Incorrect shear stress distribution in beam bending

[Hilkka]

Hello, I have modeled a beam in bending but the shear stress distribution seems incorrect. Instead of being parabolic and at a maximum in mid-thickness, it has a neutral axis in the middle and extreme values at the surfaces. How could this be possible?

I'm using quadratic solid elements. Thank you in advance for any ideas.

 

[Denial]

What software are you using?[ ] If you are new to that software, I suspect you are misreading its output.[ ] Simple equilibrium considerations dictate that the shear stress on a plane that is normal to a free surface MUST be zero.[ ] And the very use of the term "neutral axis" in the contect of shear stresses gets me wondering.

 

[Hilkka]

I'm using Ansys, and yes I am quite new to it. I was actually able to figure it with the help of my colleagues; the large deflection caused the stress components to look misleading since they were calculated using the global coordinate system. Since the beam bends to an angle of about 30 degrees, parts of the tension and compressive stresses on the surfaces (sigma_x in element coordinates) were interpreted as sigma_y stresses in global coordinates. This messed up the results. RSYS,SOLU cleared the problem.

Thank you for your help.