Help identifying this crack pattern in concrete walls 3

[Robbiee]

Hello,
I was asked to inspect and provide opinion on the cause of cracks developed in load bearing end walls of a building. Please see attached. The building superstructure is precast concrete, 23 storey high and rectangular in footprint. These cracks, as seen, start from the wall vents located near the ends of wall panels and propagate diagonally away from the support. The cracks extend through the wall. To me, they appear to be shear cracks caused by the reduced section of the beam part of the wall/frame, but can't confirm if the reduced section is adequate or not given that no design drawings of the individual wall panels are available for review. What do you think, and what would be the action plan to provide an opinion on the cause?

 

[TehMightyEngineer]

So my question is this: when one designs and or supplies something like this, do you design it not to crack under service lateral loads so as to prevent aesthetic problems? I'm guessing not but then there's the potential to end up where we may find ourselves here. And, hopes and wishes aside, it's hard to imagine that a stress concentration like these block outs could be dealt with in a serious, load carrying member without initiating cracking.

@KootK: I am assuming most precasters will be similar to our plants in this reply. From what I've seen this is fairly reasonable to assume.

It depends on the specification and project requirements how much I typically dig into a design to avoid cracking (engineering judgement, strut-and-tie, FEA, etc.). At a minimum extra reinforcement would typically be provided around openings, either through analysis or rule-of-thumb. I imagine this is similar to what most precasters would do, unless the EOR designed the panels and already called out the rebar layout. Many times I will actually defer to the guys in the shop regarding more typical pieces; they'll typically have just as good of a feel as I (if not better) for what is at risk for cracking during casting and handling.

Almost always any cracking during casting, erection, or construction is an issue (for pretty much exactly the reason you stated). Typically precast components are required to be uncracked up to construction being completed. To accomplish this I often have to consider just the plain concrete strength and even that may be at a slightly reduced compressive and tensile strength during the early life of the piece. Thus, you're correct that much of the added reinforcement likely isn't doing much other than restraining any cracking after it might occur.

After construction, I've very rarely been required to have uncracked pieces for service level loads. Of course, prestressed members often are uncracked anyway and a lot of what I design (columns, buried structures, etc) is under high compression loads which help keep cracks closed. On my own designs I've often designed for no service level cracking when it was inexpensive to do so and provided significant benefit to the owner (for me this is typically precast stairs, precast buildings/facades, and similar).

Often it's not hard to work around blocked out areas or other stress risers. Typically we adjust our lifting points to keep those areas in compression or low tension. Other things might be done as well; bolting steel reinforcements to the exterior, pretension to provide a compressive force, cast with non-metallic fibers in the mix, or even just try and cast a trial piece and see if it cracks.

Very rarely does the project actually require anything beyond just "no cracks please". Recently had a DOT project for bridge deck panels with a ton of blocked out areas for the girder shear studs. Lots of weird panel shapes and block outs within a few inches of the panel edge. Project required all handling tensile stresses be less than 5*SQRT(f'_{c_release}). Ended up having to do a full 3D FEA of each panel to identify the stresses at the bond outs. The shop guys and contractor probably hated me as I had some panels with extra lifting points to be held up with straps and come-alongs to keep the tensile stresses low.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

http://www.americanconcrete.com

 

[Robbiee]

Thanks everyone for very valuable opinions. I got today the individual wall panels drawings. Please see attached.
KootK,
I agree with you regarding the coupled shear walls you noted above, but do you see these panels put together acting as coupled shear walls? I see more of a frame behavior.
dhenger,
No, the walls don't have a ledge to support the floor. The floor sits on the wall.
Oldestguy,
I agree that differential settlement would cause this type of cracks, but why not in the lower levels like Ron noted?
Brad805,
The holes were blocked out in the plant.
My opinion:
The reduced section is half of the "beam" section and located near the support. Under gravity loads, the beam on the sides of the opening can act as cantilevers given that top rebars are present and anchored into the columns. Under lateral loads, the reduced section is only two 4"deep x 8" wide pieces with no shear reinforcing. This section is subjected to the max. shear in the panel. The reduced section could have cracked because its shear resistance is less than than the required shear. ( we would need to analyze the building to get the shear stress in these panels and check if the above is correct). No analysis has been done by us.

