Dear Opensees Community Members
I am analyzing a cantilever bridge column using "dispBeamColumn" element and RC fiber section under cyclic loading. Element 1 is the bottom most element, two nodes 1 and 2 are assigned to element 1, node 1 is fully constrained in all directions (fixed support) and node 2 is connected to to element 2. Element 2 is next to element 1, two nodes 2 and 3 are assigned to element 2. I separately recorded the stresses and strains at the common node i.e. node 2 of element 1 and element 2 using element recorder. The values of are stresses and strains recorded at the common node, i.e. node 2 of element 1 and element 2 are not same. Generally, In the finite element analysis, the stresses and strains at a common node between two elements should be same, but in my analysis they are different.
Is there any reason for that?.
I want to record stresses and strains in the element at nodes, but I think the element recorder is recording at the Integration points (Gauss points). What I should do to record stresses and strains in the element at nodes.
1. Are there any Integration points at nodes?
2. If no, How can I record stress and strains in the element at nodes
3. Is there a way to put integration points at nodes?
4. What can I do to record stresses and strains in the elements at nodes?
Thanks...
Disp Beam Column elemnt - Element strains at nodes
Moderators: silvia, selimgunay, Moderators
-
polimeruvijay
- Posts: 43
- Joined: Fri Jan 29, 2016 10:20 pm
Re: Disp Beam Column elemnt - Element strains at nodes
this is one of the answers is given previously for the subject you mentioned.
Re: stresses all over the element
Postby fmk » Fri Dec 13, 2013 10:30 am
you could apply stress extrapolation techniques to look at stresses at a number of points in the element. typically these techniques are used to extrapolate the stresses from the gauss points to the nodes for plotting stress contours. the resulting stresses are typically determined using interpolation functions and are thus approximations, i.e. not exact if the stresses at the gauss points were exact to begin with.
the question is do you need to do this. you usually make the mesh fine enough to capture results of interest. in your cases if stresses are of interest, the results at the gauss points should be fine. when you factor in the unknowns in the material, loads and element approximations, they are more than fine!
Re: stresses all over the element
Postby fmk » Fri Dec 13, 2013 10:30 am
you could apply stress extrapolation techniques to look at stresses at a number of points in the element. typically these techniques are used to extrapolate the stresses from the gauss points to the nodes for plotting stress contours. the resulting stresses are typically determined using interpolation functions and are thus approximations, i.e. not exact if the stresses at the gauss points were exact to begin with.
the question is do you need to do this. you usually make the mesh fine enough to capture results of interest. in your cases if stresses are of interest, the results at the gauss points should be fine. when you factor in the unknowns in the material, loads and element approximations, they are more than fine!