???Using the Split command for FEA preparation - causing a sliver...???

Pioneer
Pioneer

Hi all, 

 

I'm trying to create a selectable surface to constrain when in the FEA environment. I'm using the Split command to create a surface along a protrusion, but i'm getting mesh errors due to slivers that have developed as a result of using the Split command... Can anyone help me in solving this issue? Is there a better way to define a surface to select for FEA? Thanks!!The protrusion..The protrusion..The sliver..The sliver..

"The society which scorns excellence in plumbing because plumbing is a humble activity, and tolerates shoddiness in philosophy because philosophy is an exalted activity, will have neither good plumbing nor good philosophy. Neither its pipes nor its theories will hold water." -- John W. Gardner
11 REPLIES

Re: ???Using the Split command for FEA preparation - causing a sliver...???

Legend
Legend

Hi @ncsteinb,

 

as far as I know, split-command is the way to go if you want to apply a constraint or a force to a certain face of your product. Where exactly are your sliver faces? Are they where your split area merges with the rest of the part?

 

Have you tried to decrease the mesh size? If that doesn't work, try to uncheck the Geometry check by modifying your linear static study like this:

 

- Right-click your study under Simulation-sidebar and select Modify Study

- Click Options >> to see more options

- Change Geometry check from "On" to "Warning only"

Re: ???Using the Split command for FEA preparation - causing a sliver...???

Phenom
Phenom

Hi,

 

Could you share your model?

 

BR,

Imics
http://solidedgest.wordpress.com/

Re: ???Using the Split command for FEA preparation - causing a sliver...???

Pioneer
Pioneer

Hi, unfortunately I cannot share the model as it is a pre-launch design for a customer of ours. 

"The society which scorns excellence in plumbing because plumbing is a humble activity, and tolerates shoddiness in philosophy because philosophy is an exalted activity, will have neither good plumbing nor good philosophy. Neither its pipes nor its theories will hold water." -- John W. Gardner

Re: ???Using the Split command for FEA preparation - causing a sliver...???

Pioneer
Pioneer

@Teemu, The slivers are appearing at the intersection of the part and the reference plane that was used to create the split face. 

 

It's shown in the second picture where the plane intersects the body. 

 

I have tried decreasing the mesh as far as the slider will go, but haven't yet manually adjusted the mesh. I will try unchecking the Geometry Check and see if I can get anywhere... 

 

What's interesting, is that I got the model to mesh once... but i've tried about a dozen other times with no luck...

 

Thanks for the help.

"The society which scorns excellence in plumbing because plumbing is a humble activity, and tolerates shoddiness in philosophy because philosophy is an exalted activity, will have neither good plumbing nor good philosophy. Neither its pipes nor its theories will hold water." -- John W. Gardner

Re: ???Using the Split command for FEA preparation - causing a sliver...???

Pioneer
Pioneer

Are you using FEmap or an exterior FEA software? I use Ansys with no issues with split faces. I've done some very crude stuff in FEmap with split faces with success as well.

 

Can you share a little more on your construction techniques?

Re: ???Using the Split command for FEA preparation - causing a sliver...???

Pioneer
Pioneer

I'm using Femap within Solid Edge. 

 

The model is a football facemask, so the majority of features are Protrusions, profiles swept along a path. 

"The society which scorns excellence in plumbing because plumbing is a humble activity, and tolerates shoddiness in philosophy because philosophy is an exalted activity, will have neither good plumbing nor good philosophy. Neither its pipes nor its theories will hold water." -- John W. Gardner
Highlighted

Re: ???Using the Split command for FEA preparation - causing a sliver...???

Pioneer
Pioneer

Cool - The FEA load cases would be fun to figure out!

 

I assume then you're using the split command in the surfacing environment with planes. You're not trying to split the bodies are you?

 

Using sweeps and curved geometry in space can be tricky to get the parts to actually terminate as intended. You've verified the slivers only show up after the split commands? Just to double check, you put all your splits at the end of the model tree as well?

 

If you're geometry inspects to your satisfaction without the splits then may there's still a better option. If splits do not work for me I use the split faces command. I have limited expierence doing this on curves (usually just the corner rounds on HSS tubing) but it might work for you with a little more effort in the setup.

 

I'd look at my toolbar to see exactly what I use but we're short on licenses at the moment...

Re: ???Using the Split command for FEA preparation - causing a sliver...???

Pioneer
Pioneer

Yea, it's a really cool project!! http://www.zutifacemasks.com/

 

The Split command i'm using is under Modify Surfaces. Is this the correct tool to use? I basically need to create a selectable surface to constrain to represent the mounts to the helmet. I also need to create surfaces to apply loads to, so we can simulate impacts. 

 

The Geometry Inspector shows that there are slivers of 0.000" at only two of the split surfaces, and the other split surfaces are fine... 

 

The Split commands are listed at the end of the tree.

 

Cheers,

Nevin 

 

 

"The society which scorns excellence in plumbing because plumbing is a humble activity, and tolerates shoddiness in philosophy because philosophy is an exalted activity, will have neither good plumbing nor good philosophy. Neither its pipes nor its theories will hold water." -- John W. Gardner

Re: ???Using the Split command for FEA preparation - causing a sliver...???

Pioneer
Pioneer

I use the "Split" and the "Parting Split" commands. I don't immediately see an issue with what you're doing.


My guess having never swept profiles and then attempted FEA is you'd need these options for it to be successful. I could see the other ones causing goofy complexities that might fail.

 

Sweep.PNG