Secant Pile-Local Soil Supports

[Anonymous]

Hi,

I am modelling a circular formation of secant piles with 3 m embedment below the floor slab (image attached). The piles are for a stormwater inlet structure and will eventually have openings added.

There are two things I am having issues with.

1) how to model a horizontal line load 13.4 m from the base to half of the structure (resultant seismic thrust force). I tried drawing an arc and applying it as a line load but the load would not assign to the panel properly.

2) How to include the soil stiffness for each soil layer radially (set to the local axis of the piles). I currently have spring supports assigned to nodes (different stiffness for each layer) however these are assigned to the global axis which is causing a much larger design moment than what should be there. I can't see any easy way of changing this to the local axis.

The only other way I can think of doing this is splitting the panel horizontally according to each soil layer and applying different soil stiffness to these through the 'panel thickness' option.

If anyone had any ideas I would be very grateful.

I have attached my model. Soil layers are between Horizontal Axes: AB, BC, CD, and D below.

Cheers,

Trent

[Pawel.Pulak · Accepted solution]

Topics solved outside of the forum.

Here some summary:

1) There was no problem to apply the load to arc defined coherently with the cylinder - and this load was transferred to the cylinder. The problems observed by Trent probably resulted from inaccuracies between the arc and the cylinder - the arc was not detected as coherent with the surface of cylinder

2) Supports defined in Trent's model are nodal ones and in global coordinates. The soil action perpendicular the the cylinder surface can be modelled modifying these supports to act in local coordinates with direction governed by the center of cylinder. But it would require different support labels for each Z level of nodes.

That is why it is better to use the method mentioned by Trent: "splitting the panel horizontally according to each soil layer and applying different soil stiffness to these through the 'panel thickness' option.". It can be done for his model defining panels for selections of finite elements. The disadvantage  of this method is that panels defined this way have no contours so it is impossible to modify or re-mesh them.

The best method seems re-definition of the model using multisegment extrusion of circle or arcs to model zones corresponding to various soil strata and then assigning various soil properties to appropriate faces in panel thickness

---------------------------------------------
If this post answers your question please click the "ACCEPT AS SOLUTION" button (the button visible when logged to the forum). It will help everyone to find answer more quickly!

Regards,