Shear Crack in Concrete Beams

[aguynamedbryan]

Hi All,

I was hoping to gain some insight on visible shear cracks in concrete beams. At what point does a visible shear crack become a concern? I've attached a picture of a crack in a parking structure beam; is this something to be concerned about?

Thanks for your input.

Bryan

 

[KootK]

I'm not sure where the good/bad line for crack widths lies. In my opinion, if this is a shear crack, it's not wide enough to be of any concern. As a simply supported member, I would have expected a true shear crack to telegraph through to the underside of the beam. The crack in the photo doesn't appear to do that.

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.

 

[hokie66]

That is a typical diagonal tension (beam shear) crack. It is enough to make me concerned, and I would want to further investigate the design and loading before I decided to leave it alone.

 

[JoshPlumSE]

A number of years ago (post northridge earthquake) there were a number of attempts to quantify the widths of acceptable cracks. A paper (perhaps internal to the Office of Emergency Services) by Hasssan Sassi. Essentially trying to quantify a loss of capacity with the crack width.

Maybe for immediate post-disaster assessment and the yellow or red tagging of a building, this could be useful. But, the problem with the methodology is the crack by itself doesn't really say anything. Yes, your beam has a crack. And, yes it appears to be a shear crack. But, concrete is expected to crack.

The crack width is probably not large enough to cause significant increased corrosion of the reinforcement. It might be wise to look at the original design and make sure that the beam can handle the loading properly. It might be good to look at some construction photos or as-builts (if any of that is available) to see if there was something missed in construction. But, as long as the beam was designed properly I don't see that particular crack as alarming.

 

[Guastavino]

I would check the capacity of the concrete beam for shear and compare it with the loads. I also think there are some more in-depth shear cracking equations than ACI 318 gives, since ACI 318 is more for design than analysis. A quick google search found this:

http://www.ce.memphis.edu/4135/PDF/Notes/Chap_9_1_shear.pdf

Whatever you do and/or recommend, I would include a note about sealing the crack and monitoring it as a minimum. Whenever I write a report about cracking, I ALWAYS say to continue monitoring the crack. It's a disclaimer but it's also only fair. I mean when I see a crack it's usually a 1 hour snapshot into the 30 year life of a structure. We can make very good educated guesses, but in reality I think that's about the best you can do. I also use words like "Appears" and "limited visual survey" etc. because I want people to know that this isn't an exact science. Some may think it's a cop out, but I think it's just the nature of our profession. We aren't dealing with swiss watches and no matter how detailed and diligent we are, there is still judgment involved.

 

[KootK]

I would include a note about sealing the crack

This has me contemplating something for the first time. We expect a reinforced beam to crack flexurally. And we calculate expected crack widths so that corrosion isn't a problem. So why is there no accounting for shear crack widths? Surely we must acknowledge the potential for shear cracks if we're going to bother with stirrups?

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.

 

[JAE]

KootK - for flexural cracking the crack widths can get quite a bit wider due to the more elastic nature of flexural failure (longer warning, higher phi factor, etc.)
With shear there is a much more abrupt failure, more scatter in the test data (lower phi factor) and thus less give in the beam before failure. So I would susptect that a shear crack is a bit more
worisome thing than a flexural crack.

A number of responses above are what I first thought of after seeing the crack - go back to the plans and do a design check on the shear capacity.

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

 

[hokie66]

I agree with JAE. Shear cracks are of much more concern than flexural or direct tension cracks.

Checking the shear capacity by the plans will only tell you so much, as the concrete portion of the shear capacity is now lost, at least by the code equations. After doing the code check, I would approach it as a truss analogy. The top steel is your tension chord, the bottom concrete and steel form the compression chord, the stirrups the tension webs, and concrete diagonals parallel to the crack are compression webs. A critical part of this is the development of the stirrups. Hopefully, they are closed ties.

This would appear to be some sort of precast structure. If so, access to the fabrication drawings will be crucial.

 

[KootK]

With shear there is a much more abrupt failure, more scatter in the test data (lower phi factor) and thus less give in the beam before failure. So I would susptect that a shear crack is a bit more worisome thing than a flexural crack.

I agree with JAE. Shear cracks are of much more concern than flexural or direct tension cracks. Checking the shear capacity by the plans will only tell you so much, as the concrete portion of the shear capacity is now lost, at least by the code equations.

I suppose that I should clarify that, I my comments above, I'd been thinking of beams containing meaningful shear reinforcing. For an unreinforced beam, I agree with the statements above. For a reinforced beam, I'm not so sure. Some thoughts regarding shear reinforced beams.

1) My understanding is that shear reinforcement is not fully utilized until after diagonal tension cracking takes place.

2) I contend that a shear crack does not mean the loss of the shear strength attributed to concrete. This philosophy is certainly the case in ACI where Vc + Vs implies that Vc is available as a post-cracking capacity.

3) I agree that there are important structural reliability differences between bending and shear. However, those are addressed via our phi factors and should not affect our expectation that significant shear cracks will occur as full shear capacity is approached.

4) To restate my previous post, I find it surprising that we would expect shear cracking at the development of capacity but not have any related serviceability criteria to limit crack width. Clearly at some size of shear crack, capacity would diminish and corrosion would be an issue.

