Applying lateral loads on walls

[yingying]

Dear Sir/Madam

I modeled a masonry wall using the BBarbrick element. A cyclic load was applied to the top of the wall using the method in your course "Shell and Solid Models with Embedded Rebars", i.e. using rigid links and zeroLengthContactASDimplex element, but I get incorrect load-displacement curves. Can you help to see what is the cause of this, also how should the boundary conditions be set correctly for cyclic loads applied to a loading beam on a shear wall. The boundary conditions in the test approximate an in-plane cantilever wall. Also, I'm a little confused as to why in the course "Shell and Solid Models with Embedded Rebars" the cyclic loads are applied to a point on the steel loaded plate instead of a face. Could you help me find the answer here? Thanks a lot and best wishes.

[kesavapraba]

Hi, Please help us to understand your model with more details. As of now it is challenging to understand your requirement completely without additional details.

Can you help to see what is the cause of this,

Please share the model here to debug the issue.

also how should the boundary conditions be set correctly for cyclic loads applied to a loading beam on a shear wall.

What boundary conditions do you want to simulate for the model? Please provide us with more details.

Also, I'm a little confused as to why in the course "Shell and Solid Models with Embedded Rebars" the cyclic loads are applied to a point on the steel loaded plate instead of a face. Could you help me find the answer here?

As explained in the webinar, you can use different approaches to apply the cyclic loading. The approach followed in the webinar applies the loading using zerolength contact element, hence two vertices (points) were required for the purposes with rigid beam in between. However, this (point vs face) will not affect the results much as the displacement will be uniform on the face of the steel place and the steel beam is much stiffer compared to concrete system. You can verify it with the horizontal dispalcement component from the model.

[yingying]

Dear Prabakaran Kesavan:

Thank you very much for your reply, here I attached the model. The bottom of the shear wall is completely fixed. The boundary conditions at the loaded end of the shear wall are cantilevered, meaning that the in-plane bending moment at the loading end can be released.

Thanks a lot and best wishes.
yingying

[kesavapraba]

Hi, yingying. I have changed the following three things in your last model. Now it runs fine. Please check the attached model here.
[1] Please create the selection set for the restrained region at the bottom of the wall and monitor the reaction there. The problem with the reaction not showing up in the monitor appears when the displacement reverses. This happens due to the contact being lost at the loading point during reversal of displacement loading occurs.

[2] In the monitor, Plot-Y Axis-> Operation -> Set 'Sum'

[3] Please note the solid elements have only 3 translational DOFs, so I modified the SP fixity in the condition section.

[yingying]

Dear Prabakaran Kesavan:

Thank you very much for your reply and for debugging this model. May I kindly inquire about a few more questions? What is the effect of setting the Y-axis to "Sum" and "average" on the load. I ran your modified model to get the displacement reaction force curve. The displacement reaction force curve proceeds from the negative direction, whereas the cyclic cycle load applied to the top beam is loaded first with the positive displacement. Can you help me to find the reason for this. Also, what is the basis for setting the values of Kn, Kt and mu for the zeroLengthContactASDimplex element, and what conditions can be based on. Thanks again for your generous and enthusiastic help.

With best wishes,
yingying

[kesavapraba]

Hi, You are welcome. My suggestions are following:

What is the effect of setting the Y-axis to "Sum" and "average" on the load.

Sum option would add up the reaction forces from the SP constraints at the base of the model, whereas Average option would sum up the reaction force and divide it by the number of nodes where it recorded the reaction forces. So, you need the first one to get to see the total applied load (horizontal) on the structural system.

The displacement reaction force curve proceeds from the negative direction, whereas the cyclic cycle load applied to the top beam is loaded first with the positive displacement. Can you help me to find the reason for this

You can change the direction by setting/modifying the -factor in the definition of time series to -1

Also, what is the basis for setting the values of Kn, Kt and mu for the zeroLengthContactASDimplex element, and what conditions can be based on

Dr. Massimo presents a nice explnation about the contact element and associated parameters in the following webinar
https://www.youtube.com/watch?v=tXA5guozrFU

[yingying]

Dear Prabakaran Kesavan:

Thanks a lot for your reply and guidance. I have found the reason now.

With best wishes,
yingying

[kesavapraba]

You are welcome