The OpenSees Community

Hello, does anyone know how/where to find the documentation for an older version of OpenSees? The online documentation is about the latest version (3.4.0.7 as of today), but the version I'm using is older than that. Althogh I can read the change log to find what the differences are between the two versions, it'll be much easier if I can just read the documentation that is corresponding to the version I'm using.

I'd appreacite any suggestions or ideas. Thank you!

I was wondering if there is an OpenSeesPy command for creating the MPCO recorder? I couldn't find it on the OpenSeesPy wiki page but was hoping the documentation might lag implementation. I have created my model and analysis using OpenSeesPy. I wanted to use MPCO to output a database for post-processing. Can someone point me to the right direction? Thanks.

mhscott wrote: ↑Sat Dec 17, 2022 5:07 am Why do you think Mz and My are 5 and 6?

https://portwooddigital.com/2022/11/04/ ... antilever/

Thank you very much for yuor reply and teaching me/us 'how to fish', professor. Just wanted to provide updates from my side in case that other people may find helpful. I figured this out by building a simple model with simple load following the post.

1.The numbering of the dof for the 'sectionForce' command is 1,2,3,4 corresponding to P, Mz, My and T respectively. If you put 5 or 6 which is not expected by the command, it will return all 4 forces (a bug? or maybe a trick?). This explains what I got in the screenshot. See this link for what dof are valid for your section: https://openseespydoc.readthedocs.io/en ... ction.html

2.This means that OpenSees doesn't always respect the traditional/typical numbering of dof (i.e., 3D, 6dof: Fx Fy Fz Mx My Mz). (Not sure if this is a good thing though...). Anyway, a safer way to figure this out is to use Prof. Scott's suggestion in the post he shared.

Thank you again.

I think there might be a bug in the sectionForce command. For dof=6 and dof=5, I'd expect to have Mz and My respectively. But it returns all 4 section forces. Please see the attached picture. Did anyone have the same problems before? Please correct me if I'm worng. Thank you.

https://www.dropbox.com/s/64hs83u61uz67 ... g.PNG?dl=0

Hi there,

I am using the Get_Rendering commands to make plot, and I am running analysis in Jupyter Notebook. According to the tutorial, I should get interactive plots that allow zoom, pan or download. The zoom and pan features are fine. However, if I click on the download button below the plot, the new tab is just blank.

I am just a beginner of Python and OpenSeesPy. By any chance, does anyone have insights on this issue?

Thanks in advance!

selimgunay wrote: ↑Mon Nov 16, 2020 11:40 pm If you are using a zerolength element, it means that you are introducing a flexible element (can be interpreted as a spring) between the two nodes of a zerolength element. Therefore, obviously those two nodes will have different values in the eigenvectors

Thank you for your explanation. It's so clear to me now!

Hi all, I've been thinking about two questions about zero-length element but couldn't find an answer myself. By any chance, could you show me some your insights? Thanks in advance!

Q1: The zero-length element consists of two nodes (m dof at each node) at the same location, which seems to introduce one redundant node (m dof) to the model. Consequently, I think in eigen analysis, the size of the eigenvector will increase by m. And the values of the two nodes in the same eigen vector should be the same. If I plan to use the eigen vectors to do modal analysis, should I eliminate the repetitive m entries from the eigen vectors?

Q2: If I apply a nodal load on one of the two nodes of a zero-length element, does this mean this load is also applied on the other node? (In other words, will it make any difference if applying nodal load on either node of a zero-length element?)

You ended up with a psedo time =2.0 for both loads applied. To correct this, you should make the first load constant and reset time to zero before applying the second load so that the second analysis will start from time 0. Hope this helps!