Shear Punching for Slab to Column Joint Vintage 1980 1

[tomzeb]

I'm looking at a condo building in Florida with drawings dated December 1979. It has post-tensioned 8" thick slabs (9 stories) held up by columns and some interior shear walls (no under-slab beams). A detail of a typical interior column shows some rebar and banded post tensioning tendons going through the column at just under the top surface of the slab. No vertical stirrups are shown for improving the shear punching capability at this column/slab joint, although the column is integral with "non-bearing" masonry (8" cement block) walls extending on either side of it. No credit is taken for the support from the block walls, since they were called out as non-bearing in the design drawing. Calculations show that a vertical load of 65500 lbs (DW + LL + slab) from a contributing area of 15x20 ft. goes to an 8x36 inch edge column, which results in a demand to capacity ratio of over 2.

My questions are:
(1) Did the designer just miss this shear issue, or has the code changed leaving this building in jeopardy?
(2) Is shear punching being found to be a problem as people start looking at older buildings?
(3) For a post tension system designed in 1980 did the code provide details for shear punching analysis at the "critical" section as it does in later versions? (Also note, Edge columns are supposed to be designed without considering post tensioning forces.)
(4) Could the designer have used the presence of the masonry walls to justify this connection as a one-way slab?

 

[canwesteng]

That's frightening that Florida would ding someone for using loading from the current code. Beyond the fact that it's strange that the board would even take a stand on that, are they saying you could be disciplined for participating in a seismic retrofit? Taking to the logical extreme, would they suspend someone for saying that an unreinforced masonry structure from say the 20s or 30s is unsafe?

 

[rapt]

Laxpatrie,

Yes, AS3600 also has stiffness reduction factors for lateral load analysis.

But not for Gravity Load analysis, they are specifically for Lateral Analysis. It also says that the worst case scenario should be used for each design condition. 0% stiffness does not provide the worst case scenario for Punching Shear, no matter how they try to twist the interpretation of the code.

But engineers continually refer to those factors for Gravity Analysis. And some use even less (0%).

It is common practice in Asia to design slabs and beams assuming columns are knife edge supports. In most cases that is conservative for Flexure (not always). But it should never be used for Punching Shear.

 

[lexpatrie]

Canwest read the disciplinary action in detail, there's more to it than that. One issue was a forensic report on the wind resistance saying there was a design defect and rather than evaluating as to the code at the time of construction, the report considered the "current" code, (I'm not fully convinced that made it fail and the older code made it pass) and there were also some other mistakes/decisions (exposure category? duration of load? a shim being presumed...) that rather bled into a lack of competence/care in performing the calculation, some misrepresentation of the number of window tests observed/performed, etc. So it's murkier than that.

I found it instructive because it kind of advocated an existing building as far as defects should be evaluated on the code in force at the time. If you know the legal field, something will appear to set a precedent, but it sets a precedent in that specific context, so that's not globally saying regardless what you're doing, you have to follow the original design code, it means when you're claiming something is defective, you should be basing off the code at the time of construction.

I would think it somewhat follows that a) if you are evaluating a design, say for 40 year recertification, you'd look at the original documents and codes for evaluating that design. Awareness of the changes in the code since then is important, and repair or remediation of any defects would be another issue, generally involving the current code.

@rapt - the stiffness reduction I mentioned is at 6.6.3.1.1 (ACI 318-19/22), for "factored load analysis" that doesn't really restrict it to lateral only. Even so, 0.70Ig for the columns and 0.25Ig for the slab doesn't exactly make the column less stiff, it actually makes it stiffer than the gross section analysis, (by 0.7/0.25) so the column would draw more moment due to stiffness? I mentioned this because there'd been some discussion of reducing the column stiffness without any calculations and I thought it might have originated with that code item (even if it's not appropriate).

Regards,
Brian

 

[canwesteng]

What is even the point of recertifying a building if you are just going to analyze to old codes? If you don't care if the building is deficient or unsafe, just leave it as is. Certainly quite puzzling.

 

[tomzeb]

The state of Florida requires recertification of buildings at the 40 year mark as a result of the Surfside collapse. So, the do nothing option is not available.
I think Brian's point wss related to calling something defective; if a building passed requirements from a code many years ago, but fails to meet current code requirements, it is probably not technically "defective". Although, if shear punching is involved, it definitely seems like the building is defective. Since, actions need to be taken to improve the safety of the structure. Not sure how this would play out legally. Anyway, when going back 40 years, there is probably going to be cases like this.

 

[lexpatrie]

The 40 year recertification is due to a Federal building in Dade county (after 1997: Miami-Dade county) collapsing, like in 1974, at the 39 year mark?, not Champlain Towers South, in the Town of Surfside, FL. I want to say John Pistorino had an article in the FBPE newsletter on it some years ago. Yeah, I got that right.

https://fbpe.org/a-look-at-building-recertification-in-south-florida/

Hypothetically you can find a deficiency in the structural system (lateral or gravity) in the original design, based on the original code, which would then be corrected during recertification (I wouldn't advocate "fixing" to a 40 year old code, that's a different thing, it's for diagnosing issues). What one does with concrete columns that are beyond 4% or even 8% vertical reinforcement ratio, well, I don't know. I guess for one you would have them scanned/sounded for voids as the reason for the 4% limit is to allow splices and still get reasonable workability to the concrete through it so there are no voids. (I thought there were vibrating forms at one point, but I don't know when they were first done.) Anyway, not really relevant here. (I suppose I'd entertain load testing, if I had to).

Slab punching shear could be dealt with, as we are discussing here, so I'll try not to divert the conversation.

You could also find "easy" stuff, like the slabs are too thin to omit the deflection checks, slab rebar is too low, too high, too little, incorrect cover, etc. Drafting errors, a lack of shear walls, all manner of things. A penthouse that's not on any approved plans? A deleted column, a missing beam?

