link180 ansys workbench

[nicnoc]

Hello, everyone. I have a model in ansys workbench. In practice it is a wired cover. I created beam elements to simulate cables. I'm going to do the analysis and everything goes well.
However some cables are compressed.
to simulate the correct functioning of the cables I created the following snippet command, applied in the geometry environment:

et,767,link180,1,1
sectype,767,link
secreted, 314.159

....
....
....


et,796,link180,,1,1
sectype,796,link
secreted, 314.159

mp,ex,5,2.01e6
mp,dens,5,7850
mat,5
767,768,769,770,771,772,773,774,775,776,777,778,779,780,782,783,784,785,786,787,788,789,790,791,792,794,795,796
(I apply the material to elements 767,768,...,796)


the analysis however does not converge and a error message appears to me in which it is said:
"
an internal solution magnitude limit was exceeded
"
and then
"
solver pivot warnings has been encountered
"

 

[nicnoc]

solved.
then for the snippet I think it is enough:
et,matid,link180,1,1
sectype,796,link
secreted, 314.159

the material is automatically assigned (take the one assigned to the element).
However, it is necessary to tell workbench to create a single mesh element for beams that will be transformed into link180.

so performs the analysis and looks correct. .
hello to all and good day

 

[nicnoc]

I didn't actually solve it. because the good ansys continues to consider me is the element link180 but reagent both traction and compression. . .
I exported to ansys and saw that I have all links180 reagents both traction and compression

 

[nicnoc]

I think it's a convergence problem.
I made a simple model to check that the process and results were correct.
in the linear field react both to c and to traction. while with nlgeom,on I get the exact result.
Now the problem is that having to go to the field I miss the dear overlap of the effects, but this is another story.
first I have to see if at least the simple load conditions converge

 

[stefano.garbin]

et,matid,link180,1,1

That's weird. the command you indicated seems correct!

et, number, name_element, keypot1, keypot2, keyopt3, keyopt...

If you try to write
et,matid,link180,,1

 

[nicnoc]

I have rechecked, everything comes back, the binding reactions that are not null in some points are due to the proper weight of the cables (back with the hand plug), I applied acceleration g.
in other load conditions the solution does not converge me. I'll try to increase substeps.
Thanks, you're a light for me! :

 

[nicnoc]

to simulate the tension-/compression-only options, a nonlinear iterative solution approach is necessary.

vuole dire che nlgeom deve essere acceso?