PDA

View Full Version : Mesh Problems



Erick Scarpone
08-24-2005, 04:25 PM
So I'm trying to get started on the FEA studys of our new chasis (first time doing it), finally got all the loads that we need, how to put and dont put constraints, blah, blah, blah. Spend a week in front of the computer with out goin out to see the sun. And when the model is finally ready and were goin to star the FEA, we have problems with the mesh. y have 17 faces and 54 edges with probleems Cosmos Works tells me to do a failur stats and then a aplication control, i now have to days plaing with the numbers a the aplication controls and at this far i have reduce the faces to 15 and the edges to 41, any tips out there?

Erick Scarpone
08-24-2005, 04:25 PM
So I'm trying to get started on the FEA studys of our new chasis (first time doing it), finally got all the loads that we need, how to put and dont put constraints, blah, blah, blah. Spend a week in front of the computer with out goin out to see the sun. And when the model is finally ready and were goin to star the FEA, we have problems with the mesh. y have 17 faces and 54 edges with probleems Cosmos Works tells me to do a failur stats and then a aplication control, i now have to days plaing with the numbers a the aplication controls and at this far i have reduce the faces to 15 and the edges to 41, any tips out there?

Denny Trimble
08-24-2005, 05:59 PM
Which version of Cosmos are you using?

Are you doing a shell mesh, or a solid mesh?

Can you post any screenshots?

Erick Scarpone
08-25-2005, 07:37 AM
Ok I'm Doing a static study with a solid mesh, using Cosmos works 2004, will post some screenshots during the day....When the computer at one time stops trying doing the mesh
Thanks!!!

Erick Scarpone
09-03-2005, 10:11 AM
I was trying to post some pictures of the FEA we been doing, but it seems that the only way to do it is opening some other post, how do you include pictures in these normal text replay???

So anyway we finally done the mesh all our probes were in tubing boundaries and some brackets boundary with the tubing. We were getting some work done but still are interpreting the numbers we been getting.

So I hope these are my final two questions on this topic.

1.- How do you put the chassis inertia on the model, we were thinking of putting it in every tube, but there are like 30 different tube so its going to take for ever, and it seems like the wrong way of doing it, is there any other way?

2.- Putting constraints for acceleration on the back side of the car only to see how it deforms form the path front to the back, vice versa on braking and braking on curves.

How do you usually put your constraints on these particular cases the numbers rely to much on where the constraints are.

Our explanation on why the constraint only in the back or front is because we want to have the total deformation produced of the loads traveling from front to back or back to front.

This coming from one of the multiple frame definition, but the one were mostly adopting
"It is the structure that holds the front and rear suspension together"

Penny for your thoughts thanks to everyone's help!

Denny Trimble
09-03-2005, 11:08 AM
You're trying to do a solid mesh of your spaceframe? I don't think you can get it done running Windows, because of the 2GB memory limit for applications. And I bet you don't have 2GB in your machine, either http://fsae.com/groupee_common/emoticons/icon_smile.gif

You should look into Algor or Ansys, and model your frame as beam elements. Import your tube centerlines, tell the program what size tubing each line represents, apply loads and constraints, and off you go. Algor solves our frame in 30 seconds, and meshing is instantaneous.

Erick Scarpone
09-03-2005, 05:55 PM
I wont say it was easy but I did the solid mesh on my space frame on Cosmos Works.
The mesh takes like 1 hour to be done but running the studys take like 5 minutes

Horace
05-26-2007, 01:54 PM
Erick,

Would you mind explaining how you did it? As well as your computer specs?

Thanks,
Horace

Steve Yao
05-26-2007, 04:06 PM
Horace, this thread is from fall of 2005.