Problem with creating new faces on a body

[orphefs]

Hi everyone,

I have been trying to create new faces on an existing body, see image below:



Uploaded with ImageShack.us

The red line indicates where I want the new 2 faces to be. The reason I need two faces in the cross section is because of a requirement of an electromagnetic FEM software; particularly, it requires the user to define the planes into which current flows and out of. Thats why it needs 2 faces that are coplanar to each other.

I have been trying to find my way through this, looked at tutorials, fiddled with it myself, but still cant get it. If anyone out there knows how to tackle this, please let me know! I would be very very grateful!

Kind Regards,
Orpheus

 

[Kevin De Smet]

I don't think it is possible to do that and maintain a solid, you would need to make a tiny rectangular cut along the top, this could be what you want for the simulation.

I have kept it one continuous solid but made two surfaces, when extruding a surface you can make just an infinitely thin piece of geometry. I used the move/copy bodies to rotate copy another surface to ensure the way the surface is pointing is opposite, should that be of concern.


2010-07-21_125523_body_faces.zip

 

[orphefs]

Hi Kevin,

Thanks so much for your help! However, I have tried this and it does not seem to work...I think my best bet is to cut the coil along the axis of symmetry and then mirror it, so that I get two separate entities...Hopefully that should work...Thanks again,

Kind Regards,

Orpheus

 

[jimshaw]

Kevin,


Spot on. I would have done it the same way. However, you bring up a great point...


Orpheus - Which FEA tool are you using? My experience with Ansys, for example, is that in this modeling manner, the surfaces are not associated to the volume, so when you mesh, the elements will not start/stop at the surfaces. It is literally just two surfaces floating inside of a solid. Withoutcontinuity between the boundary condition faces and the mesh elements, the program will not be able to properly calculate the voltage direction...


If this tru with your setup, perhaps you can create two half-cylinders, then put your boundary conditions on the upper co-existing planes, and use contact elements on the lower planes to imply continuity. Like this:


View attachment 5827


In SW, I mirrored the body, and unchecked "merge solids", this allows there to be two separate solid bodies.


Does this make sense?


Jim

 

[jimshaw]

Orf - Great minds think alike...


Jim

 

[orphefs]

Hi Jim,

I am using Infolytica MagNet v6 to simulate magnetic fields. It seems like the method you proposed as well is the one with the most chances of success...I am still trying to figure out how to assign the coil parameters in MagNet, but that's a different problem altogether!

However, thanks so much for your quick replies and willingness to help, this is after all a great forum! Very impressed!

Kind Regards,
Orpheus

 

[Kevin De Smet]

The two halves do sound good, keep us up to date