on rubber dynamic stiffness

[candychan]

Hello,everyone.
I have a question on the calculation of rubber dynamic stiffness.
Presently I am doing FEA by ABAQUS,but it's mainly for static analysis,now I have to do some work for the dynamic stiffness for rubber mountings,for it's the working stiffness.
My question is what's the material input for dynamic analysis for rubber?and what kinds of material tests should be done for it?and what kinds of software is suitable for this work?
I will be very grateful if anyone could proivde some suggestions.
Thanks.

Candy

 

[tomaspin]

The rubber mounting manufacturers I have worked for over the last 9 years have all used MSC Marc for non linear FEA.

The reason for this is that Marc were heavily involved in a government backed UK research programme (ENGELAST), which was set up to research computational methods for analysing the design of elastomeric mountings. I therefore feel that they should be your first point of contact.

Ruber moduli (tension compression, shear) are all strain related, so you will need this data for your compounds before you start. Also phase angle ( a measure of the internal damping) will also be required. Marc should be able to advise further.

I hope this helps

Regards,

Tom

 

[candychan]

Thanks Tom.
I read the previous posts,and I know you are the expert in rubber field.So maybe I could learn a lot from you.

Presently I have am using ABAQUS,and also get some imformaton from the support.It's said the hyperelastic property as well as the viscoelastic property should be defined for the simulation.But I couldn't get the corresponding testing parameters to the coefficient of the viscoelastic constitutive model.i.e,I couldn't input the loss angle directly in ABAQUS.

I am not sure if Marc could deal with the loss angle.
I have the necessary datum for static biaxle tension,simple tension,plane tension and volmetric tension,with a varied strain.So the elastic modulus could be easily obtained.

Do you think the dynamic elastic modulus is also necessary?
Have you ever tried to do the dynamic FEA for rubber mountings?

Thanks very much.

Candy

 

[tomaspin]

Candy,

Dynamic elastic modulus is a function of the loss angle, which as you know is a measure of the internal damping.

So it may be that some software uses the loss angle and (static) elastic modulus to calculate dynamic modulus, where others may use dynamic modulus directly. It is still worth contacting Marc to discuss the type of modelling you wish carry out, and the data that you have available, to see if their code offers any advantages.

Dynamic modelling will add a whole new level of complexity to the calculation, as the dynamic modulus is dependent on the dynamic strain. In practice, you will probably have to take a mean value, and accept the limitations of the model.
In my experience, non linear FE is mainly used to analyse rubber stresses over the full load range and to calculate load deflection curves. The results are then used to identify stress concentrations to be identified, and the rubber section to be optimised.

I assume you are doing the dynamic FEA in order to look at the cyclic stress and therefore make a fatigue life prediction. Is this correct?
Even with improved modelling techniques, accurate fatigue predictions are not easy to make, and the usual course of action is to undertake a series of destructive tests to prove the design.
Non linear FEA is s specialist area, and most companies ( including the one I work for) employ engineers full time to carry out this work. These days I am mainly concerned with inital mounting design, installations, and rigid body analysis for off highway vehicles, so I have no time to play with FEA as well!

I would also recommend that you do a database search for technical papers on the subject of non-linear FEA, as these will describe the modelling techniques used.

I hope this some help to you.

Regards,

Tom

 

[tomaspin]

Candy,

The following paper might be useful:

http://www.mscsoftware.com/support/library/conf/adams/euro/1998/euc98_05.pdf

Tom

 

[brad]

Tom,
I find it curious that you would so quickly dismiss ABAQUS in this field. While there may be some argument as to which one is best in rubber modeling, most people in the field acknowledge that both of these analysis codes are exceptional for this type of work. (In fact many of the original developers for MARC's rubber capabilities went on to develop ABAQUS's capabilities).

Candy--if you have the budget for it, you may want to contact AXEL Products in Ann Arbor, MI. They do material testing of rubber products to extract properties for analysis codes. I know that they do such tests for ABAQUS and MARC. (In fact, they have even taught rubber testing courses with both ABAQUS and MSC.MARC). NOTE--I have no affiliation with them; I only know from others that they do good work.

Brad

 

[candychan]

Thanks Tom and Brad,
My purpose is to calculate the dynamic stiffness and the cyclic stress,and then predict the life of the products by FEA.
I have the same opinion,a good software and a good material input are the key of the dynamic analysis.I think both ABAQUS and MARC are suitable for such kind of work,because they are excellent codes in nonlinear analysis.
I will contact Mr Kurt in AXEL company for the dynamic material test.
Thanks.

Candy

 

[craigj]

http://www.mscsoftware.com/support/library/conf/auto00/p04500.pdf

Ansys also has a good nonlinear code

There is a test lab called datapointLabs that might be worth checking into.

http://www.datapointlabs.com/

Axel products is also, as you know from a previous reply, a good testing lab. (they also teach classes with Ansys...)

Cheers,
CJ