Dear Alessando, Thanks fo the feedback. My understanding is that the shear locking in finite elements is due to low-order elements unable to properly represent curved bending deformation and leading to errors in shear strain calculation. However, shear locking, as the name indicates, should result in higher stiffness and the understimation of the deflections. What we are observing here is the opposite effect. For some reason, for these Timoshenko beams, Chrono expiriences deflections much longer than expected.
Also, it is important to note that fo the rectangular beams that I was using in the above example, they are not slender. Indeed, the cross-section dimensions are larger than the beam length. Could you maybe confirm if the way to create these beams using the ChBuilderBeamTaperedTimoshenko is the right one? Thanks for the support. Roger On Wednesday, January 14, 2026 at 1:40:58 AM UTC-7 Alessandro Tasora ([email protected]) wrote: > Dear Roger > > This is a consequence of the current quadrature of the Timoshenko beam, > which is potentially affected by shear locking if the length/width ratio is > high. As we always used Timoshenko beams in scenarios where the > discretization is fine anyway, this was not a problem (the slender ratio is > small anyway, ie. elements are "thick beams" regardless if the entire beam > is thin) and we had no need to implement anti-shear-lock formulas. Maybe in > future. > > Regards > > Alessandro Tasora > > ------------------------------ > *From:* [email protected] <[email protected]> on behalf > of Roger Bergua <[email protected]> > *Sent:* Wednesday, January 14, 2026 12:14:09 AM > *To:* ProjectChrono <[email protected]> > *Subject:* [chrono] Re: Timoshenko beams issue > > I tried to benchmark the results of rectangular beams against other codes > (SubDyn, SAMCEF, HOTINT) and I observed the same issues as commented above. > Somehow, the Timoshenko beam formulation needs a significant number of > nodes to have the proper behavior. Is someone else observing this behavior? > Is this due to an incorrect use or implementation of the builder = > fea.ChBuilderBeamTaperedTimoshenko()? > > Attached to you can find the comparison between codes. And here the > definition of the test case: > https://github.com/OpenFAST/openfast/pull/2646#issuecomment-2688827936 > > Thanks for the support. > > Roger > > On Friday, January 9, 2026 at 1:15:09 PM UTC-7 Roger Bergua wrote: > > Hi all, > > I have been trying to model a cantilever beam in pyChrono. > > When using Euler-Bernoulli beams, everything seems to work as expected. > However, I get large displacements (~1.83 m instead of ~0.89 m) when using > the Timoshenko beams. For reference, I have added the response from a third > code (SubDyn). See the attached image: > *Wrong_Timoshenko_beams_behavior.jpg* > > Interestingly, if the discreization is increased very significaly, the > expected results are obtained. See the attached image: > *Fine_discretization_Timoshenko.jpg* > > I'm not sure what is the issue. But it does not look consistent. > > Attached you can also find the pyChrono input file used in *.txt format. > > Thanks in advance for the support! > > Roger > > -- > You received this message because you are subscribed to the Google Groups > "ProjectChrono" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion visit > https://groups.google.com/d/msgid/projectchrono/ec17dae3-bb6d-45f0-8130-ae216f61894cn%40googlegroups.com > > <https://groups.google.com/d/msgid/projectchrono/ec17dae3-bb6d-45f0-8130-ae216f61894cn%40googlegroups.com?utm_medium=email&utm_source=footer> > . > > -- You received this message because you are subscribed to the Google Groups "ProjectChrono" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/projectchrono/bd9ffeb9-2f91-482a-aa51-c8a965f179e5n%40googlegroups.com.
