Concrete Beam Question 2

[XR250]

Suppose I have a 24" wide x 12" thick strip footing with a point load from say, a shearwall chord, that can be either up or down. If I design top and bottom rebar to resist the demand, do I need to add ties to confine the reinforcing? The rebar is not necessary to increase the compression face capacity of the beam - it is only there to take care of the stress reversal. Shear reinforcing in not required based on the demand and footing size.

Thanks in advance!

 

[hardbutmild]

I second the retired13's idea. I love talking about seismic and I think I'd have a few things to say.

Almost every two way slab has untied bottom bars running near the columns and you never hear of those buckling out the bottom.

I might be wrong, but it doesn't seem exactly right to me. If you have a two way slab you'd have compression in both directions, right? Wouldn't the compression perpendicular to the reinforcement act as confinement? Some codes even allow reduced anchorage if lateral pressure exists.
Also, another idea cam to me. Since near those columns you usually have beams or hidden beams (don't know if you call it like that in english, basically a beam that's not visible from the outside, large amount of steel in the hight of a plate) wouldn't this buckling redistribute? If plate reinforcement started to buckle, it would lose its stiffness as well as the part of a plate that's heavily cracked (while the beam is confined and has higher stiffness) and redistribute it to the beam.
I might be using the wrong terminology, but the codes might help me. Some codes say that during an earthquake only a small part of plate near to the beam may be considered when determining the T section (effective width). Where high ductility is required, this effective width is as wide as a column. I think this is exactly because of that. Again, maybe I'm using "redistribute" in a wrong way, but I hope my point is clear.

Sorry if I'm not exactly on topic.

 

[KootK]

If you have a two way slab you'd have compression in both directions, right?

Right, uniaxial compression.

Wouldn't the compression perpendicular to the reinforcement act as confinement?

Only in the lateral direction. The lowest level bars are still more or less free to pop out of the bottom without any restraint from ties or confinement.

Since near those columns you usually have beams or hidden beams (don't know if you call it like that in english, basically a beam that's not visible from the outside, large amount of steel in the hight of a plate) wouldn't this buckling redistribute?

In my market it's mostly beamless, flat slabs. Regardless, per the sketch below, the problem still exists (to the extent that it is one).

Again, maybe I'm using "redistribute" in a wrong way, but I hope my point is clear.

I don't see the redistribution helping with the affect that I'm talking about. These are bars not intended to resist any load but encouraged to buckle by what is effectively imposed compressive strain.

 

[hardbutmild]

oOh, okay. Now I understand your concern better. It's usually mesh, right? It's welded with bars in the opposite direction which are because of poisson effect in tension (about 0,2*compression stress, right?). This should stabilize it as I drew in my picture. Now I see that I drew it the other way around than you did, but if wires are welded and the weld can sustain the forces I think it should be okay.

I'm just trying to understand why it doesn't happen because as you mentioned it doesn't. It's interesting certainly.

 

[KootK]

It's always bottom rebar in my area rather than mesh. Agree that, with mesh, you couldn't buckle the lowest layer without also taking the second lowest layer along for the ride. I don't really understand the Poisson part of it though.

I'm just trying to understand why it doesn't happen because as you mentioned it doesn't.

I'm fairly confident that this is why.

3) As for the "why" of it, I think that it simply takes more strain than concrete's 0.003 US limit to induce bar buckling in situations with common densities of rebar and amounts of cover. If your flexural design limited your compression side strain to 0.003 at ultimate, you probably never even see 0.002 there.

For moderate, mostly monotonic levels of strain, the cover itself provides the restraint via concrete's tensile capacity. I'm always drawn back to the fact that it takes next to nothing to restrain an Euler column from buckling so long it's fairly straight to be begin with. I think that concrete in tension is the "next to nothing" in this situation. Sketchy but, apparently, sufficient.

 

[hardbutmild]

Sure, I agree with your point 3). That's certainly true. But I've never heard of buckling occurring in plates even during earthquakes where the part right next to it (grey area in your last post) is at idk., 1% strain.

I don't really understand the Poisson part of it though.

Well if you compress something, isn't it supposed to extend in perpendicular direction? So when you press the section in the red direction (call it x) it'll extend in the blue direction (call it y, z is vertical in this example). I always thought that because of that you need at least 20% of steel in the opposite direction (this is a provision in my code. Since this is approximately the value of the poisson's ratio I figured it's because of that). Was I wrong all this time?

 

[KootK]

But I've never heard of buckling occurring in plates even during earthquakes where the part right next to it (grey area in your last post) is at idk., 1% strain.

Not buckling of the plate, buckling of the rebar. And you'd certainly buckle any steel element at 1% strain as that's about 500% more strain than it would take to yield the entire cross section and reduce its flexural stiffness to zero.

Was I wrong all this time?

I couldn't say as this still appears to me to not affect rebar buckling in any meaningful way.

 

[hardbutmild]

Not buckling of the plate, buckling of the rebar. And you'd certainly buckle any steel element at 1% strain as that's about 500% more strain than it would take to yield the entire cross section and reduce its flexural stiffness to zero.

Yeah, I'm aware that we're talking about buckling of the rebar, but I've never seen concrete cover spall in that area so I concluded that rebars didn't buckle, even at high strains.

I couldn't say as this still appears to me to not affect rebar buckling in any meaningful way.

Well, blue bar is holding the red bar. If the blue bar is in tension it's "stiffer" than it would be if it was at zero stress (this is how all the computer programs define it, tension force increases the stiffness). This means that a red bar has a stiffer support holding it against buckling. It's like having a rope. If you stress it in tension it'll require larger perpendicular force for it to move.

 

[KootK]

Well, blue bar is holding the red bar. If the blue bar is in tension it's "stiffer" than it would be if it was at zero stress (this is how all the computer programs define it, tension force increases the stiffness). This means that a red bar has a stiffer support holding it against buckling. It's like having a rope. If you stress it in tension it'll require larger perpendicular force for it to move.

You're getting too fancy for me. I'm restricting my consideration to non-mesh, lowest level bars under compression.

 

[KootK]

I though of another effect acting to prevent compression bar buckling in flexural members: geometry.