Pan and Zhong 1991
The authors presented a definition of CFT strength under axial loading, based on a load-deformation curve they derived in earlier papers. They strove to present a definition of ultimate strength which would allow a direct comparison of results obtained by investigators using disparate experimental methods. To accomplish this, the authors recommended using load-deflection curves.
Theoretical Discussion
The authors defined two types of behavior for axially loaded CFT members based on the ratio of the column length to its diameter (L/D). Load-deformation curves for L/D 4-5 showed an unloading portion of the curve as the deformation increases beyond the maximum load. Specimens of this type failed by overall buckling. As L/D increased beyond L/D = 5, the curve showed a sharper drop-off indicating a more sudden failure and the unloading occurred at lower and lower loads. For tubes with an L/D ratio of between 3 and 3.5, the curve ascended to failure. This indicated a specimen dominated by shear failure in the concrete and local outward buckling of the tube. The authors used these latter specimens to study the fundamental behavior of axially loaded CFTs. For such specimens, they showed that the steel ratio may also affect the shape of the load-deformation curve. A steel ratio below 6% or 7% produced results much like the more slender columns, showing a descending curve after the maximum axial load was reached. In practice, though, the steel ratio will generally be higher than 7%.
Axially Loaded CFTs. A number of different definitions of bearing capacity for axially loaded CFTs have been given in the literature. The authors briefly touched on several of these, ranging from a low value of axial load given as the yield strength of the tube plus the load on the concrete at the time of steel yielding (Furlong, 1967) to a high value obtained by assuming the steel yields entirely in hoop tension with no longitudinal strain and the concrete supports the full load under triaxial compression (Bondale and Clark, 1966). In fact, this steel stress state will never occur even if only the concrete is loaded because of interaction between the steel and concrete. The authors noted, however, that the values obtained by this formulation may not be far off since the steel goes into strain hardening and the concrete is able to reach the triaxial stresses in the aforementioned definition. The definitions were compared by plotting the ultimate load points on a load-deformation curve which allowed a legitimate comparison between the methods to be made.
Further Research
In conclusion, the authors stressed the variation existing in different methods and cautioned against direct comparison between tests. They lauded the load-deformation relationship as one of the better ways to obtain knowledge of CFT behavior.
Reference
Pan, Y. and Zhong, S. (1991). “Discussion on the Definition of Strength of Concrete Filled Steel Tubes,” Proceedings of the Third International Conference on Steel-Concrete Composite Structures, Wakabayashi, M. (ed.), Fukuoka, Japan, September 26-29, 1991, Association for International Cooperation and Research in Steel-Concrete Composite Structures, pp. 7-12.