420 G. Campione et al. Table 3 gives the maximum bearing capacity evaluated with EC4, LRFD and the model proposed. For the latter the strength of the columns is penalised to take into account the effect due to the slenderness of the columns through the ~ coefficient proposed by EC4. In Table 3, ~, and ~ are the slenderness as evaluated according to EC4 and LRFD, ~ NplR and P= the bearing capacity of the composite columns according to EC4 and LRFD respectively and ~ Ar the bearing capacity evaluated according to the method proposed penalising the axial plastic force with the ~ coefficient, and finally N~, the experimental values. CONCLUSIONS The strength of circular columns filled with plain concrete and FRC depends on the mechanical and geometrical properties of the steel columns (type of steel, thickness, diameter) and also on the strength of the concrete and type and volume percentage of fibres. Experimental data from tests on materials (compression and tension tests) and from compressive tests on steel columns filled with plain concrete and FRC have shown the advantages of using FRC rather than plain concrete. Simplified analytical model to predict the load-deformation curves in compression of composite members is proposed. It is based on monoaxial state of stresses for concrete, in which the confinement effect due to the lateral pressure f'l of the steel pipes, is variable at each loading step and steel pipe is in biaxial state of stresses. From the comparison between the experimental and analytical results obtained using the simplified method good agreement is found in term of maximum strength and maximum displacements, but different values are recorded in the softening branch, because of the instability phenomena that the model do not take into account. REFERENCES Campione G. Scibilia N. & Zingone G. (1998) Comportamento ciclico in compressione di colonne composte in acciaio e calcestruzzo fibrorinforzato. 3 ~ Workshop Italiano sulle Strutture Composte, Ancona, 29-30 Ottobre. Campione G, Mindess S, Scibilia N., &.Zingone G, (1999). Compressive behaviour of circular steel columns filled with fibre reinforced concrete: experimental investigation and comparison with EC4 code. Costruzioni Metalliche 5. Cosenza E., Pecce M. (1993). Resistenza a compressione di colonne con sezione di acciaio riempita di calcestruzzo. Atti delle Giornate Italiane delle Costruzioni in Acciaio, Viareggio 24-27 ottobre, 290- 303. Eurocode 4 - Common Unified Rules for Composite Steel and Concrete Structures, European Committee for Standardization (CEN), ENV 1994-1-1. Frassen J.M., Talamona D., Kruppa J., Cajot L.G. (1998). Stability of steel columns in case of fire: experimental evaluation. Journal of Structural Engineering ASCE, 124:2, 158-163. Lie T.T. (1994). Fire resistance of Circular Steel Columns Filled with Bar-Reinforced Concrete. Journal of Structural Engineering ASCE 120: 5, 1489-1509. Manual of Steel Construction:Load and resistance factor design (LRFD) (1994) 2nd Ed., Am. Inst. of Steel Construction, Chicago, IL 1. Mander J. B., Priestley, M.J.N. & Park, R. (1988). Theoretical stress-strain model for confined concrete, Journal of Structural Eng#leering ASCE 114:8, 1804-1825. Pecce, M.(1993). La modellazione del comportamento in condizioni ultime di colonne composte con sezione circolare in acciaio riempita di calcestruzzo e sottoposte a sforzo normale centrato. 1 ~ Workshop Italiano sulle Strutture Composte, Trento, 17-18 maggio. Schneider S.P. (1998). Axially loaded concrete-filled steel tubes. Journal of Structural Engineering, ASCE, 124: 10, 1125-1138. Shakir-Khail H. (1994). Experimental study of concrete-filled rectangular hollow section columns. Structural Engineering Rewiew, 6: 2,. 85-96. AXIAL COMPRESSIVE STRENGTH OF STEEL AND COMPOSITE COLUMNS FABRICATED WITH HIGH STENGTH STEEL PLATE B. Uy 1 ISchool of Civil & Environmental Engineering, The University of New South Wales, Sydney, NSW, 2052, AUSTRALIA ABSTRACT The design of tall buildings has recently provided many challenges to structural engineers. One such challenge is to minimise the cross-sectional dimensions of columns to ensure greater floor space in a building is attainable. This has both an economic and aesthetics benefit in buildings, which require structural engineering solutions. The use of high strength steel in tall buildings has the ability to achieve these benefits as the material provides a higher strength to cross-section ratio. However as the strength of the steel is increased the buckling characteristics become more dominant with slenderness limits for both local and global buckling becoming more significant. To arrest the problems associated with buckling of high strength steel, concrete filling and encasement can be utilised as it has the affect of changing the buckling mode which increases the strength and stiffness of the member. This paper describes an experimental program undertaken for both encased and concrete filled composite columns, which were designed to be stocky in nature and thus fail by strength alone. The columns were designed