One way constraints --> Mechanica WF 2.0 1

[Guest0527211403]

Hi Everybody

After a bit of a hiatus, I'm back in the Pro/E saddle, but now I've moved on a job that has me performing more FE analyses with Pro/M.

I'm analyzing a steel shell structure that sits on a number of bridge bearings, similar to those at

http://www.fabreeka.com/products/expansion.htm

. These bearings allow for translation/rotation in the horizontal plane, and the structure can "lift off" of the bearings as well. So I'm interested in constraining the structure vertically, but in one direction only.

I'm performing modal analysis to determine the natural frequencies of the structure. I've tried constraining it vertically at the bearing locations (assuming that the weight of the structure would prevent any appreciable liftoff), but I'm not confident that the results are accurate. I'm concerned that my structure will experience resonance that might cause liftoff to occur, but I'm having a hard time verifying this through the model.

Can anyone suggest a method to model a constraint that allows for positive translation only along a specific axis?

Thanks in advance!

 

[seymours2571]

Hello,

One way constraints are not possible to create. The only way to somewhat simulate a one-way constraint would be a contact analysis. However, non-linear contacts are not allowed in a modal analysis.

You could try a transient analysis, but I am sure if contacts are allowed in that analysis as well. The output from that analysis would be in the form of dispalcement versus time, which using some form of FFT you could extract the modal frequencies. It is most definately easier said than done.

Perhaps the only possible way to get the results you required would be to run the modal analysis both with and without the constraint that you are needing to be one-way. Then you would know that the modal frequency would lie somewhere between the two sets of results.

Good Luck,

Steve

 

[Guest0527211403]

Thanks Steve

Your idea about using a transient analysis to apply an impulse is certainly an interesting one, but I'm afraid I won't be able to pursue that.

I've bounded the problem as you suggested, and I'm a lot more confident in the results.