Acceptable level of flexural cracking in grade beam

[jeffhed]

I have a concrete wall of a subgrade parking structure that is on expansive soils. The foundation wall also acts as the grade beam spanning between piers. The voids have been compromised and we are in the process of restoring the proper void depth. One side of the garage is like a walkout basement with large ventilation openings in the wall. At this point the grade beam goes from a 10' tall grade beam to a 42" tall grade beam. Instead of placing piers at each side of the opening, one pier was placed in the center of the opening. I had previously observed vertical cracking at the bottom corners of the openings. Now that the concrete slab has been removed to restore the void spaces, the cracks are clearly flexural cracks that start at hairline widths at the top and taper to up to 1/8" at the bottom. Looking at the bottom of the grade beam and the exterior face of the grade beam, the crack extends completely through the grade beam and it looks to be the same on each side of the wall. So I checked the grade beam right at the point where it tapers down to 42" tall and it doesn't work according to the details on the existing plans. So we are going to install additional piers on each side of the openings. Now I finally get to my question, say that the grade beam did work calculation wise. Obviously the steel inside has yielded, there is no concrete crushing damage at the top. How much crack width is acceptable? In the event the calculations work, would it be reasonable to calculate the strain in the steel and determine the elongation from that. Does anyone know of any documents, text books, etc. that provide guidance on this? I have done a lot of searching on the internet and can't seem to find anything useful.

 

[Teguci]

Use a strut and tie model per the Appendix in ACI 318. If the crack is the full height of the beam then it sounds like a shrinkage crack to me. Make sure to provide properly spaced joints. As for acceptable crack width, structurally I be concerned with leaqkage and overall deflection. Otherwise, it becomes a visual issue.

 

[jeffhed]

Teguci,
A strut and tie model would be appropriate for the deep beam segment that was the same height the whole way. However, the strut would run into the air at the ventilation opening. I was just trying to provide some background as to why I wanted to find some guidance on acceptable crack widths for future reference. In this case it doesn't matter, the beam isn't properly reinforced. But if it were, what would be an acceptable crack width?

 

[jeffhed]

Here is a pdf of what I have. Tried multiple times to get the pdf rotated properly, but was unsuccessful.

 

[ishvaaag]

From old EHE 98 (would have to look the latest to see if something has changed), for growing exposure classes

Exposure wmax (a crack width only exceeded 5% of the times)
I 0.4 mm
IIa,IIb, H 0.3 mm
IIIa, IIIb, IV, F 0.2 mm
IIIc, Qa, Qb, Qc 0.1 mm

Being a garage, deicing salts can be present so quite likely wmax<=0.1 mm where exposed to deicing salts, elsewhere follow the exposure class

Respect the nature of the crack, being wider at the bottom, it seems dishing action and the expansive character of the soil may also have a say.

 

[ishvaaag]

wmax is the average crack width got at service level multiplied by 1.7 for wind and weight solicitations, or
by 1.3 if of rheologic, earthquake, settlement, temperature cause.

 

[rowingengineer]

jeffhed,
You may wish to review the strut-tie method in more detail. you can detail the strut tie around opening.

here is a starting point;

http://icsge2009.com/2007/icsge/RC/RCS039p F.pdf

How could you do anything so vicious? It was easy my dear, don't forget I spent two years as a building contractor. - Priscilla Presley & Ricardo Montalban

 

[jeffhed]

rowingengineer,
As I stated before and show in my pdf i posted, the opening is not in the span of the beam, it is at the end of the beam where it is supported by the pier. If it was in the middle of the beam and was in between the bottom steel and the compression area, it would be similar to when TJI manufacturers let you drill holes in the web of the TJI as long as it is doesn't take away parts of the top and bottom chords and it is in the middle third of the span. The opening I am talking about is not in the midspan of the beam, it is directly above the pier that is supporting the grade beam. There is no way for the compression strut to go around the opening and get to the supporting pier in the middle of the opening. Would anyone let a contractor remove the web of a TJI over the bearing point? I think no matter how you slice it, the smaller beam portion below the opening has to be strong enough to transfer the load from the deeper portion of the grade beam to the pier.

