Simple Tension test û why is my model too stiff?

[Ld3Ake]

Hi, I’m trying to validate my FEA models by using experimental stress-strain data from a composite tension test. I’m using NEI Nastran but for some reason the FEA results are way too stiff. I thought initially it was a material model issue so then I changed to an isotropic material – aluminium.

I have now done one of the simplest tests possible: 2D rectangle (with quad elements), fixed BC along curve at one end, force load along curve at the other end (low value so the model is always in a linear regime). The tensile modulus is 70.3 GPa, so logically after outputting the stress and strain results I expected to get the same answer back. However my results were once again too stiff and I had a predicted stiffness of 88GPa.

I really can't understand what is going on. I thought this was such a simple test.

Thanks for any help

 

[rb1957]

you're plotting elastic strain ... looks like something like 0.003".

are you accounting for poisson effects in your hand calc ?

Quando Omni Flunkus Moritati

 

[SWComposites]

So perhaps you could post ALL of the details of your hand calls from the FE results

 

[IceBreakerSours]

Are you sure your units are consistent?

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[Ld3Ake]

Hi guys, I got a reply from my Nastran reseller. Basically the problem was related to how Nastran how output results. Apparently it can't get stress and strain to output in the same analysis (well it can, but it generates "effective strain", i.e. von mises strain, which is not correct for a uniaxial test) so I have to run the model twice: once for stress, the other strain. This seems ridiculous to me that I can't get the right results in a single analysis, but on the other hand I do now get correct stiffness results. Has anyone else found this peculiarity with Nastran?