Differences in Determining Internal Forces of Downstand Beams

[Dlubal Software]

When performing control calculations and comparing the internal forces and the resulting required reinforcement of downstand beams, it might happen that large differences occur. Although the same load assumptions and spans are applied, some programs or the manual calculation display very differently internal forces compared to the FEA model. The differences already occur in the case of the centric member and without considering the internal forces components from the possible effective slab widths.

[Milos]

I am trying to verify similar model, but timber beams and clt slab using line release. Thanks for a piece of advice.

[Frank Faulstich]

Hello Milos. I'm not quite sure I understand the problem. Would you expect the same bending moment in member 2 as in member 7?

In any case, it is necessary to refine the FE-mesh.



Frank Faulstich

[Milos]

Hello, thanks for your reply. Yes, I expect the same bending moments and I get same bending moments, but not the same reaction forces due to convergence issue. I'm probably doing something wrong with line releases I will appreciate if you can help to identify my mistakes. I would like to transfer only vertical forces from slab on my timber downstand beams. I attach two very simple models which I am trying to verify.
Best regards,
Milos

[Frank Faulstich]

Hello Milos. With the file "Downstand beam analysis" you have to consider that it is a statically indeterminate system. The basic rule for such systems is that stiffer parts attract forces.

The plate is supported at lines 2 and 4 with line supports that are rigid in the Z-direction. The support in the middle is a rather flexible beam. The beam withdraws the force and the rigid supports attract the forces.

If you increase the stiffness of the beam, it also attracts larger forces.



The file "Timber Downsatnd beam" is unstable, because the surface S1 is freely movable in X and Y direction due to the line releases. As this is a pure plate load and the members M8 and M2 are centred, the line releases are not necessary.

I have deleted the line releases. Then the system is no longer unstable.

The bending moments are then 0.03 kNm and thus only half as large as in the M7 member.



The reason for this is similar to the one in the first file. Also the plate S1 has a stiffness in X-direction. Therefore it also transmits a part of the forces in this direction. If you increase the stiffness of the bars, then they also transfer a larger part of the forces.




Frank Faulstich

[Milos]

Thanks a lot.
Milos