Beams joining at an angle 4

[ruane]

Good day all

I've got a question which is a bit difficult to formulate and hopefully isn't stupid, but here goes...

If I have two beams that join at an angle in plan, can I transfer moment between them where they join? Say for example I have two beams each supporting a floor load that join at a point which is unsupported. Seen in plan the angle between beams is 110 degrees. Will there be moment transfer between the beams at the point where they join? In my head I can't see how the forces will "change direction". If you have a straight beam and you imagine the forces at midspan, they run straight from one beam to another. A straight beam (two pieces joined at 180 degrees) can be simply supported and loaded. If you try and simply support two beams joined at an angle (without torsional restraint at the support) they would just fall of the supports.

My question is: to what degree can transferring of moment between beams at an angle (in plan) with concrete and steel actually be done.
Is there an angle where the bending moment at "midspan" (the point where the tow beams meet) is so unreliable that the beams should be cantilevers fixed at the support or have sufficient torsional restraint.

I've attached a sketch. The angle isn't drawn correctly - it should look more like 90 degrees.

Thanks all

 

[Agent666]

Any moment in one beam that develops turns into torsion in the other beam. How much moment that develops depends on the torsional stiffness and length of the return beam, so if they are open sections then potentially not very stiff torsionally so relatively little moment/torsion will develop. Develop a moment on one beam and it wants to twist the other beam in simple terms.

It's not stable unless at least one of the beams cantilever from the previous support, unless I'm misunderstanding the configuration.

 

[KootK]

I agree with A666 that, in the absence of alternate torsional restraint, moment transferring from one beam into the other transmutes itself into a combination of strong axis moment and torsion consistent with vector decomposition. In most practical cases where you'd do this, however, there will be alternate torsional restraint. It's important with these systems to keep your wits about you with respect to loads tending to following stiffness and the realistic torsional stiffness of the various parts of your system.

 

[TehMightyEngineer]

In this fun thread:

https://www.eng-tips.com/viewthread.cfm?qid=443393

KootK pointed out that the above building is essentially "a giant, single bay moment frame with a kink". I would imagine that a variation on concrete moment frame detailing could be applied here to ensure continuity; with the above comments by KootK and 666 regarding the torsional forces created.

Ian Riley, PE, SE
Professional Engineer (ME, NH, VT, CT, MA, FL) Structural Engineer (IL)
American Concrete Industries

https://americanconcrete.com/

 

[MIStructE_IRE]

I don’t believe that you can properly transfer the moment around the corner like that with a concrete beam. When it cracks I don’t think that anyone in this world can predict what will really happen in that case!!

I would probably look at cantilever options, hanger options, additional columns or flat slabbing it in some way (although not sure of the rest of your layout).

 

[BAretired]

There is no reason why the plan shown cannot work. If the geometry is entered into a 3D frame analysis program, the required resistances would emerge. It may not be easy to satisfy the analysis with a choice of members and the problem of transferring moment around the corner may be difficult to achieve but a theoretical solution is clearly possible.

An additional beam such as the one shown in Kootk's second sketch would reduce the problem of moment transfer considerably and may be more acceptable architecturally than some of the ideas suggested by MIStruct above.

BA