asdeasoft.net

Dear STKO staff, dear Colleagues,
I am working on the modelling of a wall strengthened with a fiber-reinforced mortar. To avoid excessive computation time, I decided to work in 2D using 2D faces and model "Damage TC3D".
With this approach, I obtained satisfactory results for the configuration without the coating.
For the strengthened configuration, I used the same base model and added two additional layers on both sides, placed 20 mm away from the wall. I introduced the equaldof constraint, assigning the wall as the master and the coating layers as the slaves. Although this modelling choice seems reasonable from a mechanical point of view, the results are clearly incorrect. The stiffness of the wall + coating system not only appears excessively high, but also shows no sensitivity to the coating input parameters.
Attached below some pictures of the one-side strenghtened wall and the load-deflection curve (the background plot refers to the unstrengthened configuration).
I look forward to your kind support.

From your description your approach should be fine (although it could be done more efficiently using layered shells).
Without more details, I cannot give you any advice.

Thank you for your reply.
Please find below some additional details:
-the middle layer is the masonry wall;
-the other two layers are coatings made of fiber-reinforced mortar;
-the two layers are connected to the middle one using equaldof constraints, with the middle layer defined as the master;
-the load is applied to the upper corner as an edge force;
-the lower corner has UX and UY fixed, while the upper corner has UX fixed only;
-the analysis is displacement-controlled, with the force applied at the vertex of the upper corner.

Despite the analysis runs, the results still appear to be incorrect because:
- the wall + coatings system shows lower performance in ultimate displacement and the same peak strength of the URM wall (in the attached figure the experimental test of the strenghtened wall is depicted with a dashed line and the numerical results with a solid line);
- the analysis does not show any sensitivity to the coating input parameters.

I have attached the file I am working on for your convenience.

Thank you.

Unfortunately there is a mistake in your model.
The middle surface has the normal vector (local Z axis) pointing downwards. In a 2D analysis this is not possible and generates a negative jacobian (all results will be messed up).

Also you attached with EDOF incompatible meshes.
Why don't you try with Shell elements and multilayer shells? it will be much easier