Regards,
Brian

 

[lexpatrie]

Side note - I wasn't doing design work in the 1980s, but when I was learning it in school (later) 2D analysis via something like SAP was, and you'd lay out the entire structural systems side by side and add "rigid links" between them, so they'd all deflect laterally together, as a sort of implicit rigid diaphragm approach. These might have used large areas and small moments of inertia to prevent taking moments from the frames, but I'm going off blind recall.

So if you try to replicate design forces, a 3D analysis might not return well, it'd be state of the art for now, so to speak, but wouldn't necessarily fully replicate the original design process.

I have no idea when Adapt-PT became a thing, but first I saw it was around 2000.

 

[lexpatrie]

I had a moment of clarity, so I think the program we had for concrete design in Uni was Sap 90.

Wait, WHAT????

https://archive.org/details/csi-sap-90-v-5.40

 

[tomzeb]

In order to improve shear punching without interior drop panels, would construction of a column on an outside end wall (shown below) of a multi-story building be doable? Walls are 8" CMU sitting on the slab. Removing the CMU at each level gives access to the edge of the floor slab. Adding the column decreases shear/moments on adjacent columns which have shear punching issue. Incrementally forming and pouring the concrete to create the column and getting concrete under the slab at each level may be the challenge. Thoughts?

 

[lexpatrie]

Side note - I can't edit my previous post where I couldn't find the code archive for Broward. Here it is now:

https://www.broward.org/CodeAppeals/Pages/Library.aspx

Regards,
Brian

 

[lexpatrie]

I'm not convinced there's an awesome solution to dealing with this issue, "anything" could work and if it's done reasonably founded in principles of mechanics, it won't make it worse, it feels like someone's Plan B Master's project, honestly.

Are you intending the column and footing to be "cut in" to the structure , so this column going to be floor-by-floor inserted with the slab concrete continuous through the slab-column joint? I suppose settlement could influence things, but you should still get some force transfer "at ultimate". Maybe the "nonshrink" grout could be an element, (it expands then shrinks resulting in more-or-less zero shrinkage). If you could get enough of a shelf, or even insert a beam on the edge it seems like you could transfer force/load into the new columns and relieve some of the load at the adjacent columns in a rational way.

Alternately, what about some steel angles that are epoxy bolted a bit down the existing column, then that portion above it is filled with concrete and some reinforcing bars to transfer the load in bearing and expand the b0 perimeter in the punching shear calculations at the existing columns? Ugly, perhaps, but safe(r)? How much added perimeter would you need to get the existing column-slab joint to work?

Live load reduction will probably not do any good, plus the square footage and the fact that I don't think slabs are even listed for live load reduction....

What about actual strength of the as-built concrete? Sometimes the concrete is stronger (or weaker), maybe some testing? Would another 50 psi concrete strength get you a "pass" under the original code? (Not that that's what you are actually after, I guess).

 

[tomzeb]

Yes, the column will have to be "cut in" to the existing CMU wall. Sounds like you think it is feasible, just tricky at high elevations. Feasibility is what I'm after. The existing b0 is 66 inches for an 8x36 inch column and the demand/capacity ratio is above 2. It would take adding 5.5 ft to long axis of column to pass the SP calculation or adding 8 inch of thickness of concrete extending out in the critical perimeter region. Having 500 psi extra concrete strength doesn't improve things that much. Definitely a Master's project.

 

[lexpatrie]

Side note: Are you looking at the existing via Chapter 27? They allow larger phi factors than during design, if it's based on measurements of as-built conditions?

 

[tomzeb]

I'll checkout out, but it seems that its missing limit by too much for a small factor adjustment.

 

[lexpatrie]

It's worth noting, mostly for the "next one" as going from 200% overstress to 190% overstress won't really do much. If you can get more on the existing strength it might help somewhat, as it all contributes.

I thought you were working with a PT slab. As I recall a lot of reductions are taken for relaxing of the strands, is there any chance you can measure the PT tendon stress and get more N for the shear force? Now that I look at it, it isn't glaringly obvious there's any increase allowed for PT anchorage, although if this is at a column it's probably not an anchorage, except you said it was at the corner of the slab. It gives an increase for shear (0.75 to 0.80)

.....................................

.....................................

 

[tomzeb]

The increase in strength reduction factor to 0.8 is helpful. The column is on an end wall so ACI refers to it as an edge column. The code does not seem to give any procedures for taking advantage of post tensioning for the shear punching calculations. I guess they assume the positioning of the tendons coming through a column are not effective or reliable in producing shear stresses on the full cross section. The post tensioning would act on the concrete x-section in a direction opposite the gravity load, which would be very helpful. Is there anything in the code that would allow such a treatment?

 

[lexpatrie]

I wonder if there aren't "two" approaches here and I'm not sure how you'd reliably combine them, the compression force from the post-tensioning would seem to increase resistance in a shear-friction sense, but punching shear doesn't really consider "shear friction",

But the punching shear check doesn't contemplate it that I know of, (possibly as it's a reinforced concrete check with minimal PT research). So I hesitate if there's not some measure of double counting involved here.

I thought there were some "additional strength" in some of the beam shear equations that involve normal force when it's compressive.

I'm not convinced any of the punching shear tests used normal force to try to increase the shear capacity.

I did find another article you might want to look at, has some reinforcing photographs of how they fixed it. In essence a drop panel and a column capitol together. I skimmed so I'm not clear they designed the new stuff for 100% of the required load or tried to pro-rate it somehow. My impression is it's generally a pro-rated approach, and I suppose at ultimate it would work, though I can't say that for sure.

https://ascelibrary.org/doi/10.1061/9780784482018.045