ACI 318-11 13.5.3 - Unbalanced Moment Flexural Reinforcing

[PEinVA]

General question regarding ACI 318 13.5.3 which states the following:

When gravity load, wind, earthquake, or other lateral foces cause transfer of moment between slab and column, a fraction of the unbalanced moment shall be transferred by flexure in accordance with 13.5.3.2 through 13.5.3.4.

I read the previous thread here ( thread726-155029 ) which seems to be similar to my concern.

I'm running a well known slab analysis software and they have made the claim that the statement made above is not required in a finite element analysis and only applies to "manual methods". Now this is not described in any of the literature I found regarding the software. We had to email and ask the question on where it is included in design.

Yet they do include the Munb in the punching shear check, which would be the remaining unbalanced moment.

If the software, as most analysis software, including finite element programs, claims to satisfy ACI 318, and they feel as though this requirement does not apply to them, wouldn't that put them outside the scope of the code or not satisfying the code?

Does anyone know of any recent research into this that may support this view that there is no need for a concentration of bars at the column to transfer this unbalanced moment? We've done our analysis in an old strip based design software and in the finite element software. Flexural moments are within 3-4% points including the reported unbalanced moment, but they do not account for this portion of the code. Strikes me as odd. Maybe its just me, I don't know.

Appreciate your input.



RC

 

[JAE]

The ACI method uses a system to determine the required negative moment reinforcement needed in a slab and then distributes a proportionate amount in the column strip and the middle strips.

The column strips take more moment naturally because the "flow" of bending gets concentrated as you get closer to the column.

However, the column strip is 0.25L wide and I believe that the additional reinforcement needed over the column is there to account for the fact that the column strip distribution is still too wide to account for the focused concentration of load directly at the column.

The alternative would be to have more "strips" in the design (i.e. a "column line strip", a column strip and a middle strip) or even more strips to better distribute rebar where it is actually needed.

In a finite element analysis, it results in a precise analysis of the slab and provides an exact map of the flow of the loading.

I presume that the software places the actual reinforcement needed to account for this column area concentration of loading.
So with FEA, no "additional" rebar is needed because the FEA method doesn't approximate the rebar into wide strips.

 

[PicoStruc]

In my opinion, as engineer, when using FEA software to design slab, you need to apply the load with patterns to generate maximum unbalanced moment at column location.

 

[PEinVA]

JAE - I agree with every thing you have said. The software distributes moment based on strips set forth by the engineer. So it is possible to input the strips as you have indicated. We agree that if there was some distribution based on actual moments found by the more rigorous analysis one could justify the position of the software vendor because it would basically applying the same principle.

However, there is still unbalanced moment that is not being accounted for, specifically at the exterior columns. No where in Ch. 13 of ACI 318 does it limit this requirement to manual methods or to DD or EF and can be ignored if a FEA is performed. Heck, the FEA comes up with the the same damned unbalanced moments as an EF strip based analysis. Obviously we can add the third strip that you mention, which is the same thing as 13.5.3 is trying to accomplish but shouldn't that be part of the software already?

The vendor appears to recognize that they need to account for unbalanced moment as they do it in the punching shear check. However they don't account for any increase within a specific band at the column for this unbalanced moment in flexure. It seems silly to account for 40% of this unbalanced moment in shear then ignore it in flexure? Where does the other 60% go? In rebar you hope is there? You have this fancy, powerful piece of software that seems gimped in some way.

Our problem here in the office is that, like all vendors, they claim this software will design to, and meet specific codes. In our eyes, it appears as though it doesn't. And worst part, it doesn't make it clear either way. I don't know maybe we're being too critical, but we're just thinking about how many "newer" buildings we have reviewed for renovation work where reinforcing was short in many areas, and wondering if people assuming the powerful software is doing too much. Which isn't the fault of the software vendor, necessarily, but it's still a problem that should be resolved.

I was just wondering if there was any research which looked at these issues.

Not trying to paint any vendor or person in a bad light, just trying to talk through the whole process. Don't want to give the impression that I'm being an a$$.

RC

 

[PicoStruc]

What software are you refering to.

I Use CSI SAFE, and all unbalanced moment are accounted for edge, corner and interior column in design for punching and bending

 

[JAE]

RCinVA,

No where in Ch. 13 of ACI 318 does it limit this requirement to manual methods or to DD or EF and can be ignored if a FEA is performed

ACI 318-11, section 13.5.1 refers to "any" procedure that satisfies conditions of equilibrium and geometric compatability and the commentary there refers to FEA (discrete elements).

So I'm guessing here that your vendor thinks this way:

[red]If the column/slab interface is correctly modeled, your FEA analysis can show the build-up of bending in the slab at vicinity of the column.
Thus, the FEA software essentially accounts for 100% of the unbalanced moment via flexure and ignores the fact that the slab/column interface can transfer some (40% ??) of the moment via shear action.[/red]

 

[SethGuthrie]

I would add that many FEA based approaches don't deal with ACI Chapter 13 at all. For typical slabs the reinforcement from the software is naturally clustered over the columns and section (13.5.3.2) rarely controls. But there may be cases where it does control and the engineer should probably double check it. And if the software allows, redistribute the reinforcement to comply and then re-check the model.