Control arm analysis

[josh2417]

hi,
I need some help with the analysis of control arms. I am using ansys but there is a "negative equation , insufficiently constrained model" message every time solve. I want to confirm if the constraints and loads im applying is correct.
The loads are combined vertical with acceleration, braking and lateral force.
I made a dummy arm for the hub and applied the force at the end of hub
Constrained all the suspension inboard points on all dof. Lower ball joint and spring mount on lower arm are constrained in all dof. constrained the ubj on all sides except the vertical direction. How to constrain the spring chassis mounting point?
Thanks and cheers

Suspension team-2012
GS Racing
MIT,India

[Sormaz]

"Constrained all the suspension inboard points on all dof"

Do you weld your arms to your chassis?

try gaining an understanding of the physical system before running analysis on it

[shark.ashwa]

Originally posted by josh2417:
hi,
I need some help with the analysis of control arms. I am using ansys but there is a "negative equation , insufficiently constrained model" message every time solve. I want to confirm if the constraints and loads im applying is correct.
The loads are combined vertical with acceleration, braking and lateral force.
I made a dummy arm for the hub and applied the force at the end of hub
Constrained all the suspension inboard points on all dof. Lower ball joint and spring mount on lower arm are constrained in all dof. constrained the ubj on all sides except the vertical direction. How to constrain the spring chassis mounting point?
Thanks and cheers

Suspension team-2012
GS Racing
MIT,India

Josh,

As Sormaz put it, you need to gain an understanding as to how forces are transmitted from the tire contact patch to the chassis. A simple FBD should do it. As for constraining the chassis inboard points, think about what degrees of freedom are still available to the arm in actuality.
Why would you constrain the LBJ in all degrees and not the UBJ. There are several threads on the forum on loadings and constraints applied, I suggest you check them out.

Regards,
Sharath

[josh2417]

sormaz and sharath,

Yes you guys are right, we are using spherical ball joints which means there will be moments but no displacement.

We are finding it extremely difficult to simulate the analysis in ansys though.

Using FBD, the forces at tire is transfered to the hub, then to the uprights and arms.
I have read that the vertical load is entirely taken up by the spring while the arms take up the cornering and accler/braking forces.
In the same way most of the load is taken by the lower control arms.

I used a dummy spindle to replace the tire at the uprights. Is this method correct?

[shark.ashwa]

" moment but no displacement"
No I'm saying there will be rotational degrees of freedom in the ball joint which you have to appropriately constrain as such.....be it outboard or inboard.
I'm not much of an FEA guy.......

Sharath

[CameronBeaton]

Originally posted by josh2417:
sormaz and sharath,

Yes you guys are right, we are using spherical ball joints which means there will be moments but no displacement.

We are finding it extremely difficult to simulate the analysis in ansys though.

Using FBD, the forces at tire is transfered to the hub, then to the uprights and arms.
I have read that the vertical load is entirely taken up by the spring while the arms take up the cornering and accler/braking forces.
In the same way most of the load is taken by the lower control arms.

I used a dummy spindle to replace the tire at the uprights. Is this method correct?

At you trying to replicate the whole suspension? It's a little needless complictated in opinion.

I would create a spread sheet or matlab file with all the hard points as inputs and then calculates the forces acting on each element using moments, ect. This is a little tricky but will save alot of time in future years.

Then model the parts seperately as I feel you error in probably coming from the parts interacting (or not interacting) and you upright moving ifinatly as there is no restrant there. Or (if you really want to use FEA) you could attach the parts with ridgid links at the contact points of the different parts and see if the error goes, there is a good chance you'll get some new errors but if that one goes then thats then what I said is most likely correct.