to consider the strength in axial compression and were fabricated from high strength steel plate. In addition to the encased and concrete filled columns, unencased columns and hollow columns were also fabricated and tested to act as calibration specimens. A model for the axial strength was suggested and this is shown to compare well with the test results. Finally aspects of further research are addressed in this paper. KEYWORDS Columns, composite construction, high strength steel, steel structures, tall buildings INTRODUCTION The design of tall building gravity load systems is influenced heavily by the ability to resist axial force with the smallest cross-sectional sizes available. Recent developments in the quality of high strength steel have seen it become extremely attractive for the design of tall buildings in Australia and future landmark buildings are earmarked to utilise high strength steel in Japan. The benefits of the use of 421 422 B. Uy high strength steel can be utilised in a braced frame where the external spandrel frame is used to resist gravity loads alone. High strength steel is most efficient when it is allowed to develop its' full yield stress. Thus high strength steel is efficient when local and overall buckling can be eliminated in a column design. This paper will summarise the previous applications of high strength steel in tall buildings in major cities throughout the world. The summary includes a description of the column type and grade of steel used to illustrate the methods in which high strength steel is being used. Based on the previous applications, a detailed experimental program was conducted which reflected current and future uses of high strength steel in composite columns. The experiments were based on the behaviour of high strength steel columns in pure compression and consisted of both bare steel sections and composite sections. These experiments will be described here and a numerical model for the calculation of the pure compressive strength will be presented and shown to provide a conservative estimate of the column cross-section strength. COMPLETED AND PLANNED PROJECTS Previous projects, which have been completed and planned, are summarised herein. This list will identify the type of projects and the potential benefits achieved from the use of high strength steel. In particular, this table reflects tall building projects in Australia where high strength steel has been used. In the design of Star City, the major benefits derived from the use of high strength steel were in providing additional car space in the basement levels. This was a mandatory requirement for the project by the Sydney City Council. The use of high strength steel in the other Australian buildings was justified in reducing column sizes and thus providing additional floor area and car park spaces in the building. The Shimizu Super High Rise (SHR), which is a proposed project in Tokyo, Japan, will use high strength steel in box columns for the exterior spandrel frame. Figure 1 also illustrates the cross-section geometries utilised for each of these projects. TABLE 1 PROJECTS UTILISING HIGH STRENGTH STEEL Building Grosvenor Place City Sydney Perth Year Number of Completed Storeys 1988 50 Central Park 1989 50 300 Latrobe St. Melbourne 1990 30 20 Star City Shimizu SHR Sydney 1997 Tokyo Proposed 120 Column Type Encased Encased Encased Encased Filled Steel Grade (MPa) 690 690 690 690 600 Figure 1: High strength steel composite cross-sections Axial Compressive Strength of Steel and Composite Columns PREVIOUS RESEARCH 423 Previous research into high strength structural steel for columns has been mainly carried out in the regions where it has been applied in practice and this includes research in both Australia and Japan. Firstly Rosier and Croll (1987) considered the benefits of high strength quenched and tempered steel being applied in structures such as bridges, buildings and silos. This study included consideration of the economics of the material over conventional mild structural steel and showed the significant advantages that could be derived from its use. Rasmussen and Hancock (1992 and 1995) conducted tests on both high strength steel fabricated I- sections and box sections. These tests established local buckling slenderness limits for these high strength steel sections. Furthermore, slender columns were tested and the behaviour of these was compared with the slender column curves of the existing Australian Standard AS 4100-1990 (Standards Australia 1990). It was found that providing the local buckling slenderness limits were adhered to, then the slender column behaviour could be described using this standard developed specifically for mild structural steel. Hagiwara et al. (1995) and Mochizuki et al. (1995) considered the behaviour of high strength structural steel for the application in super high rise buildings in Japan. These studies considered the reliability inspection and the welding process for heavy gauge steel plate. These studies are pertinent to the application of the use of high strength steel