Column strip bottom bars & Integrity Steel

[ma3355]

Hello all.

I have a question (actually need a solution) regarding bottom "integrity" steel provision for two way slab design. As per Figure 13.3.8 of ACI 318, at least two bottom bars in each direction must pass within the column core and be anchored at exterior supports.

However, we recently poured a slab and this requirement was not followed by our field personnel. the rebar was placed as per the required spacing (#5@12" O.C), but some of the columns only had ONE bar going through the core instead of two. Unfortunately, the field-inspector did not realize the mistake until the last minute, but it was too late to do anything.

Now, we need to come up with a solution to fix this issue, but we never encountered this problem in any previous projects. Has anyone ever come across this issue?

I will really appreciate your help with this matter.

Additional Information:
- Total Area of the Slab that was poured = 100,000 square feet.
- Thickness of Slab = 20% is 8” and 80% is 12”
- Number of columns = 300
- Typical Columns size: 18”X24”
- See attached sketch for more information.

 

[JAE]

Tough one.

I suppose you might be able to drill a hole through the column, directly under the slab, and install some sort of supplementary "continuity" rod that could be extended out past the column some distance and bolted/bracketed to the bottom of the slab.

I think the intent of running two bars through the core of the column rebar is to create a pinched-bar condition should a column support fail below....creating a catenary suspension condition with a missing column.

You would have to know the position of the column vertical bars and also possibly allow for drilling in two orthogonal directions if the missing bars occurred in both directions at one column.

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

 

[KootK]

My understanding of the integrity requirement is that the intent is to provide a back up mechanism for resisting punching shear such that a progressive collapse doesn't spread around the floor plan should one connection give way. Unfortunately, I just don't see any practical way for you to get all the way back to full code compliance without getting serious invasive.

One, painful alternative might be to install fire-proofed, steel angle collars at the top of the columns designed to deal with 100% of the punching shear load. Even at that, you'll likely need a dispensation of some sort from the authority having jurisdiction.

 

[JoshPlumSE]

I'm looking at the commentary to the 2014 version of ACI 318. Section 8.7.4.2 (structural integrity).

The continuous column strip bottom reinf provides the slab some residual ability to span to the adjacent supports should a single support be damaged. The two continuous column strip bottom bars or wires through the column may be termed "integrity reinf" and are provided to give the slab some residual strength following a single punching shear failure at a single support. ACI 352.1R provides further guidance on the design of integrity reinforcement in slab-column connections.

It doesn't seem like punching shear strength itself is the focus here. Rather making sure that the slab has residual flexural strength post punching shear failure.

I would recommend taking a closer look at that ACI report 352.1R, it may give some guidance on alternate ways of providing integrity. Off hand, I don't immediately see how missing the column joint slightly really changes the integrity aspect very much. But, I haven't ever focused on that aspect of the code before.

 

[Tommy385]

It doesnt influence punching shear strength, this reinforcement ist there to prevent immediate colapps after the punching shear failure through catenary action (for a while..)

 

[ma3355]

JAE / KootK / JoshPlum / Tommy385.

First of all, thank you for responses.

As stated on your replies, this has nothing do with other structural requirements. The only purpose of having these bars going through the core of the support is to provide the slab the ability to span to the next support in case a single column be damaged.

We are still looking into trying to bring the slab into compliance, but it seems like the fix (scanning, identifying location of bars, cutting concrete, drilling new bars through column, etc) can be very costly and time consuming.

 

[haynewp]

It looks like something that would happen on one of my projects. I would contact Quakewrap for progressive collapse retrofit options. Simpson has also gotten into PC retrofits.

 

[KootK]

As stated on your replies, this has nothing do with other structural requirements. The only purpose of having these bars going through the core of the support is to provide the slab the ability to span to the next support in case a single column be damaged.

This certainly wasn't stated in my reply. I believe your statement to be fundamentally in error. And, as I see it, you stand no real chance of meaningfully rectifying your problem unless we can straighten out the integrity steel misconceptions presented in this thread that have culminated in this error. So I'm going to have to change some hearts and minds here.

To restate my opinion as clearly as possible, I believe that the intent of ACI 318-14 section 8.7.4.2 is twofold:

1) Continuous bottom bars are provided through the column cage so as to provide a secondary mechanism of resisting punching shear should conventional means fail AND;

2) Continuous bottom bars are provided throughout the column strips to provide for some caternary resistance should #1 fail and column support be lost altogether.

Put my level of confidence in my interpretation around 95%. Punching shear resistance is very much the focus of the integrity steel provisions dealing with continuous bottom bars passing through column cages.

As evidence of this, I submit the following:

1) See the attached PCA document dealing with integrity steel. In particular, not the paragraph clipped from that document and shown below. Sadly, like the source provision, the exact wording hasn't quite been articulated in a way that precludes all possible avenues of misinterpretation. That said, I still very much read that as supporting my view.

