Brace - Pin Connection

Forum for OpenSees users to post questions, comments, etc. on the use of the OpenSees interpreter, OpenSees.exe

Moderators: silvia, selimgunay, Moderators

Post Reply
yigitozcelik
Posts: 15
Joined: Tue Sep 16, 2008 8:32 am

Brace - Pin Connection

Post by yigitozcelik »

Hi;

I want to model a pin for brace connection in a 2D frame
As far as I know I need to use
Zero-Length Element with a low stiffness elastic material
and equalDOF constraint
My question is that how i can define the orientation of the local axis of the brace ?
I am confused with the explanation in the user manual

Thank you in advance
silvia
Posts: 3909
Joined: Tue Jan 11, 2005 7:44 am
Location: Degenkolb Engineers
Contact:

Post by silvia »

why not just use a truss element?

for a truss, the orientation doesn't matter anyways, but in 2D the local z axis is parallel to the global Z axis.

No, do not use a zero-length element with low stiffness, that messes up your stiffness matrix, just leave the rotational dof free. again, i'd use a truss element instead.
Silvia Mazzoni, PhD
Structural Consultant
Degenkolb Engineers
235 Montgomery Street, Suite 500
San Francisco, CA. 94104
yigitozcelik
Posts: 15
Joined: Tue Sep 16, 2008 8:32 am

Post by yigitozcelik »

According to the paper
"Model for Cyclic Inelastic Buckling of Steel Braces"
Patxi Uriz; Filip C. Filippou; and Stephen A. Mahin
to represent buckling behaviour, I am using nonlinear beam-column element for braces

so I can't use truss element for braces

I used infinite small length elastic member with very large area and very small inertia to connect the braces to the joint where column and beam intersect to model pinned connection,
but it also did not work properly.

Do you have any suggestions for my problem, Dr Mazzoni?

Thank you in advance
yigitozcelik
Posts: 15
Joined: Tue Sep 16, 2008 8:32 am

Post by yigitozcelik »

I figured it out.

Thank you for your precious help, Dr Mazzoni
silvia
Posts: 3909
Joined: Tue Jan 11, 2005 7:44 am
Location: Degenkolb Engineers
Contact:

Post by silvia »

oh, yes, for buckling you do use flexural elements.
Silvia Mazzoni, PhD
Structural Consultant
Degenkolb Engineers
235 Montgomery Street, Suite 500
San Francisco, CA. 94104
Zephyre
Posts: 10
Joined: Tue Sep 22, 2009 10:26 am
Location: UMD

Post by Zephyre »

Could you share how you solved the problem. I'm also using nonlinear beam column element and have problem to model buckling for thebrace.
thanks
Thanks!
silvia
Posts: 3909
Joined: Tue Jan 11, 2005 7:44 am
Location: Degenkolb Engineers
Contact:

Post by silvia »

you need to do some research on this topic, both in this forum and in the literature, as I have not done it myself.
Silvia Mazzoni, PhD
Structural Consultant
Degenkolb Engineers
235 Montgomery Street, Suite 500
San Francisco, CA. 94104
yigitozcelik
Posts: 15
Joined: Tue Sep 16, 2008 8:32 am

Post by yigitozcelik »

Hi Zephyre;

You need to use equalDOF command for pin connection.
First assign slave nodes to where braces are connected to other members
so master and slave nodes have same coordinates
then use equalDOF command to achieve same DOFs in x and y direction between master nodes and slave nodes (not in z ).

You need to use zerolength elements also if you need to add rotational springs between master and slave nodes.

For buckling behaviour;
you need to divide brace into at least two sub elements
and assign a initial imperfection at the mid-length of brace in direction normal to brace direction.

Please refer to
"Model for Cyclic Inelastic Buckling of Steel Braces"
Patxi Uriz; Filip C. Filippou; and Stephen A. Mahin
for further information

Finally, please notice that slave and master nodes are not same nodes.
For example if you need to calculate base shear; you need to add the contribution of slave nodes at the base level. Let s say you have a master node with tag 1 at coordinate (0,0) and a slave node with tag 2 at same coordinate, for base shear at coordinate (0,0); you need to sum the forces of both nodes.
Vb_coordinate(0,0) = F_node1 + F_node2

That is all I got
I hope it will be helpful
yigitozcelik
Posts: 15
Joined: Tue Sep 16, 2008 8:32 am

Post by yigitozcelik »

I forgot to say
For braces, you need to use corrotational transformation and use fiber section to account for moment axial load interaction.
yalham
Posts: 44
Joined: Sun Jan 06, 2008 6:37 am

Post by yalham »

Hi all

Modeling of an individual brace element (simulating buckling) with pin, can be performed using 3 nodes, simply define hinge support at end supports.

If you want to model a braced frame, and braces have pin connection, the best way is to use zeroLength-element witch restrained beam-column intersection node and brace end node in dof 1 & 2 only, a command like this:

element zeroLength $eleID $Node1ID $Node2ID -mat $matID $matID -dir 1 2

and $matID refers to a pre-defined high-stiffness elastic material.

The above command is sufficient. it connect brace end node in translational dof. “equalDOF” command only coordinate displacement of nodes and is not suitable for brace connection, here for braced frames, loads from brace must be transfer, otherwise you will have wrong response.

Good luck
Post Reply