Morishita and Tomii 1982

From Composite Systems
Jump to: navigation, search

An experimental study on bond strength of square CFTs was presented. The effect of cyclic shearing force, concrete compressive strength, and magnitude of axial load on bond strength were examined.

Experimental Study, Results and Discussions

A total of twenty-four specimens were tested. The specimens were divided into six groups having different concrete compressive strength and magnitude of axial load. One specimen in each group was filled with concrete up to the top of the steel tube and identified as the confined specimen, while the others were identified as unconfined specimens, as they had a gap at the top of the steel tube. The measured yield strength of steel and the measured compressive strength of the concrete were varying from 49.2 to 50.9 ksi and 3.32 to 5.08 ksi, respectively. All the tubes were annealed from residual stresses. The ratio of the applied axial load over the nominal axial load capacity of the columns ranged from 3 to 22%. The D/t ratio for the specimens was 35.1. The cantilever specimens were tested under constant axial compression and cyclic shear force, both applied at the top of the column. The axial load was applied through end plates attached to the columns. Thus, the axial load was applied on to the steel tube alone for unconfined specimens and on to the steel tube and concrete simultaneously for the confined specimens. All the specimens were supported by steel and concrete simultaneously at the bottom.

The chord rotation versus shear force diaphragms of the confined and unconfined specimens showed good correlation under low axial load. However, the confined specimens had a larger ultimate shear force than the unconfined specimens in the case of a high level of axial compression. It was also found that when the axial load level was high, the ultimate shear force of the confined specimens increased with an increase in the compressive strength of the concrete. Nevertheless, the unconfined specimens were not affected with concrete strength at high axial load levels.

When the specimens were subjected to axial compression only, the confined specimens exhibited continuity between the axial strains of the steel and concrete along the whole length of the column. For the unconfined specimens, the continuity was limited to the bottom portion of the CFT, where both the concrete and the steel were in contact with the endplate at the bottom of the CFT. However, when high axial load was applied, a portion of the load still shifted to the concrete towards the bottom of the CFT, but the continuity of strains was not satisfied along the length of the columns in the unconfined specimens.

In the case of cyclic shearing force, which resulted in bending moment in the specimens and thus in cyclic stresses being applied to the CFTs, it was found that the constant axial compression acting on the steel tube decreased along the column length as the cyclic shear load value increased. The rate of decrease was larger at the bottom than it was at the top. However, the steel tube recovered its axial load after the shear load was removed.

The inelasticity of the steel tubes was examined at 1% and 2% chord rotation levels through assessment of the strain gage readings on the steel tubes. It was found that more sections along the length of the specimens towards the top exhibited inelastic response as the axial load level increased.

The following formulas were proposed for the mean bond stress and average slip between steel tube and concrete, respectively:

where l0 is the height of the column and l3 is the length from the fixed end to the section numbered as 3, which was close to the mid-height; and:

where dso is the slip at the top of the column, lo is the length of the column, and dz is the infinitesimal column length.

From the experimental results, it was found that average bond strength varied noticeably with the shear force. The bond strength was also a little larger in the case of high axial load, due to enhanced confinement, and it was not affected by the compressive strength of concrete. As the slip got larger in later cycles of loading, the bond strength within each cycle remained approximately constant, thus reaching a plateau . The range of the bond stress at the peak chord rotation was 21.3 to 49.8 psi and 21.3 to 42.7 psi for small and high axial load levels, respectively. Therefore, it was decided that 21.3 psi was a reasonable, conservative estimate of bond strength.


Morishita, Y. and Tomii, M. (1982), “Experimental Studies on Bond Strength Between Square Steel Tube and Encased Concrete Core under Cyclic Shearing Force and Constant Axial Force,” Transactions of the Japan Concrete Institute, Vol. 4, pp. 363-370.