PSI - Issue 44

5

A. Viskovic et al. / Procedia Structural Integrity 44 (2023) 1348–1355 A. Viskovic et al. / Structural Integrity Procedia 00 (2022) 000–000

1352

3.3. Hooping system material properties The confinement system is realised through high versatile textile ribbons capable to be shaped as required and to experience large deformations without significant stress induced. In view of the multiple applications offered by the market, there exists the need to select the best ribbon for the strengthening interventions. This preliminary study is devoted to provide a first answer to the above questions. Two different confinement systems will be subjected to the same experimental sequence and the results compared between the two and with the unconfined reference sample. The sequence foreseen: (i) design of the system; (ii) installation of the confinement system; (iii) testing in simple compression. The values of the mechanical properties are given in Table 3 and 4. The two systems considered differ mainly for the tensile strength and the elastic modulus values. The ratio of these two quantities is quite high and will be used to verify any dependence of the hooping system on the hooping system behaviour.

Table 3. Material Properties Dyneema-Nylon

Property

Dyneema 970 kg/mc 110000 GPa 3500 MPa

Nylon

Specific weight Elastic modulus Tensile strength

70 MPa

2200 MPa 1350 kg/mc

Elongation

2-5%

30-200%

4. Numerical model and analysis Geometry. The 3D numerical model has been implemented in the Midas FEA environment using “solid” type f.e. The mesh discretization, being constrained by the mortar dimension, comes out to be very dense. So that in order to get a computationally efficient, yet precise numerical model, the dimension of the solid f.e. has been varied according to the deformation gradient: the lesser the gradient is, the higher the solid dimensions are and vice versa. Nonetheless, a huge number of f.e. (Nc ≈ 280’000 by 6 dofs each) is still required for proper modelling in view of the mortar thickness (10mm) and its layers subdivision (min 6 layers of about 1.2 mm each, for unbiased results). A drastic reduction of the f.e. number is obtained if a proper prism extracted from the column is used for the analysis, Fig. 3a. This prism is extracted from the column through cuts according to the symmetry planes of the column and constitutes the geometry of the numerical model, Fig. 3b. Finally, the model (prism) comprises two bricks and one mortar layer with a total number of f.e. (Np ≈ Nc/4). To comply with the symmetry, 2 vertical faces of the numerical model are free to move whereas the other 2 faces are prevented to move. The vertical faces allow only vertical displacements. Constitutive models. The total strain crack constitutive model was used for brick and mortar using the material properties listed in Table 3. The same has been done for the reinforcing materials. The analyses were carried out for the column with and without the hooping system. In the first case a prestress equal to 4’200 N was given to the ribbons through an equivalent thermal change. Numerical simulations. Nonlinear numerical analysis was performed under displacement control up to 1 mm. The analyses were carried out for the column with and without the hooping system. The evaluation of the improvement of the column behavior in terms of strength and ductility is the main concern of the numerical simulations.