Discussion Groups

Robot Structural Analysis

Reply
Member
FilipePagaimo
Posts: 4
Registered: ‎10-13-2011
Accepted Solution

Unable to get soil reactions with elastic planar supports

899 Views, 4 Replies
10-23-2011 04:06 PM

Hi,

I’m working on a model that needs to use a foundation slab with an elastic support. I used Planar Supports to assign such kind of support:

1.png

For this purpose (put a question) I created a quite simple model of a rectangular foundation slab 5x6m. It’s loaded with self-weigh and with a Live load of -10kN/m2:

2.png

 

I would like to get soil reactions - K map but it seems not possible. I got no results, only zeros:

3.png

 

 

I think the reason for this is because I assigned elastic supports with planar supports dialog box and not using Parameters of foundations elasticity at New Thickness Finite element dialog box:

4.png

In fact I tried to use it, but the original model has horizontal loads, so I would need to restrain horizontal movements (X and Y) as well as Z rotation. I tried several ways but couldn’t get convergence of the model, so I decided the approached with planar supports.

 

Getting back to the question, I would need to find out if stresses induced to soil are below its capacity. Maybe I’m missing some way to get those values, but until now I could only get reactions (kN) values at supports.. No soil stresses/ reactions:

5.png

 

Can you help me getting results as soil stresses in spite of forces at supports?

Thank you in advance,

Filipe

 

PS:

I'm attaching the model I used for this question.

Please use plain text.
Product Support
Artur.Kosakowski
Posts: 4,596
Registered: ‎12-17-2010

Re: Unable to get soil reactions with elastic planar supports

10-24-2011 12:03 AM in reply to: FilipePagaimo

Filipe,

 

If you use nodal (surface) supports then you don't get maps for elastic soil as it is not defined. In such case the values of reactions are listed in the table of reactions. Maps are available only for a definition of elasticity assigned to a panel thickness. I assume that the lack of convergence is simply the instability due to lack of horizontal restrains in the model with the elastic soil you created. The solution is the definition of the elastic soil in the panel thickness and either definition of additional nodal supports that restrain horizontal movement (see the attached file) or add KX and KY values in the definition of the elastic soil itself.

 

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

 



Artur Kosakowski
Please use plain text.
Member
FilipePagaimo
Posts: 4
Registered: ‎10-13-2011

Re: Unable to get soil reactions with elastic planar supports

10-24-2011 02:54 AM in reply to: Artur.Kosakowski

Thank you Artur for your time,

At the model without convergence I tried to assign some nodal supports to restrain horizontal movements and rotation around Z but it didn't work.

Regarding your recommendations I suppose I can blend approaches so I can use planar supports (instead of nodal) to put those restrains like picture below:

1.png

 

and then assign the KZ at Finite Element thickness menu:

2.png

 

I tried it and got immediate convergence, even with uplift. :smileyhappy:

 

Thank you, Filipe

Please use plain text.
Product Support
Artur.Kosakowski
Posts: 4,596
Registered: ‎12-17-2010

Re: Unable to get soil reactions with elastic planar supports

10-24-2011 03:07 AM in reply to: FilipePagaimo

Filipe,

 

So you decided on the 'either' approach :smileyhappy:

 

For the possible lack of convergence with uplift in the future:

1. Try to mark matrix update after each iteration (see the picture)

2. Check if the overturning moment (moment caused by upwards and horizontal forces about the edge of the raft foundation) is smaller than the stabilization one (due to downwards loads such as e.g. selfweight) about the same axis.

 



Artur Kosakowski
Please use plain text.
Member
FilipePagaimo
Posts: 4
Registered: ‎10-13-2011

Re: Unable to get soil reactions with elastic planar supports

10-24-2011 03:21 AM in reply to: Artur.Kosakowski

Thank you once again,

I used to check that list :smileyhappy:

I think that at other previous models my problem is that I used to assign a Kx and Ky value and only restrain global Z rotation with a nodal support. Usually (taking with care the statement) we "never have a foundation slab sliding" due to its dimensions, weigh and position so blocking horizontal displacements seems a possible solution.

Just wrote this note trying to "complete" my approach if it becomes useful for someone.

 

Cheers,

Filipe

Please use plain text.