Why do we need angular velocities in collision analysis?

General Questions related to the Momentum Based Analysis programs
#pc-crash #virtualcrash #crash
brian
Posts: 499
Joined: Tue Jul 14, 2009 10:52 am

Why do we need angular velocities in collision analysis?

Post by brian »

Q: May 22, 2010: According to some researchers, one of the bases for superiority of planar impact mechanics to CRASH3 is, rotational motion is examined and angular velocities can be found. But why do we need angular velocities? Why do we need to find angular velocities of vehicles after collision? Is it about laws or something else?

A: Consideration of angular velocities in any collision analysis method is to be sure that angular momentum is conserved and that the impact configuration produces moments which are adequate to produce the proper amount of rotation of each vehicle subsequent to the collision interaction. Simple Linear momentum analysis methods do not calculate or consider angular velocities.

When you begin consideration of angular momentum approximations, additional items must be approximated like the separation positions. See our research related to CRASH-97 - Refinement of the Trajectory Solution Procedure, SAE Paper No. 97-0949. In that publication we included angular momentum in CRASH3 and found that actually simulating the separation to rest trajectories of the vehicles helped improve the correlation of results. It also pointed out several factors of importance for any analysis technique for collision analysis:
  • 1) The approximation technique used for determining the separation positions and orientations.
  • 2) The approximate duration of the time from impact to separation.
  • 3) The magnitude and duration of the forces and moments used to approximate the effects of external tire-forces during the collision.
  • 4) Creation of evaluation terms to objectively determine when you have an “acceptable” match.
Lastly, I’m not sure I agree with ‘superiority of (some forms of) planar impact mechanics’ over CRASH3. Here a distinction must be made between non-simulation forms of approximation techniques and simulations such as SMAC. The non simulation forms include a number of 'fudge factors' which when applied in the "real world" collisions can not be directly measured and verified.
A couple of examples:
  • 1) angle of the damage surface - why not simply measure the damage like in CRASH3 and let the program internally determine the 'angle of the damage surface?
    2) The impulse ratio, MU - how is this to be determined? It appears to be inconsistently applied (see the ECF thread) where in a 10 degree front to rear impact, MU values of .038, .031, -.065, -0.50, -0.09 are used? It appears to be a subjective input).
See our response to a question on the CRASH ECF which includes a presentation of some of the and also see the thread
Question? Comment? Please email us
MORE TOPICS see:Forum Index & McHenrySoftware.com
(c)McHenry Software, Inc ALL Rights Reserved.