in projects such as the Shimizu Super High Rise in Tokyo, Japan. Uy (1996) considered the behaviour of concrete filled steel box columns filled with concrete. These studies considered the advantages derived from filling the sections with concrete to increase the local buckling stresses. Furthermore, the members were considered under combined bending and compression to assess the strength of short columns. The results of these columns were compared with columns designed with normal strength structural steel, to show the reduced cross-sectional dimensions able to be achieved. Furthermore, comparisons of the cross-sectional ductility were made and showed that composite members composed of high strength structural steel still had a large degree of reserve of strength after peak loading conditions. EXPERIMENTS This section outlines the test program undertaken which included column tests and numerous material property tests. The test set-up for the columns will be described and the results will then be presented. A general review and description of the failure modes will then be provided. Column Tests The test program consisted of eight columns, of which four columns were fabricated I sections and four were fabricated box sections. The columns and their pertinent dimensions are shown in Figure 2 Figure 2: Column cross-sections 424 B. Uy Tensile Coupon Tests To determine the stress-strain characteristics of the steel plate in tension a series of tensile coupons were produced from the virgin steel plate and tested in an Instron uniaxial testing machine. Pertinent data for these test coupons are provided in Table 2. Four tests were conducted with a mean value for yield stress of 784 MPa being established. Whilst high strength steel is not considered to have a defined strain hardening region, the tests revealed an increase in stress after yielding and the mean ultimate stress of the material in tension was determined to be 817 MPa. TABLE 2 TENSILE COUPON TESTS Specimen Number Yield Stress, cry (MPa) 765.3 Ultimate Stress Cru (MPa) 809.7 2 781.3 816.8 3 796.4 808.9 4 793.7 830.9 Mean 784.2 816.6 Standard Deviation 14.2 10.2 Compressive Stub Column Tests The stress-strain characteristics can vary in tension and compression, which can be due to the effects of inelastic local buckling and/or residual stresses. To try and identify these differences a series of stub column tests were undertaken. These stub column tests were able to establish a clear reduction in the mean yield stress of at least 30 MPa, which could be attributed to the effects of inelastic local buckling. However, the ultimate stress of the stub column tests was virtually identical to the tensile coupon tests with a mean value of 817 MPa being attained in both. TABLE 3 STUB COLUMN TESTS Specimen Number Average Standard Deviation Yield Stress, oy (MPa) 757.3 757.3 750.0 754.9 4.2 Ultimate Stress Cru (MPa) 833.0 839.1 779.0 817.0 33.1 Column Test Set-Up Eight columns were tested in pure compression in an Amsler 5,000 kN capacity compression testing facility. The column test set-up is illustrated in Figure 3 which shows the general characteristics of the testing machine platens and the instrumentation used in the testing. The test set-up highlights the end conditions, which were provided to ensure a uniform loading surface to the column. Using steel plates with recessed edge, the plates were filled with a very stiff plaster and a small preload was applied until the plaster cured and reached an appropriate stiffness. In addition to this strain gauges were used on Axial Compressive Strength of Steel and Composite Columns 425 the steel surfaces, which was useful in tracing the load-strain characteristics for both the determination of yielding and local buckling of the steel plates. Furthermore, linear varying displacement transducers (LVDT's) were used to measure the load-axial shortening characteristics which was also useful in the determination of yielding and ultimate loading of the column members. Figure 3: Photograph of column test set-up Load-Deflection Results The axial load - axial shortening of each column was recorded and these were useful in being able to ascertain the point at which yielding took place and the point of ultimate failure, which was usually characterised by concrete crushing and softening. Figure 4 illustrates typical load-deflection results for the columns tested. For both the fabricated I section and box section columns, one can see that the composite sections have a larger stiffness, as well as achieving a larger ultimate capacity. Furthermore, the apparent ductility of both the bare still sections and the composite sections is shown to be quite adequate with a significant post peak reserve of strength being displayed. Figure 4: Load-deflection results 426 B. Uy Load-Strain Results The load strain results were used to identify yielding of the steel sections in compression, and the strain gauges also proved useful in identifying the onset of inelastic local buckling. Figure 5 shows a set of typical load-strain results for the columns tested. The fabricated I section columns were designed so that the plates were compact for local buckling, however inelastic local buckling was still evident after the ultimate load was reached and this is defined by the erratic behaviour of the strain gauges as shown below. The fabricated box columns were also designed with compact plates, however, inelastic local buckling was more controlled and greater stress redistribution capability was noted by the smooth nature of the strain gauges after the peak load was reached as shown below. Figure 5: Load-strain results Failure Modes All columns were tested in pure compression and thus failure was essentially primary compression. Failure was initiated by compressive yield since most plate sections were compact. Once yielding began, concrete crushing and inelastic local plate buckling of the plate elements usually followed this. Whilst the failure mode was primary compressive each of the members displayed a significant reserve of strength and thus highlighted a ductile failure plateau. Figure 6 shows the failure modes for six of the columns tested in this paper. This photograph highlights the local buckling modes for both the bare steel and composite sections as well as highlighting the concrete crushing for the composite sections. Figure 6: Failure modes of columns Axial Compressive Strength of Steel and Composite Columns COMPRESSIVE STRENGTH MODEL 427 All the columns tested in this paper were tested in pure compression and in order to predict the strength of these members an axial compressive strength model is proposed which considers both the steel and concrete contributions to axial strength. The model shown in Equation 1 was used to provide a prediction of the axial strength of each of the columns N u = Nuc + Nus (1) where Nu is the ultimate axial strength of the composite column, Nuc=fc.Ac is the concrete contribution to axial strength and Nus=fy.As is the steel contribution to axial strength. COMPARISONS Table 4 summarises the specimen names, pertinent geometric and material properties and results of the tests. The strength of each of the tests, Nu.test was determined from the peaks of the load-deflection graphs, whilst the theoretical value for ultimate load Nu.theory was determined using Equation 1. The ratio of the test results to theoretical results is also calculated in Table 4 for each specimen. The theoretical results generally provide a conservative estimate of strength for the steel section columns. Furthermore, the composite section columns are also well represented by the model. There is a slight value of non-conservatism in the strength determination of the composite columns, which could be due to the assumed maximum concrete stress. Eurocode 4 (British Standards Institution 1994) suggests a maximum stress equivalent to the cylinder stress be used and this generally accounts for some confinement effect. The maximum stress utilised herein was equal to the cylinder strength and thus illustrates that some confinement may be present. However, it may be necessary to impose a factor to account for creep in the composite columns and thus a value less than the cylinder strength may need to be applied for design. The mean value of the ratio of strength shows that the model overestimates the strength of the columns by 1% and there is a 6 % standard deviation associated with the model. The model is therefore shown to be quite acceptable for use in design. TABLE 4 STRENGTH COMPARISONS Test Name HSSI1 HSSI2 1 2 3 HSCI1 4 HSCI2 5 HSSH1 6 HSSH2 7 HSCB1 8 HSCB2 me (mm 2) 0 0 9,500 9,500 0 0 10,000 10,000 NA-NOT APPLICABLE AS (mm 2) 1,500 1,500 1,500 1,500 2,100 2,100 2,100 2,100 f~ (MPa) NA NA fy (MPa) 750 Nu.test (kN) 1,163 1,140 Nu.theory (kY) 1,125 1,125 Nu.test/ Nu.theory 1.03 1.01 750 50 750 1,408 1,600 0.88 50 750 1,590 1,600 0.99 NA 750 1,644 1,575 1.04 NA 750 1,561 1,575 0.99 50 750 1,940 2,075 0.93 50 750 2,132 2,075 1.03 Mean Standard Deviation 0.99 0.06 428 B. Uy CONCLUSIONS This paper has described the advantages of the use of high strength steel in tall buildings. A brief overview of projects to utilise these structural forms has been provided and the reasons for their use have been given. In particular the use of high strength steel plate in composite column forms was identified as a potentially useful application. An extensive experimental program was conducted to consider the axial compressive behaviour of both high strength steel columns and high strength steel composite columns. A compressive strength model was then proposed and found to be conservative in its prediction of the axial compressive strength of all the columns tested. Further research is however necessary to consider the behaviour of these columns in combined bending and compression. Furthermore the effects of interaction buckling due to local and global buckling also need to be considered as part of future research in this area. ACKNOWLEDGEMENTS This project was sponsored by an Australian Research Council Grant and supported in kind by Bisalloy Steels at Unanderra. This assistance is