GLOBAL NO and Transform

General Questions
1

Hi,

I don’t understand why stresses with GLOBAL=NO do not deliver S12 all the same in this model. Each cube has its own coordinate system.
I’m looking at the rotational invariance when applying a Transform card.

Thanks

Rotational invariance.pmx (131.6 KB)

EDITED: By the way, it shows that aparently graphical issue.

2

GLOBAL=NO for stresses makes sense only when *ORIENTATION is applied. *TRANSFORM doesn’t change the elemental outputs.

3

Thanks Jakub,

This note on the manual is puzzling me.

" With the parameter GLOBAL you tell the program whether you would like
the results in the global rectangular coordinate system or in the local nodal
system. If an *TRANSFORM card is applied to the node at stake, this card
defines the local system. If no *TRANSFORM card is applied to the element,
the local system coincides with the global rectangular system. Default value for
the GLOBAL parameter is GLOBAL=YES, which means that the results are
stored in the global system. If you prefer the results in the local system, specify
GLOBAL=NO."

It may suggest that Transform card by itself makes unnecessary the definition of “orientations” to define the local coordinate system.

Note it makes a difference if one request GLOBAL=NO for the displacements.

1 Like
4

This quote is from *NODE FILE description because nodal results (such as reaction forces) can indeed depend on the transformation. But for elemental results (*EL FILE) you need *ORIENTATION. So:

  • *TRANSFORM → *NODE FILE, GLOBAL=NO
  • *ORIENTATION → *EL FILE, GLOBAL=NO
1 Like
5

I see, I will continue from that point.

Shouldn’t the GUI always be in Global coordinates for displacements.?

Imagine there is a Cylindrical coordinate system. First displacement component becomes the radial displacement. If one request GLOBAL=NO the screen is ploting the radial displacement value U1 into the X direction, representing incorrectly the model deformation as it happens in this case.

6

I have add the Orientation card and now S12 is the same for all the elements but for some reason the displacement colour map is not the same. That seems more a graphical issue.

imagen
imagen349×638 16.6 KB

7

That is indeed a graphical issue. The problem is that a rectangle is split into 2 triangles. And the result is different if the split is done on the other diagonal. I have been looking for a better solution for a long time but haven’t found it.

8

The default is Global = Yes. However, there are cases where a local coordinate system output is needed.

9

More clear to me ,Thank you Matej