Morishita, Tomii, and Yoshimura 1979
Two series of experiments were conducted to investigate the bond stress between steel and concrete in CFTs. In the first series, only circular columns were tested, while the second series covered both square and octagonal columns. The changes in bond strength due to the variations in cross section shape and the compressive strength of concrete were examined.
Experimental Study, Results and Discussions
The specimens were tested under monotonic axial compression, which was applied to the steel tube alone at the top of the CFT. However, both the steel and concrete was supported simultaneously at the bottom. The D/t ratios of the specimens ranged between 34.9 and 46.9. The measured yield strength of steel was either 36.6 or 37.0 ksi and the measured compressive strength of concrete was ranging from 2.73 to 4.85 ksi. The steel tubes were annealed so that they were free from residual stresses.
Strain compatibility between the steel and concrete along the whole column length was observed when small strain values were applied at the top of the steel tubes. However, as the strain level increased, bond stresses and some slip started to develop between the steel and concrete. Thus, the strains became the highest at the top of the steel tube and decreased gradually toward the bottom. The strain compatibility was observed only at the bottom region, which indicated that the axial force transfer from the steel tube to the concrete was completed at the bottom region. For a strain level at the top of the CFT of 5x10-4, no strain compatibility between the steel and concrete was observed. The circular specimens with high strength concrete, as compared to those with low strength concrete, had lower strain values at the limit of exhibiting strain compatibility along the entire length of the column. This was attributed to the greater modulus of elasticity of high strength concrete. Moreover, the bond strength was smaller when high strength concrete was used for circular specimens. This was due to the shrinkage of high strength concrete, which was larger than that of low strength concrete. In addition, it was also due to the fact that Poisson’s ratio of steel was greater than that of concrete in the elastic region. The same trends for the strain distribution along the length of the column were also observed in the case of square and octagonal CFT specimens as were observed in circular CFTs . However, in these specimens, high strength concrete was observed to have no effect on the bond strength and on the strain values at the limit of strain compatibility along the entire length and
Analytical Study
The following formulation was presented for the average bond stress (fmb):
where the first term in the parenthesis represents the normal stress of the steel tube at the top of the column, while the second term σsl is the normal stress of the steel tube at the point where strain compatibility between steel and concrete occurred. The length from the top of the column to the point of strain compatibility is represented by la.
The formulation for the average nominal slip (Ds) was given as:
where dso is the nominal slip at the top of the column, l0 is the length of the column, and dz stands for the infinitesimal column length.
Using the equations presented above, the range of bond strength for the circular specimens was determined to be 28.5 to 56.9 psi, and it was found to be 21.3 to 42.7 psi for the square and octagonal specimens. The bond stresses were found to decrease at the initial stages of slip, after which they were observed to remain constant with increasing slip.
Reference
Morishita, Y., Tomii, M., and Yoshimura, K. (1979a). “Experimental Studies on Bond Strength in Concrete Filled Circular Steel Tubular Columns Subjected to Axial Loads,” Transactions of the Japan Concrete Institute, Vol. 1, pp. 351-358.
Morishita, Y., Tomii, M., and Yoshimura, K. (1979b). “Experimental Studies on Bond Strength in Concrete Filled Square and Octagonal Steel Tubular Columns Subjected to Axial Loads,” Transactions of the Japan Concrete Institute, Vol. 1, pp. 359-366.