Cervenka Consulting Forums — polar coordinate and boundary condition in the radial direction

Re: polar coordinate and boundary condition in the radial direction

null@example.com (Teresa) — Mon, 06 Dec 2010 11:13:05 +0000

Dear Sir,
That's exactly so! How I'd never notice this! I should review the theory of elasticity. Thanks for your help and patience.
Regards,
Teresa

Re: polar coordinate and boundary condition in the radial direction

null@example.com (dpryl) — Mon, 06 Dec 2010 09:03:13 +0000

Dear Teresa, I am not sure if I understand your question - when the X and Y strains are identical, it means a uniform expansion. For a uniform expansion, the values are the same in any coordinate system (and therefore, it does not matter if using the global or some local CS).

Re: polar coordinate and boundary condition in the radial direction

null@example.com (Teresa) — Mon, 06 Dec 2010 04:01:28 +0000

Dear Sir,
I checked the input file about the defined boundary condition, and the initial strain value x and y component for every finite element is the same. Maybe I made a misunderstanding to think the x and y is global. Here the x and y component is local x and local y so that a uniform expansion can be assigned. Is that right?
Regards,
Teresa

Re: polar coordinate and boundary condition in the radial direction

null@example.com (dpryl) — Thu, 02 Dec 2010 09:39:50 +0000

The comon way to model corrosion expasion is to simply apply expansion to the whole steel crossection, see, e.g., the article Peter KOTEÅ , Miroslav BRODÅ‡AN / University of Å½ilina, Civil Engineering Faculty, Department of Structures and Bridges in our 2/2009 newsletter http://www.cervenka.cz/newsletters/news … 09-october
You do not need to prescribe the volume change through temperature, you can directly apply the condition Initial Strain for 2DElem Surface.

If you think a uniform expansion of the whole steel cross section (i.e., a diameter increase) is not good enough in your case, please explain why and we try to help you find a way of modelling.

Re: polar coordinate and boundary condition in the radial direction

null@example.com (Teresa) — Wed, 01 Dec 2010 15:29:45 +0000

Dear Sir,
Thanks a lot for your reply and thorough consideration. I am trying to model the steel corrosion induced cracking of concrete cover. In fact, I am modeling the problem on a single cross section. (I am sorry for the incorrect expression by "cylinder model") I want to apply the radial expansion to the ring outside the rebar (this ring representing the rust) for uniform corrosion. I've tried "the initial strain for surface", it's also a Cartesian coordinate which will result in constant x and y initial strain component for every point of the ring. I just thought of that the expansion effect could be modeled by temperature loading.
Regards,
Teresa

Re: polar coordinate and boundary condition in the radial direction

null@example.com (dpryl) — Mon, 29 Nov 2010 08:26:20 +0000

Hello, you can of course model the cylinder as an axisymmetrical (2D) task.
If you for some reason need a full 3D calculation, I am afraid in the current version, you can only define a pressure load perpendicular to a surface using the local coordinate system in the "Load Force for Surface" condition, but not easily prescribe such a displacement. However, if you tell us more about the problem setup, we might be able to suggest some other way of modelling it?

In the worst case, you have to define a simple distinct displacement load on the surfaces (e.g., 1m in X and 2m in Y), and edit it in the .inp file to introduce the correct functions for the dependence of the load on the position (coordinates).

polar coordinate and boundary condition in the radial direction

null@example.com (Teresa) — Sun, 28 Nov 2010 09:30:14 +0000

Hello, I am trying to apply a radial displacement in my cylinder model by GiD. But it seems that there is not polar coordinate system in the GiD. Do I need to simulate the circle boundary with polygon and apply boundary condition on each side? Or is there any easier way for this problem?
Many thanks!
Regards,
Teresa