 

[KootK]

I agree with you regarding the coupled shear walls you noted above, but do you see these panels put together acting as coupled shear walls? I see more of a frame behavior.

Yeah, I do. I puzzled over it a while myself though. Here's what I came to:

1) I expect you'll have the superposition of two types of behavior: cantilevered wall/frame for which coupling demand would be a maximum up high, and shear mode building sway (moment frame) for which coupling demand would be a maximum down low. I have trouble wrapping my head around the full complexity of it but this is why my gut tells me that max demand may well be in the middle, where you've got some of both going on. Were I in your shoes, I might try to quickly model something of similar proportions just to see what kind of behavior really does shake out in this regard. That could probably be just an afternoon's fun so long as you're not getting to hung up on trying to model the real thing accurately.

2) Without doubt, each bay of stacked panels will exhibit some coupled shear wall behavior. In my mind, the real question is whether or not the whole wall will act as one frame or the individual bays will act like several, separate shear wall systems. To explain where you do and don't see cracking horizontally, the entire wall would need to be acting as one. That makes the middle the location of peak vertical shear. I'm guessing that the connections are such that the entire wall would act as a single system but I'd need a look at the connections to be sure-ish.



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[NorthCivil]

I think kootk is on the money with his diagram screenshot he posted. this is why there is no cracking on the bottom levels. this is a coupled wall situation

whether or not its a safety issue at this point is not our judgement call to make, but as an armchair engineer back home with no skin in the game, I would say its not a big concern. I might request a concrete scanner to swing by for $500 just to make sure there is a bit of steel around that opening.

 

[Robbiee]

NorthCvil,
Thanks. The shop drawings of the wall show there are steel around the openings as per the attachment I posted previously. But, Don't you think there might be a design deficiency for lateral loads should analysis show the shear force at the location of the opening can't be resisted by the reduced section-between the two ties?

 

[dik]

Robbiee... what are the bar sizes? #3, #4, #5, 10M, 15M? There does not appear to be a lot of reinforcing across the headers. Why the hooks on the vertical bars and not the horiz ones? They both have less than 16" for development.

Dik

 

[Robbiee]

Dik,
The drawings show 4-15M bars top and same bottom in the top beam. There is one closed tie of 10M at each side of the opening. No other ties in the beams.

 

[KootK]

There is one closed tie of 10M at each side of the opening. No other ties in the beams.

Really? If you're right about that, it may indeed be a problem. I'd assumed that there were regularly spaced stirrups that you hadn't bothered to show. Strange choice for a major load bearing beam element in my opinion.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[NorthCivil]

Robbie,

Are these panels the main intended lateral force resisting system for the building?

if so - then yes, we do have a major problem here. I was under the impression these are cladding elements that have been taking some lateral because they are stiffer than the main LFRS.

 

[Robbiee]

NorthCivil,
These walls are called shear walls on the drawings.

 

[NorthCivil]

Robbiee,

It looks like the designer of the steel layout within these walls didn't consider coupling wall behavior. that's why there are so many verticals, but few horizontals.

 

[NorthCivil]

they looked at it like this.

Or they looked at the behavior as identified as "horizontal shear failure" within this image below. they didn't consider vertical shear failure.

[

http://theconstructor.org/wp-content/uploads/2012/03/clip_image0123.jpg

]

 

[Tomfh]

Under gravity loads, the beam on the sides of the opening can act as cantilevers given that top rebars are present and anchored into the columns.

Could it be gravity deflection of the long cantilever under the weight of the floor planks?