finite element modeling of isolator

[mdhayes]

I'm performing a harmonic vibration analysis (FEA) on a structure. There is a component mounted to this structure via isolators. I have the specifications for the isolator, which includes a static load rating at a give deflection and a ratio of dynamic and static spring rates.

How can I use this information to model the isolators with finite elements?

Thanks.

 

[GregLocock]

Static rate= static load rating/deflection

dynamic rate =static rate*ratio

so put the dynamic rate calculated above into your stiffness matrix, in the correct units

Cheers

Greg Locock

 

[mdhayes]

Thanks Greg, but I also need to capture the damping effect of the rubber used in the isolator. I'd like to represent the isolator as a sping-dashpot element, but I need to figure out how to calculate the dashpot parameter.

 

[tomirvine]

Rather than using a dashpot, you can enter "modal damping" into the FEA software

A typical value for an isolated system might be Q=5, which is equivalent to 10% damping.

Tom Irvine

http://www.vibrationdata.com

 

[GregLocock]

Guess what.

You can't put a damper into a conventional FE model.

The reason is that the standard formulation assumes that the damping is proportional to the stiffness or mass matrix (I can't remember which).

Tom's workaround is the usual solution, or else you export your undamped results to a program which can handle non proportional damping.



Cheers

Greg Locock

 

[mdhayes]

Thanks for the help guys. I've been reading through the ANSYS and ABAQUS help menus and through Harris's shock and vibs handbook, so I'm starting to "get it" now.

 

[brad]

Greg--FYI...
It's proportional to the stiffness for "stiffness-proportional damping", and proportional to the mass for "mass-proportional damping".

(sorry, I couldn't resist )

Brad

 

[GregLocock]

No wonder I couldn't remember which it was.



Cheers

Greg Locock