SAG1917
Aerospace
- Apr 17, 2012
- 1
Dear all,
I am currently working with a model of 3 concentric 2D rings, of varying widths. Essentially, something like the x-section of a circular pipe whose wall is made of 3 layers.
I am cooling the outer ring, for which I have defined a thermal expansion coef. and thermal conductivity (as well as E, Poisson's ratio and density). The inner rings are insulated. The outer layer is MUCH stiffer than the inner layers, which have slightly different properties.
All layers are modeled as separate parts, assembled and then the Tie constraint is used to define their interaction.
The assembly is fixed at one node (no U1 nor U2) and U1 is stopped at another node (to impeded rigid body rotation). The temperature drop is set as a boundary condition in the analysis step.
I use a steady-state, coupled temperature-disp. analysis.
All elems are quad. Those in the inner layers are plane strain elems, whereas those in the outer layer are coupled temp.-displacement elems.
The job runs but the results show that despite the entire assembly contracting (which is intended), only the outermost (cooled) layer develops any internal stresses, the two internal layers remaining unstressed, which appears wrong, as they too are being forced to shrink (by the outer layer). Does anyone have any idea what I may be doing wrong and how to correct it such that those internal layers develop stress as they are forced to contract?
Thank you all for your time.
I am currently working with a model of 3 concentric 2D rings, of varying widths. Essentially, something like the x-section of a circular pipe whose wall is made of 3 layers.
I am cooling the outer ring, for which I have defined a thermal expansion coef. and thermal conductivity (as well as E, Poisson's ratio and density). The inner rings are insulated. The outer layer is MUCH stiffer than the inner layers, which have slightly different properties.
All layers are modeled as separate parts, assembled and then the Tie constraint is used to define their interaction.
The assembly is fixed at one node (no U1 nor U2) and U1 is stopped at another node (to impeded rigid body rotation). The temperature drop is set as a boundary condition in the analysis step.
I use a steady-state, coupled temperature-disp. analysis.
All elems are quad. Those in the inner layers are plane strain elems, whereas those in the outer layer are coupled temp.-displacement elems.
The job runs but the results show that despite the entire assembly contracting (which is intended), only the outermost (cooled) layer develops any internal stresses, the two internal layers remaining unstressed, which appears wrong, as they too are being forced to shrink (by the outer layer). Does anyone have any idea what I may be doing wrong and how to correct it such that those internal layers develop stress as they are forced to contract?
Thank you all for your time.