Positive Energy Error - R66 Rollover simulation

Hi there,

I've been trying to do a Rollover simulation according to R66. I've calculated the impact velocity and implemented it as an initial condition.

The energy diagram I'm getting overshoots a little bit after the impact. 

I've been investigating the Kinetic and Internal Energy curves to find the parts with highest such energies and making improvements along the way. so the energy diagrams have improved a lot since I started. Yet, there's this overshoot that happens a little while after the impact that I have no idea about!

Additional info on the model:

• I have added mass on pillars (to take passenger masses into account).
• There are a number of components modeled as Point Mass and connected to the structure using RBODY.
• The velocity contour is attached (GIF). I see extreme velocity increase in some areas after impact.
• Ground is a Rigid Wall (RWALL).
• I used DYREL, the results improved a bit.

I appreciate your comments or tips on this.

=====================================================================

Global Parameters:

Internal Energy (IE):

Kinetic Energy (KE):

Total Energy (TE):

Velocity Contour:

High Total Energy Components: