Modal Analysis: Why high stress/displacement?

[cvan42]

Good evening, working in Workbench. Been analyzing a small plastic (Valox) gear, and was interested in modes. Where reasonable frequencies are shown in results, 6 modes within 1200 Hz to 1500 Hz, stresses and displacements are unreasonably high, 1.5 e8 PSI with lowest stress 50 KSI. With a 7500 PSI yield, it seems this gear would've exploded. This doesn't seem right... any help would be greatly appreciated. Thanks.

 

[GregLocock]

Modal analysis as such doesn't produce meaningful stresses. That is, the analysis is for an undamped mode, and the displacement scaling is not meaningful, directly.


You can get menaingful stresses if you add damping to your model and then drive it with representative forces. That may well be an add-in module to your post processor.




Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Stringmaker]

Normally, computed modes are normalized to the mass matrix. Hence, the computed stresses and displacements are meaningful in a relative sense but meaningless in an absolute sense.

As greg suggested if you want to know something other than natural frequencies or mode shapes you'll need to do a forced response analysis or something of the sort based on the information you have at hand.

 

[cvan42]

Greg, Stringmaker,
Thanks for the insight, will be sure to do more research on forced responses. Knew there's an obvious explanation for this.

Chad

 

[GBor]

I think you are probably looking for "Frequency Response Analysis". Check

http://www.femci.gov

for some basic understanding.

 

[prost]

link

http://www.femci.gov

doesn't seem to work. Perhaps this is the correct one?

http://femci.gsfc.nasa.gov/

 

[spongebob007]

You need to perform a harmonic response analysis. If you have an ANSYS/mechanical or multiphysics license you can do this. If you have the watered down version of ANSYS than a modal analysis is the best you can do.

 

[GregLocock]

Ah, not quite true.

Producing FRFs from a normal modes analysis is fairly easy in a spreadsheet.

Fairly easy...

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.