Bolted Joint FE Model Analysis using Symmetry 1

[struclearner]

Dears,

I am analyzing a FE Model with Bolted Joint, the Bolts are modeled by beams and connected to the solid elements with Rigid Body element spider, due to the size f the model, I want to use the symmetry of the model. The symmetry plane passes right through the half of one of the bolt. The bolts are modeled by beams of circular cross section. The bolt which is cut by half at the symmetry plane, what cross section of the beam will be used to model that bolt. Is the semi circle cross section will be used for that beam or the regular cross section used for other beams will be used.

Thanks for help.

 

[NRP99]

If you have flanged joint(circular bolted joint), use cyclic symmetry with 20 to 30 degree sector model consisting of 1 bolt.

Otherwise move the cutting plane few degrees further so that it passes through the clamping plates and not the bolts.
Using symmetry at bolt location will complicate the problem since you are using multipoint constraints and symmetry boundary conditions at the same place.

Use half symmetry for the bolt cross section if you are using solid bolts.

 

[struclearner]

NRP99, Thank you very much for your reply and help.

The bolt cut by the symmetry plane could be modeled by solid elements and the remaining bolts could be represented by the Beams and attached to the structure by multipoint constraints. Although the results will be not be 100% in agreement quantitatively by the two different modeling techniques of the bolts, but the model will be more efficient computationally by using the symmetry of the structure.

 

[Denial]

Your quandry, with the centreline of a beam element lying in the plane of symmetry, is often encountered in FE analysis.[ ] Unless "precautions" are taken there is a risk that the element will be "double counted".[ ] The approach I used to take was to model the beam concerned with its FULL section properties, but modelled using a material with HALF the Young's modulus and half the density.[ ] Be aware that this leads to the calculated element actions being half of the correct value, so the values the analysis calculates must be doubled.

A similar approach can be used for plate elements that lie in the plane of symmetry.

 

[Denial]

I suggest you make up a VERY simple little symmetric model.[ ] Model it in full, then analyse it using your chosen symmetry approach.[ ] Compare.[ ] The ten minutes you spend on such an exercise will be very cheap insurance.

 

[NRP99]

Denial

How do you take care of conflicting boundary conditions of MPC and symmetry BC at same place, if beam approach for bolting is used?

 

[Denial]

I'm not going to get involved in the specifics of your problem, and even if I wanted to I don't know what MPC stands for.[ ] But as a general principle if anything conflicts with your symmetry boundary conditions then you don't have symmetry any more.

 

[NRP99]

Denial
MPC stands for multi-point constraint or spider or RBE3/RBE2 elements used to connect beam bolt node to bolt holes or washer area.

My point is you can't use beam approach for bolt when you have symmetry. Symmetry BC and RBE3 will conflict with each other. It is not clear from your previous post explanation that FULL cross section beam and half material properties can give reasonable results. That's why I asked my question.

Could you please explain your point?

 

[Denial]

NRP99.[ ] I am not, and never have been, an ANSYS user.[ ] My comments were made as broad statements applicable to FEM in general.[ ] I am in no position to comment on how planar symmetry might clash with, or be made compatible with, ANSYS-specific modelling elements or devices.

I was answering what I took to be a general question from the OP:[ ] when developing a model of a half of a structure with a plane of symmetry, how do you model a beam that lies in the symmetry plane.