Partitioning 1

[ysg519]

I am analyzing a part after importing it as a .stp file. I have created seed points along edges successfully.When I try to mesh I get a message saying that areas in magenta cannot be meshed and should be partitioned. I have tried to partition it using options such as 3 point plane and normal to edge. But the problem is that it is getting too tedious. Isn't there an easy way to mesh parts without paritioning them?
Thanks

 

[corus]

You can just use tet elements is you can't be bothered or if the geometry is just too difficult. It is better to try and get structured or swept regions though for more regular meshes by partitioning the part. There is an art to partitioning the part and if you can try and use the
least number of partitions. Using planes isn't always the best way to partition though.

corus

 

[brep]

What mesh technique are you trying to use?

 

[ysg519]

Thanks corus the tet elements worked! Thanks brep.

 

[rstupplebeen]

If tet meshing is ok for this application check out virtual topology. It will replace multiple surfaces with 1 approximate surface. This allows the mesher to use elements with better aspect ratios in some circumstances. Also, as a rule of thumb tet10s are more accurate for the time costs associated with the additional DOFs. Hope this helps.

Rob Stupplebeen

 

[brep]

+1 for Mt Stupplebeen's virtual topology suggestion. And we'll forgive for his TET10 slip, I think he meant C3D10M

 

[ysg519]

Thanks rstupplebeen and brep. And Mr.Stupplebeen will forgive brep for the "Mt" I think he meant "Mr"

 

[rstupplebeen]

brep,

Point taken. Switching between too many codes recently and talking to those not in the know.

ysg519,

You are welcome. I think you are right since I am not a mountain of a man.

 

[ysg519]

I have posted this question as a separate thread but didnt get any response. I was wondering if you could take a look and please give me some input. I am repeating it here for convenience.

THERMAL CONTACT PROPERTIES
I am defining contact properties for conducting surfaces. In the boxes where I am required to mention the conductance at zero clearance, what goes in there is it the thermal conductivity of the slave or the master surface. Also what goes in the clearance box for zero conductivity? In my case the surfaces are in full 100% contact with no clearance. Thanks

 

[corus]

If you have 100% contact with no clearance then there is no need to use contact surfaces across a non-existing gap. Just tie the two surfaces together.

corus