Warping torsion restraints

[Dmitry]

Hello, collegues.

I Have a few clarifications about ends and intermediate restraints in "warping torsion" module.

I want to calculate beam which is a part of the multi-storey frame. This beam has rigid ends restraints and two intermidiate restraints - secondary beams and horizontal bracing system. Secondary beams are restricting main beam out of plane and bracing system is restricting main beam in longitudal direction.

How to define restraints correctly?

1) I see here https://www.dlubal.com/ru/podderzhka...-znanij/001417 ends restraints of the main beam must be hinged in my case. In addition one of them must be free to move in local x direction.

2) Should I restrain main beam only out of plain or it should be restraned exactly like in the frame? I'm asking that because internal forces in module "warping torsion" are different from internal forces in "steel" module in so doing, main beam restrained exactly like in the frame.

3) How to define intermediate restraints in case of continious beam which has vertical V-bracing in the middle? I guess it should be restrained in longitudal and verctical directions.

Thank you for your time.

[Frank Faulstich]

Hello.

Can you please attach the file (without results) so we can analyze the problems.

Frank Faulstich

[Dmitry]

I can not upload file with format *.rf5t. *.rf5 should be rename to *.rf5t.

[Frank Faulstich]

Hello Dmitry,

thanks for the file. In the model file you can see that you are looking at a girder that has multiple eccentric supports. The support is arranged in such a way that, due to a horizontal beam deformation, additional loads are introduced into the beam itself via the support structure. This "circular reference" cannot be taken into account in the 7 dof degree of freedom calculation. The only change is to find a semi-equivalent support for the system, which, due to the load, produces internal forces similar to those in the overall system.

I can give you the following feedback on your questions here:

1. Yes, since a bending moment is transmitted via the frame corners in the frame plane, an correponding release is required here at the supports. To map the very slightly step-shaped normal force curve, the support must be supported on one side against the normal force and opened on the other side. The step-like normal force distribution then results from the discrete application of normal force and not due to the integral lateral support.

2. Yes, you have to support the girder from the supporting plane in order to take the Mz, Vy and Mt gradients into account correctly. In a cross-check, the "linear" 7 dof internal forces should be similar to the input internal forces.

3. A simple answer is that you need to apply equivalent spring stiffness at these points. Since the determination of this spring due to the integral connection of the girder actually to be dimensioned with the braching system makes this determination difficult, I propose an empirical specification here until the forces are similar to the input.

You can find a suggestion in the following file

Frank Faulstich

[Dmitry]

Thanks for explanation!