Regarding PLAXIS 2D model depth?

When you are modelling foundation piles connected to a raft in PLAXIS 2D (I do understand that indeed piles are a 3D problem),

what is the minimum depth of the geometry? 

For example if the piles are 10 metres in length, can we assume the minimum length of the geometry is at least 3 times the pile length i.e 30 metres, or is this too much?

What if the geometry is selected as 15 metres, this would affect the results right?

Parents
  • Dear Jamal, 

    Defining the size of the model is one of the first and important steps when modelling with any finite element software. The model boundaries should be far enough so that they do not affect the area of your model.

    Why is that? Because in a numerical analysis we need to define fixities on the model boundaries which as you can imagine they do not exist in reality. Therefore, having a model in which the bottom boundary is too close to your piles, the soil movement will be restricted by bottom (full) fixities, thus giving you an unrealistic response.

    There is no general rule that says extend the boundary by a specific value as it highly depends on the soil conditions you are modelling. In some cases, 1.5 times the pile length can be enough, in others 2 times the pile length, but that is model specific.

    Always check the results and see that the deformations (contours) because of the piles are not extending to the model boundary where you expect almost nothing to happen.

    Note that one exception for the location of the model bottom boundary is the fact that you consider a (very stiff) rock being your base. In that case, you can skip modelling the rock material and instead replace it with your model bottom boundary as it would act as a "hard" boundary in reality, too.

    Answer Verified By: Jamal Hassan 

Reply
  • Dear Jamal, 

    Defining the size of the model is one of the first and important steps when modelling with any finite element software. The model boundaries should be far enough so that they do not affect the area of your model.

    Why is that? Because in a numerical analysis we need to define fixities on the model boundaries which as you can imagine they do not exist in reality. Therefore, having a model in which the bottom boundary is too close to your piles, the soil movement will be restricted by bottom (full) fixities, thus giving you an unrealistic response.

    There is no general rule that says extend the boundary by a specific value as it highly depends on the soil conditions you are modelling. In some cases, 1.5 times the pile length can be enough, in others 2 times the pile length, but that is model specific.

    Always check the results and see that the deformations (contours) because of the piles are not extending to the model boundary where you expect almost nothing to happen.

    Note that one exception for the location of the model bottom boundary is the fact that you consider a (very stiff) rock being your base. In that case, you can skip modelling the rock material and instead replace it with your model bottom boundary as it would act as a "hard" boundary in reality, too.

    Answer Verified By: Jamal Hassan 

Children
No Data