2) If the two bars going through the column were all about the caternary action, why would it matter that they pass through the column cage? The column no longer exists as a slab support when that caternary action would be engaged.

3) When ACI 352.1-11 discusses slab to column integrity recommendations, both the form of the equation and the supporting sketch suggest that the force being dealt with at the connection post-post failure is the gravity load being delivered to the column by the slab via punching shear. Nothing about this information suggests that caternary action is being considered in the calculation.

@ma3355: any chance you have stud rails at your columns? If so, it may be feasible to count the bars placed over the rails as substitutes for the bars placed through the column cages. ACI 352.1-11 suggests a similar approach where shear heads are used. Of course, shear heads aren't often used but I would consider stud rails to be substantially similar in concept for this purpose. Obviously, this would be a great "do nothing" fix if you've got those stud rails.

 

[JAE]

KootK,
I guess I've always looked at the bar-through-the-core requirement emanating from needing a solid end-anchor for the catenary pull since inside the core you have a significant pinching action on the two bars.

The idea is that if you have a two span condition with columns 1, 2, and 3 supporting a slab - and column 2 fails for some reason (a bomb like at the OK City collapse)- then the bars going through the cores of columns 1 and 3 are pinched by the natural axial load (as opposed to adjacent column strip bars that might loose their natural bond if the slab itself is dislocated and cracked significantly.

This was in a lot of the pre-code approval commentary for ACI 318 after the Oklahoma City Murrah Federal Building collapse.



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

 

[KootK]

With the usual respect, I believe that your interpretation is in error JAE.

- the math in the calculation snippet that I posted clearly deals with the gravity load at a single column. Nothing about that suggests caternary.

- where there is caternary action, I believe that the bottom bars doing the carternary-ing would be being supported by the column strips running the other way.

- numerically, if any significant amount of the caternary action depended on just those two bars running through the columns, the situation would be utterly hopeless.

 

[KootK]

@JAE:

To clarify, I don't doubt that the integrity bars accomplish some of what you say they do. I simply do not believe that is the spirit of the integrity provisions as written and intended.

I've found that, both on this board and in the US in general, these provisions are frequently misconstrued. I did my first decade in the US and misconstrued them myself while I was there. There's just something lacking in the presentation that seems to cause confusion.

I didn't get myself sorted out on this until I returned to Canada where, frankly, the presentation and understanding of this stuff is much better. Or at least it was. ACI 352 seems to have caught up with the Canadian stuff now though which is great. The last step will just be getting the word out. It's easier in Canada as the related ACI 352 provisions are part of the main body of the concrete code in Canada.

In Canada, some firms will do the column integrity bars in drop panel systems as non-continuous bars located at the bottom of the drop panel as an aid to constructability. Philosophically, that detailing jives with punching shear treatment and would be utterly inconsistent with caternary treatment (at least in the absence of some seriously unpopular shear ties).

This is not to say that everything that is written and done in Canada is the right way to go about things in the US or anyplace else. Given the significant parallels between the code provisions, I think that Canadian practice is germane to the discussion however. In all likelihood, the provisions probably got their start with ACI and Canada just rocked the explanation part when we stole them and had trouble understanding them.

 

[MIStructE_IRE]

Just thinking out loud.. Is this a lower level in a building? Ie are the more floors above? If so, could you compensate for this at a higher lever so that in the event of a column loss, this floor could hang from the upper floors and avoid disproportionate collapse?

 

[KootK]

If so, could you compensate for this at a higher lever so that in the event of a column loss, this floor could hang from the upper floors and avoid disproportionate collapse?

In my opinion, no, you couldn't. A slab can't hang from a column with the slab to column shear transfer mechanism compromised. You might be providing more of a different kind of robustness but you wouldn't be directly addressing the intent of the continuous bars passing over the columns.

 

[JAE]

KootK,
All good points and I think your are correct on the punching shear issue.

But I think there are really two scenarios here and much of the literature out there supports both.

I think you are talking about the concept of the slab failing in punching shear around my column 2. With that event, I think all your points are valid. The continuous "integrity" steel is there to avoid a failure in the rebar at that specific column to allow tensile membrane action to occur and prevent the total collapse of the slab.

But another scenario - where you have the total loss of column 2, say from a failed column below the floor (unrelated to the floor or floor-column connection), there is no punching shear happening at this story and the tension membrane action is fully utilized here to avoid progressive collapse - anchoring of the bars at columns 1 and 3 are essential.

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

 

[JoshPlumSE]

2) If the two bars going through the column were all about the caternary action, why would it matter that they pass through the column cage? The column no longer exists as a slab support when that caternary action would be engaged.

