Equilibrium and Compatibility Torsion

[drasticxxxx]

Equilibrium and Compatibility Torsion

I have been reading in this topic a lot just to know when I can ignore torsion, and anyone can say when you compatibility torsion you can ignore it but this questions will jump to my head:
1. If the secondary beam is totally pin thin I can say that primary beam has no torsion at all and I can easily say no need to design torsion, and I found the ACI allowed for redistribution of the moment based on the reduction of allowable compatibility torsion (So I would like to ask as a designers is the general trend is to ignore compatibility torsion or what? If I want to answer this question, I will say it depends on your assumptions, if someone assume slab or secondary beams edges are pin and use steel details to support this then yes, but if someone assumed fixed then ignore torsion this will increase the moment in the center of secondary beams and cause failure.
So what is your comment guys?

2. I read in one book the beams carrying precast slabs in two edges is considered equilibrium torsion ,but I can’t understand, Since I know cantilevers are considers under equilibrium torsion but this case is not clear ,so please clarify it

3. How can I make pin connection in reality ?

 

[drasticxxxx]

for more understanding can anyone tell me why these 3 system, consider equilibrium torsion(actually i can understand the third one, because if no torsion in the beam the canopy cantilever will fail ,but the first two are not clear

http://files.engineering.com/getfil...e3-ffa60c44a7f8&file=11111111111111111111.PNG

also please see the drawn case below and judge??equilibrium or not and why?

http://files.engineering.com/getfil...c1-4614-be85-572ff93a31f7&file=2222222222.png

 

[sponton]

You can 'ignore' it in beams that have geometrical and similarly loaded adjacent bays, in other words where torsion balances [has a net effect of zero]. You shouldn't ignore it in spandrel beams since these are loaded only in one side.

If you're talking about steel, even if you have a shear tab connection [pin] to only one side of the section's web in reality there is torsion but this is minimal since it's basically the reaction transmitted from the weld rather than the center of gravity of the connection [See Salmon's Steel structures: Design and Behavior Chapter 8.6- Practical situations of torsional loading] . It's hard to really get an exact number in such cases since there are many factors that come into play [floor slabs, joists, anything really that stabilizes your section so that twisting is prevented]

2. If you're loading it in both edges that mean the stiffness of the slabs will prevent twisting, a related idea i would say is when you have a concrete floor on top of a steel beam you can assume that it is continously braced in the sense that either through friction or actual embedding [or shear bolts] will prevent movement of the flanges relative to each other.

As for the canopy, of course you would take it into account and also you'd make sure you use closed sections [HSS, TS] which are better at handling torsion. Other cases where you should take into account torsion would be beams loaded with considerably eccentric loadings [beams supporting masonry through a bottom plate, canopies, etc.]

3. You can't, you'd have to have actual frictionless pins which i am afraid do not exist in reality. All connections offer some degree of moment restraint, which is good.

 

[hokie66]

You can often ignore compatibility torsion in steel beams due to the ductile nature of steel as a material. The carrying beam twists ever so slightly due to deflection of the carried beam, but there is nothing detrimental about that.

In cast in place concrete construction, you don't ignore compatibility torsion in spandrel beams, but rather provide nominal torsion reinforcement as prescribed by the codes. For precast construction, it is common to have equilibrium torsion issues which require more than nominal torsion reinforcement.

 

[drasticxxxx]

Thank you for your respond.

Can you see the second post, the beam I mentioned is carrying precast slab not cast in-situe ,and the book is saying it is equilibrium torsion ,cant understand your respond in this issue
also in the second post there is cantilever with torsion force from point load is this equilibrium torsion as well ?

Please see the second post and give me your feedback,because I am little bit confused,and the answer of this second post will make things clear.

 

[hokie66]

First post, second post, I am not sure which is which. But in the one where you show three pictures, the cantilever with offset force is definitely equilibrium torsion. If you released the torsion at the support, it would rotate. The canopy also represents a case of equilibrium torsion, as without the resistance of the beam to twisting, the whole thing falls down. The other thing, the precast slabs sitting on the inverted tee beam can be stable, but depends on the loading. If the loading is eccentric, then torsion is required to restore equilibrium.

Not sure what the other picture represents.

 

[drasticxxxx]

Now it is start to begin more clear,one last comment to finish the puzzle.

http://files.engineering.com/getfil...c1-4614-be85-572ff93a31f7&file=2222222222.png

The cantilever here with back span,consider compatibility or equilibrium?,I am saying this because if cantilever set with no torsion stiffness the whole beam with the back span will rotate

Thank you

 

[hokie66]

Not sure I have an answer for you on that one. It would depend on the support conditions as to whether torsion is involved. As drawn, all the supports are pinned, and the structure is stable.

 

[drasticxxxx]

Thank you for your help,I really appreciate.I am really looking to hear form others about there comment on this beam,and the stability of back span cantilever if the long span beam cause torsion in the cantilever.

I also check with our senior engineer in the office,and he is telling me that we always consider compatibility torsion modifier in etabs 0.1,but as mentioned above or as mentioned in the code it should redistributed and then be design on the capacity(So,there is contradiction,thus my explanation is usually compatibility torsion can be reduced by redistribution ,and hence the ACI code allow for reduction ,while Etabs only compare with threshold and don't redistribute ,so maybe they usually change modifier of torsion to 0.1, and this maybe be analogous to redistribution)

Finally is reducing torsional stiffness to 0.1 has an effect of choosing R in Earthquake design.

Sorry for my many questions,but this is because I am really thirsty to know for some questions in my mind

 

[rapt]

In Concrete, If the torsion is being relied on to provide structural stability then you must design for it. If you ignore torsion, then you must redistribute its effects back into the remainder of the structure so that the structure is still in equilibrium. As well, you must provide minimum torsion reinforcement in the members where it was ignored to ensure that the members still performs ok under formal flexure/shear loadings.