Dear all,
I'm observing something suspicious in a simulation that I wanted to share with you. I appreciate if you could help me out with it please: I'm simulating a 2D plate which is fixed in one end and I'm displacing the other end mimicking a uniaxial test. Now, when I average the nodal stress and divide by strain, I got a Young's modulus that shows 5% difference comparing with initial input E. So, is there any plausible explanation for it or there is something I'm missing in my simulation ? I tried different types of mesh, mesh size, mesh refinement, different integration order but all ended up into same observation.
Best, Ali
Hi Ali,
what if you integrate the stress over a surface region to get the force, instead of using the nodal stress (how do you compute that)?
Can you post a small working example demonstrating the problem?
r.
On 12/08/2021 21:22, kshargh.ali@gmail.com wrote:
Dear all,
I'm observing something suspicious in a simulation that I wanted to share with you. I appreciate if you could help me out with it please: I'm simulating a 2D plate which is fixed in one end and I'm displacing the other end mimicking a uniaxial test. Now, when I average the nodal stress and divide by strain, I got a Young's modulus that shows 5% difference comparing with initial input E. So, is there any plausible explanation for it or there is something I'm missing in my simulation ? I tried different types of mesh, mesh size, mesh refinement, different integration order but all ended up into same observation.
Best, Ali
Thank you so much Robert for the reply. I was actually struggling with it for a couple of days :( I just found out that the default option for 2D condition ( thin structure) in SfePy is defined as "plane strain" meaning strain = (stress/E)*(1-vu^2) which I really didn't expect it. Thus, by adding "plane='stress'" now it is working fine ( i.e. strain = stress/E).
Ali
On 18/08/2021 16:34, kshargh.ali@gmail.com wrote:
Thank you so much Robert for the reply. I was actually struggling with it for a couple of days :( I just found out that the default option for 2D condition ( thin structure) in SfePy is defined as "plane strain" meaning strain = (stress/E)*(1-vu^2) which I really didn't expect it. Thus, by adding "plane='stress'" now it is working fine ( i.e. strain = stress/E).
Glad you sorted that out. Writing the issue down as another FAQ topic.
Cheers, r.
participants (2)
-
kshargh.ali＠gmail.com -
Robert Cimrman