Problem with the values of reaction response of reinforced concrete in ABAQUS

[T.E.Dada]

Hi everyone,

I have been trying to model a concrete retaining wall using ABAQUS. There has been progress so far, but the values of the responses I have been getting are very low compared to what is obtainable using classical equilibrium approach.

I suppose the issue was with the inconsistency of the units, however, checking through the model, I have not been able to debug the problem.

I used an SI unit of mm, N, N/mm[sup]2[/sup] and tons/mm[sup]3[/sup] for density

Can you help me out with a way to solve this problem?

 

[FEA way]

I had a look at your model. The units seem correct and I didn’t notice any other modeling errors. Can you say more about how you compare this to theoretical results ? Which variables do you compare and how do you obtain the theoretical values ? This model is quite complex as it includes metal plasticity, CDP with damage, complex rebar structure with various cross-sectional areas and spatially variable pressure loading. Thus it may be hard to get a correct estimate of its response.

 

[T.E.Dada]

Many thanks for your response.

I wanted the force - displacement curve; the force representing the reaction at the fixed end and the displacement at the free end. Initially, I was using only the reaction at a specific node and the value seems to be very low, however taking a sum of all my desired reaction components makes a lot of sense to me. The displacement on the other hand is an average of specific node across the top of the wall. Regarding the sum of the reaction, I would not know if I am doing the right thing.

With regards to the curve, there seems to be no failure point, even when I load with very high values. Instead, the higher the load, the lower the number of iterations before aborting due to "Too many increments needed to complete the step". Is there any thing I need to put right or define to have this failure point of the curve?

I have attached a typical plot of my load against displacement

 

[FEA way]

You can plot the sum of reaction forces at fixed nodes vs displacement of selected node at the free end. To see damage you may have to adjust the material properties. But I would perform the single element test with this material model first.

 

[T.E.Dada]

Many thanks for your response.

I really appreciate.

 

[Francis666]

Hi T.E. Dada. I have same problem- The plasticity/creep/connector friction algorithm did not converge at x point. How do you solve that? I also modelled a concrete damaged plasticity model.