Yes, this is exactly the trouble I have interpreting that code provision and the commentary. If I had a ton of free time this week, I would dig into that ACI report, 352.1R. Though I think the OP probably should do so. In lieu of doing it myself, I trust that KootK's probably got a pretty good grasp of it if he's really reviewed it.

In Canada, some firms will do the column integrity bars in drop panel systems as non-continuous bars located at the bottom of the drop panel as an aid to constructability. Philosophically, that detailing jives with punching shear treatment and would be utterly inconsistent with caternary treatment (at least in the absence of some seriously unpopular shear ties).

I wonder if you could create a column collar or capital on the underside of the slab so that the bars would be continuous through the punching shear zone of the collar / capital. That might be a cheaper option than some of the others.

 

[MIStructE_IRE]

“I wonder if you could create a column collar or capital on the underside of the slab so that the bars would be continuous through the punching shear zone of the collar / capital. That might be a cheaper option than some of the others.”

That’s the only “out” I can see here really.. But it won’t come cheap over 300 columns..

 

[KootK]

But I think there are really two scenarios here and much of the literature out there supports both.

I agree. I would say that there are actually three scenarios at play that may or may not be related:

Scenario 1: Conventional punching shear fails at column 2, the backup punching shear mechanism of the "two bars" kicks in, THE END. No caternary action required.

Scenario 2: Conventional punching shear fails at column 2, the backup punching shear mechanism of the "two bars" also fails, and caternary action out to columns 1 & 3 kick in.

Scenario 3: No conventional punching shear failure at column 2, column 2 IS eliminated as a support Oklahoma style however, caternary action kicks in between columns 1 & 3.

I think you are talking about the concept of the slab failing in punching shear around my column 2. With that event, I think all your points are valid.

That is indeed precisely how I've been thinking of it.

The continuous "integrity" steel is there to avoid a failure in the rebar at that specific column to allow tensile membrane action to occur and prevent the total collapse of the slab

I'm not sure that I agree. Or, at the least, I might phrase it differently. During this event, I would say that the "two bars" are there the provide a secondary punching shear resistance mechanism which, if effective, would prevent caternary action from being required at all. This would be my scenario 1 above. Are we on the same page with that?

But another scenario - where you have the total loss of column 2, say from a failed column below the floor (unrelated to the floor or floor-column connection), there is no punching shear happening at this story and the tension membrane action is fully utilized here to avoid progressive collapse - anchoring of the bars at columns 1 and 3 are essential.

Agreed. This would be my scenario 3 above. And, while it had not occurred to me previously, I do see the logic in the bar pinching at columns one and three. That said:

1) It's easy to miss the pinching business as this aspect of things does not seem to manifest itself in the associated calculations. Of course, just because a particular phenomenon isn't required to be calculated explicitly, that doesn't mean that the detailing isn't meant to address that phenomenon intelligently anyhow.

2) Were one to calculate the tensile demand on the "two bars" under canterary action, I suspect that those bars would be proven woefully inadequate for the purpose of caternary action. The bars provided by ACI 352 would get you closer to good but still a long ways off I'd think.

3) As we've discussed, I have presumed that the "caternary bars" running from column 1 to 3 are really supported by the perpendicular column strips at columns 1 & 3. If that presumption is incorrect, then what are we to make of the fact that most of our continuous, caternary action bars are not located over the columns? Or are we saying that both mechanisms -- two bars & perpendicular column strips - are contributing and that the two bar mechanism is simply better served with the pinching business in play?

 

[KootK]

I wonder if you could create a column collar or capital on the underside of the slab so that the bars would be continuous through the punching shear zone of the collar / capital.

Clearly, great minds think alike.

One, painful alternative might be to install fire-proofed, steel angle collars at the top of the columns designed to deal with 100% of the punching shear load.

The only difficulty that I see with that is that I believe that the integrity steel provisions specifically ask for the two bars to run not just over the columns but, rather, through the column cages. My presumption is that is desirable in order to keep the slab hanging bars from slipping off the side of the column when the punching shear zone cracks up terribly. With the collars, I'm not sure what would perform that same function.

any chance you have stud rails at your columns? If so, it may be feasible to count the bars placed over the rails as substitutes for the bars placed through the column cages. ACI 352.1-11 suggests a similar approach where shear heads are used.

There may be a way to make a go of this even without stud rails. Hilti has a drilled in punching shear reinforcement solution that could get you back to a ductile shear resistance system similar to the stud rails and/or shear heads. Still not cheap but at least possible as a means of being able to count slab bottom bars outside of the column cage.

 

[MIStructE_IRE]

I hope the young field inspector isn’t beating himself up over it... We’ve all been there at some point! Perks of the job!!

 

[MIStructE_IRE]

I’m curious - how did you resolve this one in the end?