Topic: Long time formed shear crack failure in continuous beams


While there are different types of shear cracks, the critical shear crack of continuous beams that is formed over a longer period (and not as abruply as the presented LSB in the Tutorial) makes it really difficult to get good results. I have tried to change different parameters, but at some point, where the LD of the experiments continue with an smaller gradient, the FEM forms the shear crack which leads to failure. At some models, the LD reaches only 70% of the force and 30% of the actual displacement.

->Is there some other way to get better results? I have tried to reduce the mesh-size until 14 elements per height (my computer can t handle more than that) and to lower the step-size. The continuous beam is modeled by 2 applied forces (where one is dependent on the other to give a pre-calculated M and V at a certain point). I have also applied symmetry along the beam, dividing it by 2. Both, the AL and NR Methods were tested.

->Is there any possibility to improve the connection between the relative displacement of the crack-sides which leads, next to the Aggregate Interlock, to a remaining support of the concrete on the reinforcement?

Thank you in advance.


Re: Long time formed shear crack failure in continuous beams

Dear Dmitry,
1. if you mention a longer period, I assume you are working with the Creep module of ATENA Science?

2. It is not clear from your question which "level" of modelling the long-time response you are using:

A. using the Static module (or ATENA Engineering) with adjusted material properties (E modulus or/and other),

B. using the Creep module with properties constant for the whole volume (i.e., only defining the average distance from the surface),

C. using the Creep module with imported temperature or/and humidity from the Transport module


3. Or does "long term" rather mean "due to cyclic loading"? If so, please see ATENA Troubleshooting, 2.2.26 How can I model cyclic or fatigue loading?

4. For force loading up to failure, you have to use the Arc Length solution method (see also ATENA Troubleshooting, Solution methods).

5. If you have further questions / seek help with your particular model, you can follow Troubleshooting, 2.1.1 to send us your model along with information about what are you modelling and what are the questions to be answered by your analysis.