Calibration of Damage Parameters by Simulation of the Kahn Test The parameters fc of the GTN model and σmax und δc of the decohesion law, respectively, were adjusted to meet the experimental load vs displacement and load vs crack growth curves of the Kahn specimen, see Figure 2. The latter was measured from three specimens for which tests have been stopped at different amounts of crack growth. Both curves are met with satisfactory accuracy by both models, taking fc = 0.018 and σmax = 580 MPa, δc = 0.08 mm, or Γ0 = 26.1 N/mm, respectively. 0.0 0.5 1.0 1.5 2.0 0 1 2 3 4 5 6 7 CMOD (mm) test CZM GTN Kahn specimen 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 ∆a (mm) test CZM GTN Figure 2: Kahn test and simulation by GTN and CZM Verification by Simulation of the M(T) Test The M(T) has now been simulated, taking these parameters, see Figure 3. No perfect agreement between test and simulation results could be obtained. Maximum load is overestimated and reached too early with respect to elongation vL. On the other hand, crack growth is too fast in the simulation. Keeping in mind that there is a factor of 20 in size between the Kahn and the M(T) specimen, and the former is notched whereas the latter is cracked, the numerical prediction is not too bad. Some additional studies of the involved elastic and dissipated mechanical energies will further elucidate the transfer problems arising between small and large specimens. 0 1 2 3 4 0 100 200 300 load (kN) vL (mm) test CZM GTN 0 10 20 0 100 200 300 ∆a (mm) test CZM GTN load (kN) Centre-Cracked Panel M(T) 30 Figure 3: M(T) test and simulation by GTN and CZM
RkJQdWJsaXNoZXIy MjM0NDE=