chapter
1 Pages

## FINITE ELEMENT ANALYSIS Since the test results showed that the shape of ruptured cone of concrete was almost Identical in all the cases where the embedment length varied from 4 cm to 24 cm, the analysis was conducted only on the case of 8 cm embedment length with 60x60x30 cm concrete block. Using the rectangular elements with four nodal points, the linear elastic finite element analysis was done taking the axi-symmetric procedure. Assuming a crack to propagate along the given route obtained by the test results, the stress distribution in concrete was computed at the each loading level. The crack propagation was judged when the principal tensile stress in the element exceeded the tensile strength of concrete. Once the cracking was recognized in the element remeshing of element was proceeded so as to eliminate the stress transfer along the cracked surface. ANALYTICAL RESULTS According to the elastic analysis, the maximum pull-out capacity was evaluated when the crack propagated up to 0.43 of the assumed full crack stress f t length. This seemed to coincide the distribution result by Eligehausen and et al [2] which showed the ratio of 0.45 by non-linear finite element analysis. cracked elastic However, the computed capacity of 3.1 failure surface tonf at the maximum load was far nodal below the test result of 6.3 tonf. crack °^rce V This was because the analysis did not count the stress transfer in the wid lth f \'< cracked region. Then, the linear elastic analysis was, again, repeated so that the cracked element could bear the tensile stress in inverse proportion to the crack width. Such resistances were applied to the Figure 3. Crack model element as the nodal forces as shown in Figure 3.

STRESS TRANSFER MECHANISM

Figure 4. Stress-crack width [3]

Figure 5. Stress distribution