In a forensic application, particularly one possibly involving prestressing and expensive repairs, I'd turn to the modified compression field theory (Link). As far as I know, it's still the gold standard for the accurate assessment of concrete shear. A coefficient of variation less than 5% is about as good as it gets in natural systems.






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.

 

[rapt]

Definitely a shear crack and definitely needs to be checked to make sure both all of the design, detailing and construction were done properly. If there is properly detailed shear reinforcement present across the crack and the tension reinforcement at the bottom is correctly developed and the design calculations show it is ok, then you are probably ok.

Also, you have not mentioned if the beam is RC or PT and if PT, bonded or unbonded.

I do not know why it was assumed by some that ACI318 was used. I commented on this on another post. There are other design codes in the world and questions come from buildings designed to those codes. In any case, I would check it in accordance with the Canadian rules and possibly Eurocode as they are more modern than some of the older shear design methods in ACI, BS and AS codes, for both shear design and end development of the tension reinforcement.

 

[KootK]

I do not know why it was assumed by some that ACI318 was used. I commented on this on another post. There are other design codes in the world and questions come from buildings designed to those codes.

We sometimes assume ACI for the same reason that we're corresponding in English here and the same reason that law enforcement agencies are tempted to employ racial profiling: we're rational creatures with an innate understanding of probability. When the OP doesn't specify a code, what are we supposed to assume? Estonian?

Besides, nobody has yet to assume anything in this thread. Those of us that have mentioned ACI have simply done so as a means of attempting to relate what we know of concrete shear theory and practice to the OP's issue. If the OP is only interested in hearing the Latvian/Australian/Martian perspective, then that needs to be established explicitly.

My native tongue is Canadian/CSA. I often switch to Yank/ACI here in deference to the fact that 90% of the concrete traffic is 318 based. When in Rome...

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.

 

[IDS]

We sometimes assume ACI for the same reason that we're corresponding in English here and the same reason that law enforcement agencies are tempted to employ racial profiling: we're rational creatures with an innate understanding of probability. When the OP doesn't specify a code, what are we supposed to assume? Estonian?

Eurocode 2 would be a good choice in general, then you'd have Estonians covered as well as everyone who speaks English, and also a document that is available to everyone at no cost:

https://law.resource.org/pub/eur/ibr/en.1992.1.1.2004.pdf

But for discussions of shear I agree that the Canadian code would be the best choice.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[KootK]

Why don't we just switch to Esperanto around here while we're at it? Eurocode would be a lousy default for the same reason that Martian would be: the majority of the traffic here is not Eurocode based, it's 318 based.

Frankly, as far as I can tell, things work just fine around here when we just let people speak whatever code they're comfortable speaking.

Canada's shear provisionS are only good because we helped devise the modified compression field theory that I mentioned above. And because we consider it our own, we were quick to roll aspects of it into our shear design provisions.

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.

 

[hokie66]

KootK,
You wrongfooted me on the "modified compression field theory". I thought you might pick shear friction.

 

[KootK]

Ah Hokie... I've been making a concerted effort to be less of a shear friction maven. But, like a recovering meth-head at a good party, I always get sucked back in by bad influences.

Check out this awesome paper by another Canadian: Link. It presents a very clever model of conventional one way beam shear predicated upon shear friction. It gained very little traction unfortunately. I see elements of it in MCFT actually. Or vice versa I suppose.

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.

 

[KootK]

@Bryan: as I mentioned at the top, it troubles me that this shear crack doesn't seem to extend to the bottom of the beam. I feel like any self respecting shear crack ought to run out to the tension face. I guess I'm just always wanting Mohr.

Looking closer at the photo, something of the crack does seem to run out to the tension face of the beam. It looks wider than the crack and whitish? Any chance that's a previous crack repair? Or is it just my imagination / debris runoff?

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.

 

[hokie66]

KootK,
I think the crack, if it exists at the bottom, is very narrow because that is the compression face. See my post above where I suggest the bottom concrete and steel form the compression chord in a truss analogy.

 

[Tomfh]

it looks like it goes all the way down, but has salts or autogenous healing or something at the bottom.

Looks like an ordinary shear crack to me. the concrete is failing in diagonal tension, and the steel is being mobilised as the crack opens. this is how reinforced concrte beams work in shear, no?

 

[KootK]

I think the crack, if it exists at the bottom, is very narrow because that is the compression face

What makes you say that Hokie? The joint deatailing strongly suggests, to me at least, that the beam is precast and intended to be simply supported. That's part of what's been bothering me about the crack not extending to the bottom, however, as the lack of extension would suggest hogging moment. Maybe the structure above has created accidental restraint?

the concrete is failing in diagonal tension, and the steel is being mobilised as the crack opens. this is how reinforced concrte beams work in shear, no?

That's my presumption, yes.

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.

 

[cliff234]

I'll echo what rapt said. Check the design, check that the steel was properly detailed and verify that the placement of the reinforcing steel was inspected. Are there any photos of the reinforcing steel before the concrete was poured. Is this a precast beam? (The beam appears to be bearing on a pad indicating that it is precast.) Are there other identical beams? Are any of those cracking? Is there anything about this beam that is different from other identical beams?