 

[hokie66]

Based on your sketch, the wall is probably cantilevering to the opening, and the shallow beam below the opening is spanning between the end of the wall and the pier. The cracks are restraint shrinkage cracks, caused by the stress riser at the reentrant corner.

 

[jeffhed]

hokie66,
The wall probably would be cantilevering on one side, but on the other side there is a corner. I have added to my sketch to show more accurately what is out there.

 

[hokie66]

Lack of coordination with the mechanical engineers is a trap all of us have fallen into at one time or another.

 

[jeffhed]

hokie66,
I agree. But in this case the opening were shown on the structural drawings in the same place. We have been hired to fix multiple problems to the existing buildings. Which is why I posted in the first place the question about maximum allowable crack widths. Most cracks are from differential movement from expansive soils, but a few cracks are from stuff like this. Wanted to find out minimum allowable crack widths that are considered acceptable even when calculations show that everything is fine. I have a couple of those situations on this project. If the member is adequate, how large of cracks would you expect to see?

 

[hokie66]

Unfortunately (or perhaps fortunately), limits on crack width is one of those things which is still subject to engineering judgment. Suggested reading...the Commentary in Chapter 10 of ACI318-08. That said, traditionally cracks in the range of 0.4 mm or 0.015" have been accepted, except in water retaining structures or where exposed to corrosive environments. This agrees with the table ishvaaag posted above.

 

[jeffhed]

Thanks, hokie66. This would correspond to what I have previously seen in the field. Basically the absence of cracks other than hairline cracks in members that are sufficiently designed and constructed. Thanks for the information and also thanks to ishvaaag for providing the information as well. This will also be a good guideline when we get down to patching the cracks in the wall that have occured from building movement.

 

[ishvaaag]

I have looked in the EHE 98 code for the class of the structure and where not protected by impermeabilization, a concrete structure exposed to chlorid salts of different origin than marine fares as class IV, hence 0.2 mm limit width.

 

[jeffhed]

Thanks, ishvaaag. All of these numbers of crack widths that are being posted appear to be to help prevent water infiltration and rebar corrosion. However, are these crack widths the EXPECTED crack widths in a structure should everything be designed and constructed correctly. I think that is really what I am after. Say we design a grade beam, it is detailed properly, constructed properly and loaded up the same as the design loads. Would the beam crack on the bottom? It probably would have to to engage the reinforcing steel. But, how wide would the crack be before we would say that it should not be cracking that much. Can an expected crack width be calculated based on the member dimensions, loading, etc?

 

[ishvaaag]

Certainly the limit crack widths given are against the calculated wmax coming from a set of specific hypotheses, i.e., the limits are for cracks appearing from mechanical action developing from the computation under a set of hypotheses.

Normally the computation wouldn't include shrinkage cracks, for example, mainly because to some extent it would be a more arbitrary input than the devised procedure to estimate cracks upon flexure itself. This means that shrinkage cracks must be prevented by proper jointing and sealing, not that the contribution to cracks from shrinkage should be dismissed, since, obviously, as the widths are set to prevent rebar attack, any contribution is meaningful.

So, for whatever the condition of your model representing the built state, all the factors contributing to wmax must be included, rheological as well, something that is more easily said than done.

As a comment, the calculation procedures for flexural crack widths have sometimes shown their nose in the codes; however, the crack widths in a shear-flexural computational context I mainly have only seen in the canadian set of programs that come with RESPONSE, a program freely available from the web. Even so it was more demonstrative technology than fit for practice, since only for specific cases.

I made a pair of Mathcad Professional 2000 worksheets that deal with crack width estimation for flexural cases, I think one in american setup and other in european; no word therein for shrinkage contribution. However if you are interested I can post them, working or as pdf printout.

 

[ron9876]

I would say that first be sure that what you have--or what you plan to do--is structurally sound then pressure inject epoxy cracks.

hokie66 looks to me like combo flexural/shrinkage cracks.

 

[ron9876]

I would say that first be sure that what you have--or what you plan to do--is structurally sound then epoxy inject cracks.

hokie66 looks to me like combo flexural/shrinkage cracks.