DavidCam
Structural
- Jun 28, 2011
- 8
Hi everyone,
I'm trying to model a simple RC beam to compare the ultimate load of this simple structure model with hand calculations and verify that the difference in the results is small enough, in an attempt to justify that the model works properly and then use it to simulate a more complex structure.
The structure is a cantilever rectangular beam, fixed at the left end and loaded at the right end. The height of the beam is 1 m, the width is 0.5 m.
The hand calculations are based on the Eurocode 2 hypothesis. The ultimate bending moment for the mentioned cross-section is 504 mkN.
The FEM is built with ANSYS APDL 16. It uses SOLID65 for the concrete and LINK(180) for the reinforcement. The concrete material is modelled with CONCR failure criteria and MISO; the reinforcement is modelled with BISO. The structure is simplified by symmetry. The reinforcement elements share nodes with the concrete elements. The load is applied in multiple load steps. At the beggining, when the load is low, the behaviour is mainly as linear-elastic. As the load increases and the tensile stress reaches the tensile strength, the concrete cracks and the stresses are redistributed. When the cracks open, the stress at the reinforcement also increases. Apparently the general behaviour is correct. But the model fails to converge at a much lower load than the hand-assessed.
I've tried all sorts of things to get the expected behaviour, like changing the material parameters, solution controls, applying the loads both as a pressure at the upper edge and as a imposed displacement, etc. But there's no way I can achieve it. If there's someone who has made a similar study I'd really appreciate some recommendation. I attach the APDL file used to create an example of the model.
Thanks!
I'm trying to model a simple RC beam to compare the ultimate load of this simple structure model with hand calculations and verify that the difference in the results is small enough, in an attempt to justify that the model works properly and then use it to simulate a more complex structure.
The structure is a cantilever rectangular beam, fixed at the left end and loaded at the right end. The height of the beam is 1 m, the width is 0.5 m.
The hand calculations are based on the Eurocode 2 hypothesis. The ultimate bending moment for the mentioned cross-section is 504 mkN.
The FEM is built with ANSYS APDL 16. It uses SOLID65 for the concrete and LINK(180) for the reinforcement. The concrete material is modelled with CONCR failure criteria and MISO; the reinforcement is modelled with BISO. The structure is simplified by symmetry. The reinforcement elements share nodes with the concrete elements. The load is applied in multiple load steps. At the beggining, when the load is low, the behaviour is mainly as linear-elastic. As the load increases and the tensile stress reaches the tensile strength, the concrete cracks and the stresses are redistributed. When the cracks open, the stress at the reinforcement also increases. Apparently the general behaviour is correct. But the model fails to converge at a much lower load than the hand-assessed.
I've tried all sorts of things to get the expected behaviour, like changing the material parameters, solution controls, applying the loads both as a pressure at the upper edge and as a imposed displacement, etc. But there's no way I can achieve it. If there's someone who has made a similar study I'd really appreciate some recommendation. I attach the APDL file used to create an example of the model.
Thanks!