Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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.

Sliding contact or Roller?

Please login with a confirmed email address before reporting spam

Hi again!

I'm trying to simulate the effect of a load pressing down on an arch-shape (made of composites with cylindrical orthogonal material properties). This arch is resting on a fix rigid plate but the arch is free to slide open when pressed down.

(See ArchSchematic.jpg)

With the load pressing down, the arch will "open" and its extremities will raise. The only contact remaining with the plate being a single point.

Try #1 (See OrthogonalArc8.mph)
I tried to simulate that by creating a contact between the arch and the plate (Contact Pair > Contact) but I end end with a Segregated Solver Steps error ("segregated solver steps do not involve all components").

I tried to replicate some settings of the model example "cylinder_roller_contact.mph" in the model library and this time I end up with a Jacobian error ("Failed to evaluate operator Jacobian").

Try #2:
Instead of working with contacts, I decided to use the ROLLER constraint. But since the roller constraint does not exist for a single point, I had to apply it on a small edge of my model which provoked a stress discontinuity.

(See PointStress,jpg)

I wouldn't mind that discontinuity if I was able to create a subset of the result and remove a couple of mesh cells for visualization purpose so that the scale factor would not be effected by it and I would be able to actually see the stress in the model (not just all blue).


So finally my question is: Is there a way to create a "Point-Roller" constraint OR a way to remove from visualization of stress some area of the model where we know that there are some mathematical artifacts?

Thanks again for all help...I'm slowly going up the "COMSOL learning curve".
:)


3 Replies Last Post 21.02.2011, 17:22 GMT-5
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 19.02.2011, 15:52 GMT-5
Hi

if your "ground" is horizontal, I would say that your "point roller" constraint is essentially prescribed displacement y=0, no restrictions on x (in 2D) for the point. Or what do you believe?

To avoid to display high or low stress values you can always "filter" part of the images out, in v4.1 you have a filter sub-node, and also the rulers for the min-max display values in the plot section.

In 3.5 we ususally added a bolean conditions

(1E5 < smsld.mises) * smsld.mises * (smsld.mises < 1E6)

--
Good luck
Ivar
Hi if your "ground" is horizontal, I would say that your "point roller" constraint is essentially prescribed displacement y=0, no restrictions on x (in 2D) for the point. Or what do you believe? To avoid to display high or low stress values you can always "filter" part of the images out, in v4.1 you have a filter sub-node, and also the rulers for the min-max display values in the plot section. In 3.5 we ususally added a bolean conditions (1E5 < smsld.mises) * smsld.mises * (smsld.mises < 1E6) -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 21.02.2011, 10:29 GMT-5
I actually tried the "Prescribed Displacement" constraint but it doesn't seem to be working well. Does the "Prescribed Displacement" constraint is blocking rotation intrinsically?

Example: If I want a point to slide horizontally but not vertically, I would choose the "Point Prescribed Displacement" and under the prescribed displacement (standard notation), I choose "Prescribed in Y direction" of V0 = 0 m (and I don't choose anything for the U0 field).

Strangely, when I do that, my model do not want to slide, as if it was fixed. But if I use the "Roller" constraint, it will slide.

Any idea why?
I actually tried the "Prescribed Displacement" constraint but it doesn't seem to be working well. Does the "Prescribed Displacement" constraint is blocking rotation intrinsically? Example: If I want a point to slide horizontally but not vertically, I would choose the "Point Prescribed Displacement" and under the prescribed displacement (standard notation), I choose "Prescribed in Y direction" of V0 = 0 m (and I don't choose anything for the U0 field). Strangely, when I do that, my model do not want to slide, as if it was fixed. But if I use the "Roller" constraint, it will slide. Any idea why?

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 21.02.2011, 17:22 GMT-5
Hi

that I do not understand, there must be another constraint somewhere ;), ceck the equations with the preferences Equation view "on", ad scroll to the lowest subnode for the two boundary cases

--
Good luck
Ivar
Hi that I do not understand, there must be another constraint somewhere ;), ceck the equations with the preferences Equation view "on", ad scroll to the lowest subnode for the two boundary cases -- Good luck Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.