Interaction and contact elements

Post Reply
OYK
Posts: 4
Joined: Wed Mar 01, 2023 7:00 pm

Interaction and contact elements

Post by OYK » Mon Mar 27, 2023 1:52 pm

Dear Forum,

I am building a model for a footing overturning from a horizontal load. At the moment I am using a fully connected model by merging however the resulting load displacement is not correct. I want to use 2 merged blocks instead (footing and soil) and create an interaction between them. SSP Elements are used for the solid elements. What elements and properties should I use and how do I draw the interaction (node to node?)? The parameters I am aiming to add is 15 degrees friction angle or 0.177 mu to this interaction, if needed the Youngs Modulus is 25 MPa and Poissons Ratio of 0.3. I have attached a test file for demonstration of what I am trying to achieve as well as an image. Note part of the footing is in the soil not just sitting on top.

Many thanks!
Image.JPG
Image.JPG (61.64 KiB) Viewed 2265 times
Attachments
Contact Element Test.zip
(455.77 KiB) Downloaded 159 times

STKO Team
Posts: 2297
Joined: Tue Oct 29, 2019 8:45 am

Re: Interaction and contact elements

Post by STKO Team » Tue Mar 28, 2023 8:04 am

Check our webinar #24 Implicit-Explicit Contact Element Applied Examples: soil-foundation interaction and RC structures here:
https://asdea.eu/software/advanced-e-learning-courses/

OYK
Posts: 4
Joined: Wed Mar 01, 2023 7:00 pm

Re: Interaction and contact elements

Post by OYK » Tue Mar 28, 2023 10:18 pm

STKO Team wrote:
Tue Mar 28, 2023 8:04 am
Check our webinar #24 Implicit-Explicit Contact Element Applied Examples: soil-foundation interaction and RC structures here:
https://asdea.eu/software/advanced-e-learning-courses/
Hi,

Thanks for the response. I have watched the webinar and tried applying it to my model. However, I have not managed to get it working. I have attached a simplified version I was using to test the elements on but it doesn't work. I managed to get 2 blocks of the same size to work but with the lower block being larger it doesn't work.
Attachments
Contact Element.zip
(128.69 KiB) Downloaded 153 times

ValentinaSB
Posts: 61
Joined: Thu Dec 09, 2021 10:53 am

Re: Interaction and contact elements

Post by ValentinaSB » Wed Mar 29, 2023 9:04 am

Hi OYK,
I see a few problems in your model:
1. you are using a contact element with orientation type from the local axis, but you don't assign the local axis.
2. I would keep geometry in a way to have uniform mesh
3. Check that you are coherent in units of measurement for your material properties and applied loads. You apply 20200 as a volume force, keep in mind that it will be multiplied by the volume to which you assign it. In your case by 1000 (10x10x10)

I have changed the contact as in the webinar in the last example for the slab-soil, in this case, you can also use finer mesh for your footing. It runs with a smaller value of the load, but you need to look into the parameters.

Let us know if you have any trouble.

Best,
Valentina
Attachments
Block interaction debug_corrected.rar
(112.31 KiB) Downloaded 152 times

OYK
Posts: 4
Joined: Wed Mar 01, 2023 7:00 pm

Re: Interaction and contact elements

Post by OYK » Wed Mar 29, 2023 11:52 am

ValentinaSB wrote:
Wed Mar 29, 2023 9:04 am
Hi OYK,
I see a few problems in your model:
1. you are using a contact element with orientation type from the local axis, but you don't assign the local axis.
2. I would keep geometry in a way to have uniform mesh
3. Check that you are coherent in units of measurement for your material properties and applied loads. You apply 20200 as a volume force, keep in mind that it will be multiplied by the volume to which you assign it. In your case by 1000 (10x10x10)

I have changed the contact as in the webinar in the last example for the slab-soil, in this case, you can also use finer mesh for your footing. It runs with a smaller value of the load, but you need to look into the parameters.

Let us know if you have any trouble.

Best,
Valentina

Hi Valentina,

Thank you for your quick response. Just a couple of questions for your response:

1. The files from the webinar do not have a local axis assigned, to check if there is no local axis assigned is the global used as default with the Z direction?

2. In terms of uniform mesh is this fully necessary as I plan to have a much larger model where a uniform mesh of same block sizes would be unfeasible is terms of processing power. The mesh I have used is inspired from the soil pile webinar. So long as the nodes match for the footing will this still work with elements becoming larger the further out they are from the interaction?

3. With the units for STKO they are not labelled. The inputs I used are in SI so the 20200 is meant to represent 20.2 kN/m^3 is this correct. For material properties, one of the blocks is 70 GPa Youngs Modulus so I input this as 7e10, would these ratios make sense?

4. I am using a Drucker Prager Material for the soil, it seems to have issues as I have got it working with 2 elastic isotropic materials. Is there a reason for this? I have taken an example from the webinar and applied a Drucker Material to it and it does not work when I do this, the file is attached. The parameters I have used for the Drucker Prager work in my other models so if you are able to help with this that would be great.
T04_contact3d_orientation with DP.zip
(251.48 KiB) Downloaded 146 times
Many thanks

ValentinaSB
Posts: 61
Joined: Thu Dec 09, 2021 10:53 am

Re: Interaction and contact elements

Post by ValentinaSB » Tue Apr 11, 2023 9:30 am

Hi OYK,

Sorry for the late response.

1. You are right, you can use it in this case.

2. Yes, it should work. But you can also use a different approach for a very big part of soli. You can create two geometry for the soil with coarse and fine mesh and use embed element between them. In this case, you will use contact only between your building and the soil with the fine mesh. A good example is in webinar #25 New Element Presentation - ASDAbsorbingBoundary
https://youtu.be/JCVExTm8Cgw

3. STKO doesn't have units of measurement, so you choose them from the beginning and you need to be coherent with them. For the big models is always better to work in meters. I can see now that you were using N and m, so it should be fine. The problem was with Drucker Prager material.

4. The Drucker Prager has many issues (you can also see other users on the forum that had tried to calibrate it). You can try to use another material. If you try to calibrate this one be sure that in that webinar example all units of measure are coherent, try to see in tester your material with more divisions.

Let us know how it is going.

Best,
Valentina

Post Reply