*INITIAL CONDITIONS, TYPE=STRESS, SECTION POINTS

[Launeben]

Hi everyone,

I am doing research on stability analysis of welded I-profiles under compression. Therefore im doing analysis in ABAQUS. I have modeled a simply supported beam under compression with linear elastic-ideal plastic material behaviour. As elements I use simple beam elements (B33, Euler-BERNOULLI type, cubic formulation). As procedure I use the STATIC, RIKS procedure (NLGEOM=ON). As imperfections I already put the geometric imperfection (L/1000) by scaling the first eigenshape from a previous eigenvalue buckling analysis. Now I wanted to put residual stresses (longitudinal stresses from welding) as structural imperfections. I have a added a picture of the stress distribution i want to apply. Therefore I use the command lines:

*INITIAL CONDITIONS, TYPE=STRESS, SECTION POINTS
first line: <Element set number>, <Section point number>, <value of the first stress component>, 0, 0

I repeat this line for defining a stress in all section points (which are shown in the beam section library, 13 for I shaped beams). In order to define stresses at more points (than the 13 section points given as default value) I changed the number of section points before by the command lines:

*BEAM SECTION, ELSET=<>, MATERIAL=<>, SECTION=I
third line: <number of integration points in the lower flange>, <number in the web>, <number in the upper flange>

However after running the calculation I always get an error. Is it possible to use command *INITIAL CONDITIONS, TYPE=STRESS, SECTION POINTS for beam elements? If so what I am doing wrong? Please help me. I already try since days, but without success. And if it is not possible with this command how else I can do that?

Best regards, I hope somebody can help me or replies
Benny

 

[burningDiesel]

There is also an Abaqus subsection where your question might be spotted sooner.
to answer your question: the ", SECTION POINTS" attribute is for shells (probably that is the error you are getting?). Although I agree it should also be available for beams (but afaik in ABQ, it isn't).
What I used was the SIGINI subroutine, which is very easy to use.

 

[Launeben]

Thanks for your answer very much

So as I understand for beams its only possible to put a residual stress or initial stress state with such a subroutine?

In that case I will try this.

Best regards, Benny