resonance frequency V beam

[Oonizuk]

Hi people,
I am trying to model the V shape fixed-free beam (view on attachments) with beam dimensions in order to get the resonance frequency.

[URL unfurl="true"]http://files.engineering.com/getfile.aspx?folder=f8744b93-ea3f-4c8b-866e-2f21be3ea0c2&file=beam-shape.png[/url]

I tried a 2 x (mass spring system) or beam FEA models without success.
Do you have any idea what I have to do?

 

[btrueblood]

What are you comparing the FEA results to? Did the FEA analysis allow for torsional (twisting) modes, esp. for the #1 segment?

 

[GregLocock]

That's quite a nice little example. There is no reason why FEA should not be able to solve it, but it would be a bit tricky to check the FEA with a handcalc, although you could check that it solves some similar configurations accurately. For example, replace the free arm with a flywheel, do you get the right modes in flex and torsuion ?

Cheers

Greg Locock


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

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

 

[Oonizuk]

I compared results with real experimentation and comsol model. Both are coherent but not with my model.

I choose a 2 flex beam model with 3 dof by node (2 translation and 1 rotation). Then I wrote down M and K matrix and try to solve det(K-w²M)=0, with computer in the first place, but results are far away from reality??

I do not know how to "model" the bend. Should I add a extra stiffness or torsion moment on node 2 or something like that??

 

[btrueblood]

I'd think in the real world the lowest frequency mode would involve rotation about the axis of the cantilevered, lower (#1?), beam. To resolve that mode would require a 3D beam model, not a 2D solution. Unless the "real world" beam was constrained against such rotation, or the beam depth was such that the torsional stiffness was much greater than the bending stiffness.

 

[GregLocock]

The dominant flexure of the first mode is very much dependent on the ratio of length of the arms to the diameter of the beams they are made from.

"I do not know how to "model" the bend. Should I add a extra stiffness or torsion moment on node 2 or something like that??
"

No, and that question indicates you haven't got the physics right.

Cheers

Greg Locock


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

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

 

[Oonizuk]

Thank you