## General Mechanical

#### Solver pivot warning or error has been detected

• rajesh10
Subscriber

Hello Everyone
I'm facing a problem related to solver pivot point error. The analysis is of 3DOF planar mechanism, consisting of two bodies (Body1 & Base) connected by vertical longitudinal spring. A remote force is applied at base and is constrained in all direction except vertical. The Body1 is free in vertical, horizontal and rotational axis. The Body1 is constrained in horizontal and rotational direction using spring elements. The solver continuously giving the pivot point error. I have tried all suggestion as provided in the discussion related to pivot point error but able to find any solution.

I have attached the Ansys files for your understanding.

It will be highly appreciable, if anyone could provide solution.

• peteroznewman
Subscriber

Hello Rajesh,

It looks to me like your model is more about rigid body motion and less about deformation within a body.  Below is a snapshot of the model.

Are you interested in the motion of the body including the effects of inertia while increasing that force over 5 seconds, or are you only interested in the static equilibrium at each level of force?

If the inertia effect is relevant, I recommend a Rigid Dynamics model. I have attached an archive where the two bodies are Rigid Bodies.  It adds a Translational Joint to Ground to the rectangular Base body and a Joint Load of the force vs. time you had.  It adds a Planar Joint to the ellipsoidal body, and the thickness is adjusted to make it 20 kg.

You don't say what the mass of the base is. I made it 10 kg. Leaving it at the default mass of a fraction of a gram creates a system with a very high natural frequency that is much harder to solve in a dynamics solver.  This is a two body system with 4 degrees of freedom, with a total mass of 30 kg. The joint load pushes on the base with a constantly increasing force. The force ramp excites one of the modes and the two masses begin oscillating.

The attached archive does not include results, so you have to Clear Generated Data and Solve.  It takes 233k time steps to solve out to 5 seconds.  Once the solution finishes you will get this plot of the deformation. Be patient, it takes a long time to plot all those points!  You can reduce your wait time by setting the End Time to 2 seconds instead of 5 seconds.

• rajesh10
Subscriber

Hello Peter..

Thanks for the quick response to my query. Yes, I do want the inertial effect. So, as suggested I used RBD analysis, but with the command snippet to activate the rotational DOF of the longitudinal spring, error is occurring.

"A general failure occurred during the solution process."

Please suggest some solution.

Thanks

• peteroznewman
Subscriber

Mechanical APDL commands don't work in Rigid Dynamics because it is not using the Mechanical APDL solver.  Delete the Command Object.

What does it mean to have a torsional spring to ground?  How does that work when the body is also translating?

Instead of a torsional spring, you could have two vertical longitudinal springs from the first body to the second body that are parallel and separated in X.  Together, they have the same stiffness as the one spring when the ellipsiodal body translates but does not rotate, but when the ellispoidal body rotates, they provide a restoring torque.

• rajesh10
Subscriber

Hello Sir...

The model is such that I can't use two vertical springs. To apply rotational constraint, torsional spring needed. Since RBD uses python scripting and I'm not much familiar with it. I think transient structure with rigid bodies may solve my problem.

• peteroznewman
Subscriber

Physically, how do you make a torsional spring that can connect between ground and a body on a planar joint to ground that allows it to translate and rotate in the XY plane?

• rajesh10
Subscriber

I'm trying to make this model...