Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Boundary Systems and large deformation / String of Yarn (solved)
Posted 28.04.2021, 05:54 GMT-4 Structural & Acoustics, General, Structural Mechanics Version 5.6 4 Replies
Please login with a confirmed email address before reporting spam
I got a time dependent study where I model a string of yarn and how it bends around a complex solid body. To implement the absence of bending stiffness of the yarn I use the membrane-physics. I works really good.
Now I want to set up a boundary load that is oriented along the surface of the yarn/membrane. If I use a boundary load on the membrane in combination with the automatically set up boundary system I can set up a force that follows the curvature of the yarn. So far so good.
While solving the time dependent study the yarn deforms as it is bending around the counterbody. When I check the orientation of the boundary system (plot the coordinate system on the boundary of the membrane) I can see that the boundary system doesnt follow the deformation of the membrane. The boundary load is no longer oriented along the yarn by the end of the study.
Motivation behind the boundary load: the membrane-solid contact node doesn't allow sliding induced friction (there is a node for that for solid-solid contacts that came with 5.6 but not for contacts with a membrane involved). So i want to set up a boundary load that is tangential to the surface of the membrane and the magnitude should be the local contact pressure multiplicated with the friction coefficient.
How can I solve this?
(In the screenshots attached you can see how the boundary system tends to keep it's original orientation under deformation of the membrane)
Please login with a confirmed email address before reporting spam
The boundary system can have either fixed directions, or directions that follow the deformation. This is determined by the Frame setting in the Boundary System node.
This, of course, assumes that the analysis is geometrically nonlinear, so that there is an actual difference between the material frame and the spatial frame. Also, the deformation scale for the plots should be 1 both for the membrane itself, and for the coordinate system plot.
-------------------Henrik Sönnerlind
COMSOL
Please login with a confirmed email address before reporting spam
Thanks for the reply.
The scale of the deformation is set to 1 in the plot (for the displacement of the membrane and the coordinate system).
The frame setting is on "deformed configuration" (default setting). When I change it to 'geometry' or 'reference' the boundary system definitely keeps it original orientation and does not adjust with the deformation. I tried that after you mentioned it.
With the frame setting on "deformed configuaration" the boundary system follows the deformation at first but at some point it diverges as you can see in the animated gif.
Attachments:
Please login with a confirmed email address before reporting spam
I solved it:
In the settings of the Boundary System Node there is a dropdown menu "create first tangent direction from". It is set on "global cartesian (spartial) / (x,y,z)" by default. When i set it to "material / (X,Y,Z)" the boundary system follows the deformation perfectly.
which makes sense when I look at the pictures in the doc: https://doc.comsol.com/5.5/doc/com.comsol.help.comsol/comsol_ref_deformedmeshes.25.05.html
Thanks again for the hints!
Attachments:
Please login with a confirmed email address before reporting spam
It looks like it is related to the setting for 'Create first tangent from'. The only orientation that is uniquely defined is the normal vector (the blue arrow). It can be a bit tricky to get the in-plane orientations that you really want for a general scenario with large rotations. Note that the actual orientation of the first tangent is the projection of the selected direction on the surface.
-------------------Henrik Sönnerlind
COMSOL