Requirements for Concrete Bearing Wall

[bookowski]

As far as I can tell ACI is pretty sparse on designing a load bearing concrete wall other than treating it as a column. If I want to design a concrete wall with untied reinforcement, say a single centrally placed curtain of steel, what is the way to approach this. The steel would not be counted in compression, that part is clear. Other than that can it be treated as a column with Ast = 0. There are the empirical provisions for walls in Chapter 14 which produce a much more conservative result - maybe this is appropriate as there is no confinement of the concrete.

I am looking at a layout where the architect would like very thin 'columns' and I can make them as long as I want and relatively closely spaced, therefore the load is relatively small. It seems that thin long walls placed in the partitions should not be a problem (similar to load bearing cmu construction) but I want to make sure I'm not missing something or violating code. Is there a minimum thickness? Any reason that a 6" or 7" wall is not possible if the calcs work (including slenderness effects)?

 

[KootK]

Wallumns! Is this a multi-story building with concrete floors? I've seen a lot of hotels in eastern Canada built this way. The minimum size that I've encountered is 8"

- The walls can be treated as columns.
- If your reinforcing is kept beneath a certain level (1% is it?) you can omit ties.
- There's a minimum thickness that is a ratio of the unsupported height. In Canada, it's h/25.
- If you're supporting concrete slabs:
---watch out for punching shear at the wall ends.
---keep an eye on in-plane moment.
---the walls will participate laterally no matter what you tell them not to do.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[JAE]

I think KootK's responses are correct.

For ACI, if you are less than 1% reinforcement then you can omit ties.

I have designed numerous tilt-up wall panels (5 to 8 inches thick) - some with single mats of rebar in the center.

The key with thin wall designs is that your second order effects can be quite significant so looking at PΔ and Pδ effects are necessary.

Once you include cracking, reduced Ie values, and second order effects to get Pu and Mu values along the height of the wall you can then check
the design...we use SPColumn but I'm sure other methods/software/hand calcs would work.

One thing I also do for walls - not mandated directly by ACI - is reduce my phi factors by 0.1 just because of the lack of ties.



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[TLHS]

I don't know about other codes, but the Canadian code has a bearing wall section that actually defines the requirements, as long as the bending is low enough that the force resultant falls inside some footprint.

 

[bookowski]

Thanks for input. Yes, multi-story with flat slabs. 15 stories but I'm guessing this scheme will only work towards the top and then they will have to progressively fatten up, likely becoming tied at some point down the bldg. H/25 is generous enough, that would allow me 5" which I don't think I'd do even if it works on paper.

Kootk - How do you check punching shear just at the ends, isn't the punching shear calc usually performed over the entire section? Similar to a P/A + Mc/I type check.

When providing a single curtain of reinforcement is it typical to run the horizontal down the middle and then stagger the vertical every other side to get a little more 'd'?

What min. thickness would allow you to put ties - I'm think at about 9" you could get ties in there (with two curtains of reinf).

 

[KootK]

The punching shear issue is a matter of some debate. Some do in fact do what you've proposed. It doesn't jive well with testing though. Eurocode has provisions that limit the shear perimeter to a three sided thing running 1.5 slab thicknesses along either side of the wall. It's pretty punitive. I've been putting stud rails at the ends of some of my walls lately.

The shear that you need to deal with can be fairly sensitive to your modelling assumptions. At a minimum, I've been assuming that any load coming in from beyond the ends of the walls is dealt with at the ends of the walls. When it's all said and done, I sometimes get better results in unit demising walls by decomposing the walls into two high aspect ratio columns at the ends. 10"x36" or something like that. It works well for punching shear, is predictable in terms of load distribution, doesn't mess with lateral much, and saves a little concrete. Just an idea.

I've never heard of staggering the vertical bars but it's a clever idea. I'm not sure how big of a constructibility issue that would be.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[hokie66]

Ties are not common in walls, and are best not considered. A problem with very thin walls is getting the concrete compacted. With bars in the middle of a 6" wall, there is not much room for a vibrator. Even with an 8" wall reinforced in both faces, the horizontals need to be in the outer layers to allow adequate space for compaction. Allowance for adequate compaction will prevent your next question...what do we do about honeycombing?

 

[JAE]

When providing a single curtain of reinforcement is it typical to run the horizontal down the middle and then stagger the vertical every other side to get a little more 'd'?

You wouldn't want to do that. Either go single mat or double mat.
Staggering vertical bars is difficult and you are essentially using a double mat of bars anyway as they need to have horizontal bars to tie to for stability anyway I'd think.



Check out Eng-Tips Forum's Policies here:
faq731-376