual displacements were relatively large in RC and PCuL with the mild strength steel corrugated panel, while the other three specimens with low strength steel showed good

7 Y. Ichioka, S. Kono, M. Nishiyama and F. Watanabe / Journal of Advanced Concrete Technology Vol. 7, No. 3, , self-centering performance. Degradation of lateral load carrying capacity after peak load was small, with 8% of peak load still supported beyond R = 4.% for all specimens. Specimen PCuL experienced abrupt lateral load decrease at R = 2 to 3% because of the slip at the gap between the beam and a column, but this had little or no effect on the lateral load carrying capacity. 4.2 Lateral load carried by dampers The lateral load carried by dampers-story drift angle relations are shown in Fig. 7. Damper yield points are also plotted. Shear stress induced in the corrugated steel panel was computed from strain readings, with plane stress assumption, and the Von Mises yield criterion. Strain gauges were attached to the flat plates, which showed elastic strain until R = 4%. The shear resistance of the damper was calculated by multiplying the shear stress by the cross-sectional area of the flat plate. The ratio of load carried by the damper to the total lateral load was 3% to 35% at R = 4% in the five specimens. The computed contribution was 25% by considering the story shear force at the formation of a collapse mechanism for the bare frame and the shear force of the damper at yielding. The dampers carried a larger lateral load than expected. The lateral load carried by the damper increased until R = 1.% after damper yielding. The increment was largest in PCbS and PCuS, which used extremely low strength steel (LY1). 4.3 Equivalent viscous damping ratio and residual deformation ratio Figure 8 shows the equivalent viscous damping ratio-story drift angle relations. The equivalent viscous damping ratio, h eq, was calculated from each second hysteresis cycle of the lateral load-drift relations. h eq of the bare frames without dampers were computed from hysteresis loops of the lateral load carried by the frame-story drift relations. Lateral load carried by the frame was obtained by subtracting lateral load carried by the damper from the total lateral load. h eq of the precast post-tensioned specimens was about 4% larger than that of the bare frames, while the increment of h eq in specimen RC was about 2%. The difference in yield strengths of the corrugated steel panels (LY225 and LY1) made only a small difference in h eq as observed in PCbM and PCbS. Residual deformation ratio, r d, which is defined as the ratio of the average residual deformation in the positive and negative directions to the maximum deformations in each cycle, was recorded to evaluate self-centering performance. Figure 9 shows the residual deformation ratio-story drift angle relations. The ratio was obtained from the second loop of each loading cycle. r d of specimen RC increased with story drift and reached 45% at R = 2.%, while r d of the precast prestressed specimens were about 18 to 25% at the same story drift angle. The columns and beams of the precast post-tensioned specimens suffered little damage. Less concrete crushing Equivalent Viscous Damping Ratio, h eq (%) Equivalent Viscous Damping Ratio, heq (%) PCbS PCbM PCuS 5 PCuL RC (a) Total PCbS PCbM PCuS PCuL RC (b) Bare frame Fig. 8 Equivalent viscous damping ratio. Residual Deformation Ratio, r d (%) PCbS PCbM PCuS PCuL RC Fig. 9 Residual deformation ratio. was observed in these specimens than in specimen RC and cracks were concentrated in the joint mortar. Almost the same amount of r d in PCbS and PCuL indicated that r d was independent of bond due to grouting. r d of PCuS was the smallest of all specimens since the small shear force carried by the low strength corrugated panel (LY1) can be easily offset by the restoring force of the precast post-tensioned concrete frame. 4.4 Damage of frames and dampers Damage of specimens PCuS and RC after experiencing R = 4.% drift are shown in Fig. 1. Local and overall buckling occurred at the corrugated steel panel, but no fracture was observed at the plate and welded part. PCuS showed minor damage with small crushing areas at the

