Joined: 03 Jun 2015
|Posted: Mon Jun 22, 2015 2:37 am
I’m doing Master of Science Thesis about progressive collapse and I’m doing FEM calculations of catenary action. Unfortunately I’m having problems in FEM calculations.
I’m solving this kind of problem at the moment:
So I’m having column loss at the same point where my nodal load P is located. The length of the beam’s are 8 m and their cross sectional area is just 2600 mm2 (size of the beam's connection plates are 26mmx100mm). I’m using following non-linear material properties:
Young’s Modulus: 210 GPa
Poisson’s Ratio: 0,3
Yield strength: 355 MPa
Tangent Modulus: 21 MPa
With analytical calculations, solution for deflection is 3333 mm (Tangent Modulus has been 0 MPa)
For FEM calculations I’m using Ansys WB release 16.0. I have tried beam (BEAM189), shell (SHELL181) and solid (SOLID186) elements and I have got unconverged solutions. I also tried to use LINK180 element with APDL command but the program wasn't able to calculate solution. I have also tried to give bit initial deflection but it didn’t help. If I’m using program controlled weak springs I’m getting solution for my problem and the deflection is 3182 mm, but in this case I’m getting vertical Force Reaction ~112 kN upwards for weak springs. Value of the vertical Force Reaction of weak springs is too large so the solution is too inaccurate, or is Ansys automatically eliminating this error? What should I do to get converged solution? What would be best element type use?
I would appreciate very much if I could get some help for my problem.
When I’m getting my model working I will turn it bit more complex, where I’m also taking slabs in account.
Student of Tampere University of Technology, Finland