Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Traction boundary condition: Linear Elasticity

Krishna Kumar Narayanan

Please login with a confirmed email address before reporting spam

Hi,

An I-section beam (h=0.01,w=0.008,l=0.125,th=tb=5e-4; all dim in m) is simulated for deformation and cauchy stresses for linear elasticity. The body load is 1 N at the open end. The length of the beam is oriented along y-axis, width along x-axis and height along z-axis. The simulation provided me the x,y,z coordinates and the 6 cauchy stresses. When i analyse the output in python against neumann (traction) boundary condition, i found a discrepancy that i could not explain.

According to the equations, at the open end the traction boundary condition is

Since the load is acting along the negative z-axis downwards, , where f = -1 N and sa is the surface area calculated using

Thus plugging in this into the equation where n = [0,1,0], one should expect

(on the boundary points at y = L),

which i did not find. Am i checking the boundary conditions wrong? Attached are the exported comsol text file and the python code to test.

Thank you very much



2 Replies Last Post 24 mai 2023, 06:22 UTC−4
Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 year ago 20 mars 2023, 09:55 UTC−4

The information in your question is not complete. How are loads applied? You mention a body load at an end, but a body load would act over a volume.

Anyway, the scatter plot looks very much like a text book example of shear stress distribution over an I-beam section: A weakly parabolic distribution over the web, and almost no stresses (in the load dirction) over the flanges.

See for example https://www.ae.msstate.edu/tupas/SA2/chA14.6_text.html

-------------------
Henrik Sönnerlind
COMSOL
The information in your question is not complete. How are loads applied? You mention a body load at an end, but a body load would act over a volume. Anyway, the scatter plot looks very much like a text book example of shear stress distribution over an I-beam section: A weakly parabolic distribution over the web, and almost no stresses (in the load dirction) over the flanges. See for example

Krishna Kumar Narayanan

Please login with a confirmed email address before reporting spam

Posted: 1 year ago 24 mai 2023, 06:22 UTC−4
Updated: 1 year ago 24 mai 2023, 06:25 UTC−4

Dear Henrik, thank you very much for your reply. Sorry to revisit this topic two months later. I saw the reference and i can immediately see that the is a weak parabolic distribution along the web. Nevertheless, maybe you can help me with one curiosity here. According to the theory, the shear stress must be maximum at the neutral axis and minimum at the either end of the web. But what is see is that the distribution has an inverted parabolic shape in that it is maximum at the web ends and minimum in the center. (see attachment) Any idea as to why this could be? Thanks

Dear Henrik, thank you very much for your reply. Sorry to revisit this topic two months later. I saw the reference and i can immediately see that the \sigma_{yz} is a weak parabolic distribution along the web. Nevertheless, maybe you can help me with one curiosity here. According to the theory, the shear stress \sigma_{yz} must be maximum at the neutral axis and minimum at the either end of the web. But what is see is that the distribution has an inverted parabolic shape in that it is maximum at the web ends and minimum in the center. (see attachment) Any idea as to why this could be? Thanks

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.