8 34 Y. Ichioka, S. Kono, M. Nishiyama and F. Watanabe / Journal of Advanced Concrete Technology Vol. 7, No. 3, , 29 column bases and the ends of the beam. However, damage of the RC specimen was severe at the hinge zones near both ends of the beam, with concrete spalling and bucking of reinforcement. 5. Design procedure of the hybrid system 5.1 Numerical simulation using M-θ Model A simple mechanical model was developed to simulate the hysteretic behavior of precast prestressed concrete frames with corrugated steel panel dampers. The model is expressed by superposition of two independent and additive shear resistances attributed to the frame and the damper. Chosa et al. (26) proved that the hysteretic characteristics of reinforced concrete portal frames with corrugated steel shear walls were well simulated by superposition of hysteresis loops of the frames and the shear walls. The hysteresis loops of the damper are expressed by the Menegotto-Pinto model (Restrepo 1993) shown in Eq. 1. Based on the hysteresis loops of R =.5% and R = 1.% cycles, the strain-hardening ratio, Q, and the control parameter for transition from elastic to plastic branches, R, were defined as Q =.5 and R = 13.5, respectively. This model considers hardening due to cyclic loading. The hysteresis loops of the frame were computed based on the M-θ model proposed by Ichioka et al. (28). This M-θ model was developed for a precast prestressed concrete member taking into consideration the bond-slip behavior of tendons caused by a large gap opening at the interface between precast members. The bond-slip relationship for strand tendons was based on research by Adachi et al. (2). The analytical results of PCbM in Fig. 11 gave a good estimation of measured lateral load resistance and residual deformations. f s = f + (ε s ε ) E m [Q + (1 Q) / {1 + E m (ε s ε ) / (f ch f ) R } 1 / R ] (1) where ε s and f s : current strain and stress, ε and f : strain and stress at reversal point, E m : initial elastic tangent, Q: strain-hardening ratio (ratio between post-yield tangent and initial elastic tangent), f ch : yield strength, R: control parameter for transition from elastic to plastic branches. 5.2 Procedure to determine adequate damper contribution Optimization of energy dissipating performance and self-centering performance of the hybrid system proposed in this paper is simple since the frames and dampers work independently for a given story drift. The hysteretic characteristics of the system are estimated by superposition of the hysteresis loops of the dampers and frames, as mentioned in the preceding section. Figures 12(a) and (b) show the equivalent viscous damping ratio, h eq, and residual deformation ratio, r d, calculated by the superposition method where the initial stiffness of the damper ranges from.25 to 5 times that of specimen PCbM. In Figures 12(a) and (b), β is the lateral load contribution of the dampers to the whole structural system (frames and dampers). In PCbM, the initial shear stiffness of the damper, G D, was 5 GPa and β was.33. (a) PCbS Fig. 1 Specimen photos after R = 4.% cycle. (b) RC 1 Lateral Load (kn) Experiment Analysis Experiment Analysis Experiment Analysis Rotation Angle (%) (a) Frame + Damper (b) Frame (c) Damper Fig. 11 Comparison between experiments and analyses for PCbM.

