Learning Forum

[xiao]

Hello all,nI have build a simple one-span frame. I want to model the horizontal element with link 180. But I get an error i don't understand. I searched the internet for a whole afternoon. Can anyone help solve my problem. nThe link contains my model. I cannot attache ansys file here.nhttps://www.dropbox.com/s/9lrtasaey9fw6dt/theoretical_co_ca_co.wbpj?dl=0nThe error message isnAn edge on a line body with its type set to Link can contain only one element. Consider using mesh controls to generate an appropriate mesh for these bodies.nn

[peteroznewman]

The number of elements along the line body must be exactly 1.nYou have a mesh with more than 1. That doesn't work.nUse a Sizing Mesh Control on the Line body and set the number of elements to 1.nAlso, you put the .wbpj file in your dropbox. That file is useless without the _files folder of the same name.n

[xiao]

Thanks Peter. nI tried what you have suggested. The model still cannot run.nFrom the solver output, the element in the horizontal is successfully set to LINk180. nI'm thinking of the geometry is not correct, so I tried to re-define the horizontal element to be beam. The model runs.nThe error messages are:nSovler pivot warnnings or errors have been encounterred during the solutionnThe selective body meshing is not being recorded,so the meshing may not be persistent on an updatenThe link contains my model.nI really don't how to deal with this error. The theory should be simple. Just replace the horizontal element from beam to link. nI have tested my code in SAP2000, MATLAB. But the ANSYS model just cannot run.n

[peteroznewman]

the link contains only the .wbpj file. This file alone is useless without the _files folder of the same name.nPlease use the File, Archive menu selection to create a .wbpz file, which combines the folder and file into a single Project Archive.nYou can put that file in your dropbox.n

[xiao]

Thanks peter.nThe link below contains the necessary docs.nhttps://www.dropbox.com/sh/icq8bxcqsbe6f29/AAD_-Oi0d212qccA6zJb1M75a?dl=0n

[peteroznewman]

,you didn't fully understand the concept of a single element link.nThe geometry you are meshing has 10 line bodies chained in series. This causes great difficulty for the solver since there is no stiffness between any links. nPlease read these posts.n/forum/forums/topic/cable-driven-parallel-robot-modal-analysis/n/forum/forums/topic/modeling-of-a-prestressed-concrete-sleeper/n/forum/forums/topic/transmission-line-simulation/n/forum/forums/topic/modeling-of-cable-in-a-bridge/nn

[xiao]

Thanks Peter. I will read the post.n

[xiao]

Hello Peter,nI read the posts and modified my model. Now it works. But there is a question about the solution. I campared the solution with my own code and SAP2000. The displacement in the Z direction doesn't make sense. In my code and sap, the displacement in z direction under the applied load is -1.62E-07 m, however, in ANSYS it is -2.6552e-002. This displacement is in deformation probe 3(z direction).nCan you help me solve this problem? Do I need to modify something?nThe attached file is the model.nBest,nXiaonn

[peteroznewman]

Hello Xiao,nLet's look at the big picture here, the top of the left column is moving 0.363 m in the X direction. Why are you focused on the 0.026 m of motion in the Z direction? nANSYS beams implement Timoshenko beam equations. Is that what your code used, or did you use Euler beam equations? ANSYS is performing iterations to take account of deformed geometry. Does your code update the stiffness matrix multiple times as the nodes move?nIf you were going to compare output of different codes, why not start with small displacements of the tip of a cantilever beam, which is easily compared with the analytical result?n