Learning Forum

[Philippe CLECH]

Good morning to all

During my training, I was told that in order to obtain a compliant result, the value of the shear stress, for example, should not differ by more than 10% between the averaged and unaveraged result. Is this rule imperative globally or locally?

overall result (Overall model):

[Ashish Khemka]

nnThere is no rule as such. But when you refine a mesh, the results change. In general, when the change is within say 5% or 10% (based on what may be acceptable by you or design needs) we say that the mesh is converged or the result is acceptable.nnRegards,nAshish Khemkan

[Philippe CLECH]

Hi Ashish.nSo if I go back to my example, I'm trying to validate the sizing of the failure zone.nIn this case, for example, the delta of the averaged and unaveraged maximum shear stress result must be 10% less if I am interrogating the bodies in this zone that I am trying to monitor. Is this correct?nAnd it doesn't matter what I'm trying to monitor if in general the 10% rule is not respected. nIs this reasoning acceptable for my case?nBest regardsnPhilippen