Could you post the closed-form formulations as well? Sorry for asking too much.
Just run the case01_ABAx.inp, why the volume of the 3D element is increasing? Elastomers are imcompressible in most time, arnt they?
Should the element contracts in 2 & 3 directions under uniaxial tension loading along 1 direction?
answers
The closed-form solution was obtained from a specialized computer program that I am working on. This program is still being developed and I cannot disclose more information about it at this time. For the NH-model, however, it should be easy enough to derive the response in uniaxial tension and simple shear.
For test case [i]case01_ABAx.inp[/i] the volume is slightly increasing due to the chosen material parameters. This volume increase is cased by the hydrostatic tension component of the stress state. The amount of volumne change is exactly as expected.
I have a question here, is that Neo-Hookean model is suitable in predicting rubber under compression?
I was given a set of data which provide me force-deflectin data. The rubber is cylindlicar shape, height=0.925 inch, Diameter = 0.984 inc, shape facter =0.2659. Initial young modulus under compression about 2050 ibs, poision ratio take 0.4997 (by guess). The rubber be compressed 0.118 inch (3 mm), 12.77% strain. When I using Ansys, neo-hookean model to compare the test data i given, that is totally unmatch with the force-displacement curve unless first 0.1-0.3 % strain. Afther that the force shoot up without stoping as the stiffness almost not softern but the given test data actually sorften after 2% of compression.
PS: I am not given any other data for using orther material model but just Neo-Hookean. And is that anybody can tell me why this model so unmatch with this case?
The Neo-Hookean material model should work reasonable well for strains up to 12%.
How did you decide which material parameters to use?
Which parameter to use?
If you mean the material parameter need to use in Neo-hookean in ansys it was already set by the program. For neo-hookean it need intial shear modulus and d, which equal to 2/K, where K is bulk modulus.
Up to 12%.... is that apply to compressin also or just tension?
With the rubber disk data given by vender, it is pretty high in initial young modulus under compression. I convert it into initial modulus with
G=E/2(1+2r) and bulk modulus K=E/3(1-2r) and r is potion ratio be set to 0.4997. It end up with the curve almost a straight line with only a bit curving and only match in begining about 0.01% strain only. The rest it just keep on going with almost linear but the graph given by vender actually soften in stiffness and two curve become further and further away.
Thank you for reply
The Neo-Hookean material model should work relatively well for the small strains that you are interested in, and the approach that you are taking seems correct. I cannot say why you dont get a better aggrement 😕
Re: vumat validation
[quote=Jorgen_Bergstrom][img] 
[/img]
Dear Jorgen,
Do you know why the stress-strain curves are not smooth for vumat 4x4x4 elements shear tests? is it related to the explicit solver?
Thanks,
Wei