FEA on chassis Help!

[mnmhalstead]

Hi All,
Before I start I would like to say that I a novice in FEA and would like to learn as much as possible through this forum.
So hence any useful comments would be much appreciated.
Software Siemens UG NX NASTRAN
I would like to run a Static analysis on a chassis successfully and this is what I tried to accomplish.
I would like someone to validate my thoughts or point me in the right direction.

Scope of the analysis is to find out if the chassis is strong enough to withstand the following loads and highlight weak areas.
Yield Stress of the material is : 450 Mpa
3D Tetrahedral mesh used.
Surface to surface contact between chassis members.

A Force of 34335 N is applied to the lower Deck and is distributed across all the beams that come in contact with the floor.
I propose to use Fix constraints on four RHS points that connect to the suspension system.
A force of 4905 N in applied on the Kingpin that connects to the upper Deck.

After solving I get this error in F06 file.
"SYSTEM FATAL MESSAGE 3000 (CNTITER2)
ITERATIVE SOLUTION FAILED DUE TO FAILURE OF PRECONDITIONER TO FACTOR.
THIS ERROR CAN RESULT IF THE STRUCTURE IS NOT RESTRAINED SUFFICIENTLY TO PREVENT
RIGID BODY MOTION OR IF INTERNAL MECHANISMS EXIST."
How do I add more constraints to achieve my goals?

See pic attached.

Thanks
Regards
Mike

 

[GregLocock]

Which degrees of freedom do your fix constraints fix?

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[rickfischer51]

I'm not familiar with NX Nastran. Assuming that "fix" on a node in a tet mesh is UX=UY=UZ = 0 and your four nodes are located so as to define a plane, then it looks like you have restrained all six DOF and the restraints are sufficient. So, something is loose. Part of the model is not attached, or not attached well. Run a modal analysis and check for zero Hz modes. If you have them , plot deflection and you should see the loose part. The error message is vague but may be the result of a very large displacement. Check thickness and material properties. Maybe you got some units hosed up

Rick Fischer
Principal Engineer
Argonne National Laboratory

 

[mnmhalstead]

Thanks guys for replying.
@GregLocock,
The fix constraints are in the +Z direction.

@rickfischer51,
Thanks I shall try your suggestions.

Thanks
Mike

 

[GregLocock]

You have undercosntrained your model, add two x and one y

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[dwallace1971]

Oh my! I think you'd probably best get an experienced stress analyst to help you with this. The answers you seek will best be found building either multiple FE models or by using your FE model as a loads model and then performing the detailed analysis using classical methods. Either way, there are many things you could do to muck this up along the way.

"On the human scale, the laws of Newtonian Physics are non-negotiable"