## General Mechanical

Topics relate to Mechanical Enterprise, Motion, Additive Print and more

#### Parts Acting Separately in Modal Analysis

• ZaferYilmaz
Subscriber

Hello everyone,

I am having difficulties in the modal analysis of simple structure(1- story building). I have imported geometry from autocad in which walls are separated parts. I set the materials and default mesh. I encountered some errors and warnings on the way but at the end i obtained frequencies and mode shapes. Even though frequencies are logical. Mode shapes are parts of building moving in absurd ways.( I am trying to obtain modes where all the structure moves in one way or bending).  My model is as follows.

First two modes are below.

1.mod

2. mod

I went up to 100 modes but none of them was whole structure's mode. I analyzed for simple brick wall and got proper mode shapes . As shown below.

Then, I thought it was something with contacts and i tried forming all structure as parts in design modeler and setting share topology but it did not change anything.

One of the warnings i faced was 'at least one body has been found to have only 1 element in at least 2 directions'. But i managed to fix it by changing Geometry>element control>manual. Thought it would be helpful to share this because it was hard to find the solution.

Regards,

Zafer Y?lmaz

• peteroznewman
Subscriber

Hello Zafer,

The analysis is correctly showing that the first natural frequency is the chimney. The problem is that your building model is very stiff. Take a flat sheet of cardboard and it is easy to flex.  Fold that cardboard into a box shape and tape all the joints together, now the box is extremely stiff. That is why your one wall can flex, but your building can't, because it is a box.

To see some lateral vibration in a building model, make four slender steel columns and put a frame around the tops of the columns, then put a massive concrete floor at the top of the columns. Now you will see some lateral modes. You can copy that geometry to make a second storey and now the modes get more interesting.

• ZaferYilmaz
Subscriber

Dear Peter,

Thank you for your answer. I was working on this for a long time, and i wanted to get professional help quickly.

I did what you proposed and yes I got the modes that is expected for structures. Such as this horizontal movement.

For my case, I cannot make any changes on it. What would you propose for stiff structures like mine? I need to get modes such as horizontal, longitudinal or bending etc. in which whole structure moves.

• peteroznewman
Subscriber

What do you mean you can't make any changes?

Maybe you can't change the outside shape of the building, but can you do any of the following...

• Change the wall thickness

• Change the wall density

• Change the wall Young's Modulus

• Change the connection between adjacent walls

• Change the boundary condition to the ground.

When I say walls above, I mean walls and roof.

• ZaferYilmaz
Subscriber
It's normally 2 story building and I started with one story of it. I cannot make any changes because it's the modal analysis of a real building and I am trying to get the modes of that structure by using workbench. Normally mechanical Apdl is used for this type of situations. I faced some problems in apdl and decided to use workbench. I wanted to learn if there is a way to make modal analysis of the building in workbench without making any changes in geometry.
Regards
Zafer Y?lmaz
Subscriber

Hello zefer,

If you have modeled all the contacts correctly, then what ansys work bench is showing is how your building behaves dynamically.

how are you modelling the ground? if your are simply fixing the base, then may be try modelling the ground stiffness with spring elements

i would suggest modeling both the floors, then you may see the the primary global bending mode you are looking for, but as peter said, with just one floor your structure is too stiff.