gratefully acknowledged. REFERENCES British Standards Institution (1994) Eurocode 4, ENV 1994-1-1 1994. Design of composite steel and concrete structures, Part 1.1, General Rules and Rules for Buildings. Hagiwara, Y., Kadono, A., Suzuki, T. Kubodera, I. Fukasawa, T. and Tanuma, Y. (1995) Application of HT780 high strength steel plate to structural member of super high rise building: Part 2 Reliability inspection of the structure. Proceedings of the Fifth East Asia - Pacific Conference on Structural Engineering and Construction, Building for the 21st Century, Gold Coast, pp. 2289-2294. Mochuziki, H., Yamashita, T., Kanaya, K., and Fukasawa, T. (1995) Application of HT780 high strength steel plate to structural member of super high rise building: Part 1 Development of high strength steel with heavy gauge and welding process. Proceedings of the Fifth East Asia - Pacific Conference on Structural Engineering and Construction, Building for the 21st Century, Gold Coast, pp. 2283-2288. Rasmussen, K.J.R., and Hancock, G.J. (1992) Plate slenderness limits for high strength steel sections. Journal of Constructional Steel Research, 23, pp. 73-96. Rasmussen, K.J.R., and Hancock, G.J. (1995) Tests of high strength steel columns. Journal of Constructional Steel Research, 34, pp. 27-52. Rosier, G.A. and Croll, J.E. (1987) High strength quenched and tempered steels in structures. Seminar Papers of Association of Consulting Structural Engineers of New South Wales, Steel in Structures, Sydney. Standards Australia (1990) Australian Standard, Steel Structures, AS4100-1990, Sydney, Australia. Uy, B. (1996) Behaviour and design of high strength steel-concrete filled box columns. Proceedings of the International Conference on Advances in Steel Structures, Hong Kong, pp. 455-460. CONCRETE FILLED COLD-FORMED C450 RHS COLUMNS SUBJECTED TO CYCLIC AXIAL LOADING X. L. Zhao, R. H. Grzebieta, P. Wong and C. Lee Department of Civil Engineering, Monash University, Clayton, VIC 3168, Australia ABSTRACT This paper describes a series of static and cyclic axial compression tests on empty and concrete-filled Rectangular Hollow Sections (RHS). Columns were made from cold formed C450 (450 MPa nominal yield stress) RHS with two different plate slenderness ratios. Two loading protocols were applied, namely Cyclic-Direct (full axial displacement applied and load oscillated) and Cyclic-Incremental (load oscillated at several accumulating axial displacement increments). First-cycle buckling loads were noted in the tests and compared with design loads predicted using various national standards and CIDECT. The paper further demonstrates that concrete filling increases the post-peak-load residual strength and reduces the rate of residual strength loss per cycle for thinner RHS columns subjected to overload cyclic situations. The authors also suggest that the AISC Seismic Provisions (1997) may be conservative with respect to compact behaviour for the specified width-to-wall thickness ratio limits. KEYWORDS Buckling, Bracing, Cold-Formed, Columns, Cyclic Loading, Concrete-Filling, Steel Hollow Sections INTRODUCTION Cold-formed tubular sections are widely used in steel structures. Cold-formed rectangular steel tubular braces have recently become popular in seismic regions, especially for high rise structures (Liu and Goel (1988)). Tests were performed in New Zealand (Walpole (1995)) and in USA (Jain et al. (1980), Sherman and Sully (1994)) on cold-formed RHS empty members. The test results showed that the capacity of cold- formed tubular members subjected to cyclic axial compression and tension reduced significantly due to local buckling in the sections. The magnitude of the local buckles also increased under repeated loading. Tests performed at Monash University have also demonstrated that the capacity of empty cold-formed tubular members reduces significantly when subjected to cyclic bending moments (Grzebieta et al. (1997)). Therefore it is necessary to study in detail the behaviour of cold-formed tubular sections during cyclic loading, and to investigate the possibility of improving the earthquake resistance of tubular members. It was suggested (Walpole (1995)) that the width-to-thickness limit for RHS under compression given in NZS3404 (1992) needs to be reduced for cold-formed sections under earthquake generated forces. In fact, a much lower limit of width-to-thickness ratio is given in AISC Seismic Provisions for Structural Steel Buildings (AISC (1997)) for RHS bracing members. 429 . high strength steel- concrete filled box columns. Proceedings of the International Conference on Advances in Steel Structures, Hong Kong, pp. 45 5-4 60. CONCRETE FILLED COLD-FORMED C450 RHS COLUMNS. specified width-to-wall thickness ratio limits. KEYWORDS Buckling, Bracing, Cold-Formed, Columns, Cyclic Loading, Concrete-Filling, Steel Hollow Sections INTRODUCTION Cold-formed tubular. testing machine platens and the instrumentation used in the testing. The test set-up highlights the end conditions, which were provided to ensure a uniform loading surface to the column. Using steel