Learning Forum

[Jean Vasco]

Hi, everyone!

I'm studying the vibrations on a free-free beam under axial compression loading. I have a physical model setup in a laboratory, with an aluminum beam axially loaded suspended from a flexible string to approximate the free-free boundary conditions. So I have a hollow tube with a steel cable fixed at one end and "free" to be tightened up at the other (see the image for the setup schematic). As I tension the cable, the end caps compress the beam reducing the overall stiffness of the system and, by consquence, the its natural frequency of vibration.



I tried doing a similar setup in Ansys but with just the hollow tube and axial loading with weak springs turned on to approximate the free-free ends condition. The results were not as close to the results of the experimental setup as I would like. There is a suspicion that the natural frequency of vibration of the steel cable is close enough to the natural frequency of vibration of the beam to affect the overall result. So I'm trying to model the steel wire and the end caps as well.

Right now, my model setup in ansys if the same hollow tube, but with the end caps and a cirular bar acting as the wire. I tried using LINK180 to model the cable, but I haven't been able to make it work (kinda of a beginner in ansys). The circular bar is connected to the caps which are in contact with the ends of the hollow beam. I applied a bolt pretension load on the circular bar to model the pretension on the wire.

With zero loading, the natural frequency is way below the zero loading result obtained with the physical model. But as I increase the bolt pretension loading, the frequencies also increase, which is not what should be happning in the overall case. I suspect that this increase in stiffness/frequency is happening at the wire level (which is what happens in the physical model). But I'm expecting a decrease in the values of the natural frequency of vibration for the beam as the pretension load increases.

For reference, with zero loading I should be getting around 20.51 Hz, but I'm barelly getting 10 Hz with my ansys model. 

What should I be doing to better represent my physical model?

I'm uploading my model as well for anyone who wants to take a look.
Beam-Cap-Cable-Setup.wbpz

Thanks!

[Ashish Khemka]

Hi,

Ansys employees cannot download the model. I would suggest using a prestressed modal analysis.

Prestressed Modal Analysis | Ansys Innovation Courses

Regards,

Ashish Kumar