Choosing FE Simulation Type for Dynamics and Impact/Force Data into FE 1

[Searc]

I am running the FEA of a wheel hitting a wall and trying to determine whether I have to use a dynamical simulation, or whether a static simulation can give reasonable results instead. I am interested in the stresses and deformation of the wheel parts.

Question 1

The Solidworks Simulation help states "generally if the frequency of a load is larger than 1/3 of the lowest (fundamental) frequency, a dynamic study should be used."

As I don't have harmonic loading, what does the 'frequency of a load' refer to in terms of a one-time impact? I imagine you use the conversion of 1 / time period of the impact = frequency??

Question 2

Am I correct in thinking the following rules apply when choosing between the simulation types for dynamical systems:
[ul]
[li]Linear Static - can be used if the frequency of a load is smaller than 1/3 of the lowest (fundamental) frequency, the materials are stressed below their elastic limit, there are no large deformations and there are no changes to the states of each contact over time.[/li]
[li]Linear Dynamic (Modal Time History) - if the frequency of a load is larger than 1/3 of the lowest (fundamental) frequency, but the materials are stressed below their elastic limit, there are no large deformations and there are no changes to the states of each contact over time.[/li]
[li]Nonlinear Dynamic - if the frequency of a load is larger than 1/3 of the lowest (fundamental) frequency, and either the materials are stressed above their elastic limit, there are large deformations, and/or there are changes to the states of each contact over time.[/li]
[li]Explicit Dynamic - if the frequency of a load is larger than 1/3 of the lowest (fundamental) frequency, and there is a great level of plastic deformation.[/li]
[/ul]

Question 3

3a. To use a Linear Dynamic (Modal Time History) simulation to study an impact you first would use a rigid-body simulation to find the graph of forces on the wheel over time. Then you would apply this forces over time data to your Linear Dynamic (Modal Time History) FEA?

3b. To use a Linear Static (using the 1/3 of the lowest (fundamental) frequency assumption) simulation to study an impact, you first calculate the worst impact force and apply this in the study, or do you find the average impact force over the timespan of the impact?

Thanks!

 

[GregLocock]

1 yes
2 seems reasonable
3a as a first cut, basically it is a spring mass system, but once you switch analysis type you'll generate a different force pulse
3b I'd use the peak force.

If you describe what you are actually simulating more thoroughly I'm sure you'll get better answers


Cheers

Greg Locock


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

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

 

[FEA way]

Something like this should be simulated using nonlinear dynamic (implicit or better explicit) analysis. Unfortunately, SolidWorks Simulation doesn’t offer much in this regard.

 

[Searc]

Thanks Greg,

The peak impact force being Fmax = (m * v^2) / s ?

Which is twice the average force = (½ m v^2) / s

 

[Searc]

Thanks for the reply FEA way. Just for context, this is a wheelchair hitting a wall and not a car crashing into a wall. The stresses and deformations are low, hence, why I was thinking of Linear Dynamic (Modal Time History) or even linear static given the 'if the frequency of a load is larger than 1/3 of the lowest (fundamental) frequency' assumption. Do you still think this requires a nonlinear dynamic study? Also doesn't Solidworks have implicit nonlinear dynamics?

 

[FEA way]

Yes, SW Simulation has a nonlinear dynamic procedure (using an implicit time integration scheme). This or maybe even static analysis (if defined properly) should be sufficient here when large deformations are not expected (like it’s usually the case in car crash tests).

 

[GregLocock]

The peak force in an SDOF collision is defined by the time constant of the system and the momentum and the spring rate.

Cheers

Greg Locock


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

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

 

[IceBreakerSours]

What experimental measurements are you using to compare your model predictions?

FYI As briefly alluded to previously, at a very high and intuitive level, explicit and implicit are time integration schemes to solve the same underlying governing equation of motion. Codes with implicit schemes give you additional useful capabilities like static, linear, and nonlinear dynamic analyses. The thumb rules you enlisted are very useful ones thumb rules - nothing more, nothing less - which is to say that the knowledge/experience/.. of the physics involved and the measurements taken from the system of interest taken within the context of the risk associated with any error is ultimately paramount. The underlined part doesn't get as much attention but this site has plenty of good discussion on the rest.

*********************************************************
Are you new to this forum? If so, please read these FAQs:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083