Incorrect shear maps when defining slabs with reduction of the moment of inertia

[MartinTrs]

Hi!

I'm using RSA 2018.02 and I've discovered that the shear force maps are incorrect for the slabs that are defined with the use of "reduction of the moment of inertia" option.

To illustrate this problem I've made a simple model with 3 slabs (see the attachment):

1. Thickness 300 mm without reduction of I
2. Thickness 300 mm with reduction of 0,25xI
3. Thickness 189 mm without reduction of I (189 mm has the same moment of inertia as 300 mm with 0,25xI)

All of the slabs are loaded with uniform planar load of 20 kN/m2

The moment Mxx maps and reactions are similar to each other and correspond to hand calculations:

However, the shear force maps are incorrect for the middle slab (thickness 300 mm and reduction of 0,25xI). In this case shear forces are not in correct correlation with moments and reactions:

Best regards,

Martin

[Artur.Kosakowski]

Hi @MartinTrs

This type of results for a panel with reduced Ig is very strongly mesh dependent:

The most accurate values (with little mesh size influence) are these from reactions.

If I managed to answer your question(s) press the Accept as Solution button please. This will help other users to find solution(s) much faster. Thank you.

[MartinTrs]

Hi @Artur.Kosakowski

I don't understand how is it possible to obtain results that ignore fundamental relationships between rotation, bending moment, shear and external loads:

Deflections, rotations, bending moments and reactions are correct independent of the mesh size. However, shear forces are strongly influenced by meshing size such that fundamental relationships are ignored.

Best regards

Martin

[Artur.Kosakowski · Accepted solution]

Hi @MartinTrs

This relationship is 'stored' in the shape functions that describes behavior of surface elements and they are sort of 'calibrated' between 'bending' and  'shear'. When you change one of these components without 'alerting' the other this 'calibration' is also affected and in the consequence the results obtained for shear are less precise. If they are your primary objective (and you can't use the reactions instead) you should consider using the orthotropic panel definition providing the 'coherent' stiffness matrix components instead. 

If I managed to answer your question(s) press the Accept as Solution button please. This will help other users to find solution(s) much faster. Thank you.

[MartinTrs]

@Artur.Kosakowski, thank you for the explanation!