Vibration Analysis

[Rod]

Hi everyone,
I need some help from people who know how to use Ansys.
In one of my course at university (Ecole Polytechnique de Montreal) we have a projet to do in Ansys9.0. I have try to find answer to my problem and question by myself, but I found nothing and I'm tired of searching and loosing my time.
My problem is: my model consist of a support for a small motor for a formula SAE car (525cc KTM motor). Between The motor and the frame of the car, we have put a support with rubber element to reduce the vibration (like a filter). We know the values of the vibration in hz and the amplitude, but we don't know how to use this in Ansys.
We are thinking about using a spectrum analysis (single pt), but we are not sure about the way it work. There are table with frequencies, but no amplitude table? In the load step opts there is a lot of options and we don't really know how to use them.
If someone can help me, I'll appreciate it.
Thank you
Rod

[Jorgen]

I am not quite sure I understand what you are trying to accomplish. Are you trying to design a vibration damping rubber support? Is your goal to determine the dimensions or deciding what material to use, or perhaps both?

As you know, to use Ansys you need to specify both the dimensions of the rubber daming support, a constitutive model for the the rubber material, and the material parameters that go into the constitutive model.

The vibration data that you mentioned can be used to simulate the dynamic response of the car, but you still need to specify the material model for Ansys.

Best of luck,
Jorgen

[Rod]

Thank you Jorgen,
Yes, I'm trying to design a vibration damping rubber support. I already have the design of the support. We have the vibration in hz and their amplitude. We want to know the stress and displacement of the node between the frame of the car and the support. We don't want vibration in the frame of the car.
I have read in the Ansys help about the analysis. I have try the spectrum analysis with no luck (I think it's not the analysis I want to do). I had luck with an harmonic analysis and find some good result, but I'm not sure of anything! I have follow the instruction about how to do this kind of analysis in the help.
For the support, I use an elastic suspension from Paulstra:
www.paulstra-vibrachoc.com (the radiaflex)
For the material, I use natural rubber 50% HAF black with Ex=4MPa, PRXY=0.499 and density=1100kg/m3 . I also read in the help that in dynamic analysis (harmonic), Ansys just use linear propreties of material.
But now, I don't know how to find damping constant for rubber.
Thank you
Rod

[Rod]

I have an other question,
On my model I have done a modal analysis and I have found some frequencies values (the one that we don't want, "resonance" frequencies). I have found a value of 300hz and some others.
After, I have done an harmonic analysis. I have some value in hz, like 33.33hz,... and 300hz (the motor at 9000rpm). When I solve my model, It give me the vertical displacement in mm of my rubber support at those values in hz.
Do I have to do a superposition of my modal and harmonic analysis to see at the same time all the effect on the rubber? I want to see what happen at 300hz. I don't know if the harmonic analysis take into account the modal analysis?
And I don't realy know how to do a superposition, i'm not an expert at all :(, I just take a little course on FEM like everybody in eng.
Thank you very much
Rod

[Jorgen]

A few pointers:

- You can find the damping behavior through experiments and model the material behavior using linear viscoelasticity. That approach should be the simplest.

- Once you have your FE model setup you can quite easily extract the eigen frequencies of your system. You can then compare those frequencies to 300 Hz.

- You can also drive the hole system by applying any arbitrary loading frequency and amplitude. The idea here would be to drive the system as realistically as you can and observe the resulting system behavior, amplitudes, etc.

Jorgen

[Carlos]

Hi,
what I am doing is the following
once you have the frecuency and amplitude of the excitation, you should characterize the material. To do this I considered the bulk modulus constant (since I have no access to testing machines for measuring bulk modulus) and realized some compression experiments at the frecuency and amplitude of interest. With this information one can compute the elasticity modulus, Poisson ratio and, as a consecuence, the shear modulus. The bulk and shear modulus graph over time is asked by Ansys to fit the curve to a prony model. The constants given by Ansys will be used to represent the material in your model.
After this (I have not done it, but I hope it works) you should correlate a Ansys model representing your specimens in the tests with the experiments (it somehow should fit :lol: ). By doing this you can become confident in your results.
Next, you can finally model you CAR dynamically. I will do it in a explicit manner since my model is not too big. Here you apply the excitation frecuency and amplitude you desire. I will measure the system response to compute the transmissibility, which hopefully will correlate the experimental vibration data as well.
Finally, and if required, you can perform a change in the design in the computer.
Hope this helps.

Carlos.[/i]