Torque

[akanakapally]

I am performing a simple static analysis on a hollow cylinder constrained on one end in all DOF and torqued on the other end using a *cload option and a rbe2 element. The stresses I observe on the cylinder are 1.5 times more than what I get through hand calculations. Any idea what could be happening? When I look at the rotational DOF values about the axis of the cylinder I see that there is a uniform value along the length. Shouldnot it be linearly varying?

Abhishek

 

[corus]

If it's constrained at one end then the rotational values should be zero at the constrained end.
Unless you're kust using a beam element with cylindrical properties, then the rotation of the cylinder can only be expressed in terms of translation values whether it's modelled in shells or solids, surely?

corus

 

[akanakapally]

Thanks for the reply Corus.
I am using solid elements for the cylinder so I understand that the rotational DOF values must be zero on it. However my main concern is why are the stresses so high?

 

[corus]

With solid elements there are no rotational dof, so I'm not sure what values you're looking at.
With RBE2 elements you're effectively restraining the end by which you're applying the loads, radially. If this distance is 'reasonably' close to the constrained end then you may have some stress added to the constrained end. Add a picture of your stresses then people can make a judgement if this could be the cause .. or something else.

corus

 

[akanakapally]

I found the answer for this. I was looking at the von mises stress values. I learnt from abaqus tech support that I must be looking at the shear stress contour plots instead. Thanks for the help Corus.