Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Beams subjected to bending and torsion 2

Status
Not open for further replies.

HassanZebari

Civil/Environmental
Apr 2, 2020
16
Hello,

Is there a maximum limit for longitudinal torsional reinforcement in beams? because in some cross sections in beams, i see a very high reinforcement ratio ( due to bending and torsion) which could lead the beam to be over reinforced which we are not willing to have.

So how can we have tension failure in beams that have both longitudinal torsional reinforcement and bending reinforcement?

Thanks
 
Replies continue below

Recommended for you

Is there a maximum limit for longitudinal torsional reinforcement in beams? because in some cross sections in beams, i see a very high reinforcement ratio ( due to bending and torsion) which could lead the beam to be over reinforced which we are not willing to have.

Usually before you get to that level of longitudinal reinforcement.....you've reached the limiting torque as per code. (In ACI 318-11, it's about 4 times the threshold torque (IIRC). See Sect 11.5.3.1) However, the steel is additive (i.e. flexural+long. torsion).....and I am unaware of any limit to this.
 
I'd say that the code upper limit on longitudinal reinforcement is inclusive of that required for flexure and torsion.

Not withstanding that, is the torsional component required for equilibrium (are you stuck with requiring to satisfy the equilibrium torsion requirements), or is it just required for compatibility?
 
This is the catch in ACI, which does not have clear upper limit on reinforcement for flexural members. But one shall stick to the preferred behavior - the steel has to yield before crush of the concrete. I agree with both of above, that bending steel and torsion steel are additive, since the applied forces are not smart enough to distinguish which steel is for which purpose.
 
Agent666 said:
I'd say that the code upper limit on longitudinal reinforcement is inclusive of that required for flexure and torsion.

I don't know much about US codes, but I suppose they don't differ much from european on this point. I'd disagree, torsional reinforcement is equally distributed along the whole section so I assume there's equal amount of torsional steel in tension and compression area. This surely must amount to something. This maximum value is to ensure tension controlled failure, if more steel is provided in tension, steel is needed in compression also. Since for torsion you provide equal amount of tension and compression reinforcement I'd say that torsional steel is irrelevant in this check (of course, constructability is a different question).
 
You are right guys, That's why i thought about this since steel (bending and torsion) are additive and loads can't distinguish between the purposes of placing them.

I think that "hardbutmild" has made a good point about the distribution of longitudinal torsional steel in tension and compression where it can be equal so the extra steel that is placed in tension zone to resist torsion can "cancel" the extra steel placed in compression zone to resist torsion too so we could surpass the over-reinforcement issue for the beam

Thanks for the answers
 
While hardbutmild suggestion is logically sound, I'll be wary on making a conclusion based on that. I could be wrong, but with As the same for each face, will the strain/stress in the cracked tension zone equal to that in compression zone, thus cancel each other? Note that as the concrete has cracked, the stress-strain relationship is no longer linear. I think it needs a closer look/exam.
 
I agree that it needs a closer examination, but that shouldn't be hard. It's a simple doubly reinforced section. I based my conclusion partly on the fact that when doing seismic design, ductility of a beam is ensured if you put in compression at least half of tension reinforcement (at least in europe). If half the reinforcement will be ductile enough for seismic elements, I don't see how a full reinforcement wouldn't be for non-seismic ones.
Additionally, shouldn't the equal amount of steel in tension and compression mean that if concrete is completely disregarded, the section would fail at the same time in both tension and compression? This means that even if concrete crushes, you still have a "balanced failure" (i know i'm using it wrong). Considering that you provide confinement (stirrups for torsion) means that concrete can go beyond 0,35% (or 0,3% whatever's the limit).
It seems like it shouldn't be a problem at all, but as I said, do the doubly reinforced section to figure out the strains, set steel to yield and check if concrete strain is below the ultimate limit for equilibrium. If it is, you're fine. This should be easy.
 
hardbutmild's point was validated through a parameter study. The result indicated, for a given cross section cracked up to the neutral axis, while both the bending and compression steel were doubled (+100%), the concrete compressive block increased at a much faster the same rate (+100%). Thus, the addition of torsional reinforcement, evenly placed on top and bottom faces, has no effect on the ultimate failure mechanism, that is, the flexural reinforcement will yield prior to the crush of concrete.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor