Modelling viscoelastic material attachment

[mechie23]

Hi,

I am trying to model a cylindrical casing with a neoprene sheet attached to it (through adhesive tape). Some masses are mounted on the neoprene intermittently, again with adhesive tape, with a hope of anchoring the neoprene.

Objective is to see how far the neoprene sheet and added masses would be effective in damping the vibration. As a first step I intend to see variation in strain energy in neoprene sheet in modal analysis step. I am using viscoelastic properties from a dynamic mechanical analysis test.

The problem I am finding is that elements where masses are mounted are severely distorted for any mode shape. And very low strain energy is developed in neoprene as compared to the base casing. I am using a 'tie' constraint to connect casing surface and neoprene surface and also that of neoprene and mass. Is this way of modelling correct and how to minimise element distortion ? I can understand that neoprene Young's modulus is very low as compared to steel masses, but is there any other way?

Thank you for your time and help.

Regards
mechie

 

[mechie23]

I thought these pictures would clear my explanation

Fig.1 - Total assembly strain map in a mode shape

Fig.2 - Only neoprene sheet with its distorted elements at the location of mass

 

[burningDiesel]

If you are not interested in the masses, maybe you can add them as point masses to nodes in your model?
And if your geometry is simple (cylindrical), why don't you make the nodes of steel+neoprene match, removing the need for the tie constraint?

 

[mechie23]

sdebock, thank you for your response.

Actually the masses being tuned masses (for a frequency), I would like to have them occupy the surface area for that mode shape at antinodes. I am really not sure if I can simply put a point mass at one FE node at each antinode of mode shape.

Can you please elaborate what you meant by matching the steel+neoprene nodes? In fact during meshing I made sure they are matched node-to-node visually. But even then we have to define a constraint. Right? Otherwise how they will remain attached?

Kind regards