9 Y. Ichioka, S. Kono, M. Nishiyama and F. Watanabe / Journal of Advanced Concrete Technology Vol. 7, No. 3, , Equivalent Viscous Damping Ratio, h eq (%) GD= GPa GD=5 GPa GD=25 GPa GD=17 GPa GD=5 GPa Lateral Load Contribution of Damper, β (a) h eq -β relations Residual Deformation Ratio, r d (%) GD= GPa GD=5 GPa GD=25 GPa GD=17 GPa GD=5 GPa Lateral Load Contribution of Damper, β (b) r d - β relations Fig. 12 Optimization of amount of corrugated shear panel dampers. To maximize the energy dissipating capability evaluated by h eq, the lateral load contribution of a damper is defined from Fig. 12(a), however, r d in Fig. 12(b) sets a limitation on the value of β that can be selected. For example, if r d < 1% and h eq > 15% is desirable, β >.35 is needed from Fig. 12(a) for a G D = 5 GPa damper. However, in Fig. 12(b), r d = 15% for β =.35. In the case of using G D = GPa damper, β =.3 meets all requirements. 6. Conclusions Static reversed-cyclic loading tests on five portal frames with corrugated steel panel dampers were conducted. The following conclusions were obtained. (1) The combination of corrugated steel shear panels with precast prestressed concrete frames resulted in a high restoring force and large hysteresis loops. The performance of specimens with bonded tendons was as good as that with unbonded tendons. All specimens showed ductile behavior and reduction of the load carrying capacity was less than 2% of the peak load even at R = 4.%. (2) Performance of the extremely low yield strength (LY1) steel panel damper was as good as that of the low strength (LY225) steel panel damper. The energy dissipation capabilities of the two steel panels were almost the same. However, a smaller residual displacement was observed in the specimen with the LY1 steel damper since the smaller shear force carried by the damper can be easily offset by the restoring force of the precast prestressed frame. (3) The precast prestressed columns and beams suffered little damage. Less concrete crushing, compared with specimen RC, and less crack concentration in the joint mortar were observed in these specimens. Almost the same amount of r d in PCbS and PCuL indicated that r d was independent of bond due to grouting. (4) The hysteretic behavior of the hybrid frame could be accurately predicted by superposing the hysteretic behaviors of the precast prestressed frame and the dampers. The method accurately predicted the energy dissipation and the residual displacement. Acknowledgments Part of this research was financially supported by two Grant-in-aids of the Ministry of Education, Culture, Sports, Science and Technology (PIs: H. Tanaka and S. Kono), Development of Innovative Seeds, Japan Science and Technology Agency (PI: S. Kono), and Collaborative Research Projects of the Materials and Structures Laboratory, Tokyo Institute of Technology (Prof. S. Hayashi). References Adachi, M., Takatsu, H. and Nishiyama, M. (2). Idealization of bond characteristic of prestressing strand. Summaries of technical papers of Annual Meeting Architectural Institute of Japan, C-2 Structures IV, (in Japanese) Architectural Institute of Japan (1985). Design Guidelines for Composite Structures. (in Japanese) Chosa, K., Kashiwai, Y., Kono, S. and Watanabe, F. (26). Fundamental study on corrugated steel webs used as shear walls. Summaries of Technical Papers of Annual Meeting Architectural Institute of Japan, Vol. C2, (in Japanese) Cormack, L. G. (1988). Design and construction of the major bridges on the mangaweka rail deviation. Transactions of the Institution of Professional Engineers New Zealand, Civil Engineering Section, 15(1), Ichioka, Y., Kono, S. and Watanabe, F. (28). Structural system enabling prompt recovery after earthquakes. Proceedings of the 14 th World Conference on Earthquake Engineering, ID Englekirk, R. E. (22). Design construction of the paramount A 39-story precast prestressed concrete apartment building. PCI Journal, 47(4), Mo, Y. L. and Perng, S. F. (2). Hybrid RC

10 36 Y. Ichioka, S. Kono, M. Nishiyama and F. Watanabe / Journal of Advanced Concrete Technology Vol. 7, No. 3, , 29 frame-steel wall systems. Composite and Hybrid Systems, ACI SP-196, Pampanin, S., Amaris, A. and Palermo, A. (26). Implementation and testing of advanced solutions for jointed ductile seismic resisting frames. Proceedings of the 2nd International Congress, ID 8-2. Priestley, M. J. N. (1991). Overview of the PRESSS research programme. PCI Journal, 36(4), Priestley, M. J. N. (1996). The PRESSS program current status and proposed plans for phase III. PCI Journal, 41(2), Priestley, M. J. N., Sritharan, S., Conley, J. R. and Pampanin, S. (1999). Preliminary results and conclusions from the PRESSS five-storey precast concrete test building. PCI Journal, 44(6), Restrepo, J. I. (1993). Seismic behaviour of connections between precast concrete elements. Research report, 93-3, University of Canterbury.