Control Joints & Curling in Gymnasium Concrete Slab on Grade 2

[KHoff]

A gymnasium addition I designed is currently under construction. The slab on grade in the gym consists of 4" concrete with fiber mesh over a 15 mil vapor barrier and granular fill. In addition, I show sawed control joints at 12 feet on center in both directions. The contractor recommended to omit the control joints under the gym floor to prevent curling. I understand curling can occur at a construction joints, so I specify dowels at construction joints. Should I be that concerned about curling at control joints as well? Should I be more concerned about the random cracking that will occur without joints?

 

[jayrod12]

Curling can also occur at control joints but with proper reinforcement you can evenly distribute the cracks that will happen and keep their size to a minimum.

 

[Ron]

Curling is caused by drying shrinkage at the top of the slab that exceeds the drying of the bottom of the slab. You have done a good job of specifying the control joint spacing....make sure they do it and saw the joints the same day of concrete placement.

If they will cure the slab properly, preferably with moist curing for at least 7 days, there will likely be minimal curling. It is more often a problem in exposed slabs. Also, pay attention to the mix design that the contractor will be using. If he is pumping the concrete, he will likely use a smaller aggregate. This requires more cement and more water...thus more shrinkage. If you have not already done so, require that they use the largest aggregate practicable for the placement. No. 57 stone would be the minimum size I would use for this application.

Make sure they positively cure the slab!!

 

[KHoff]

The slab was specified with fiber mesh rather than W.W.F. or standard reinforcing bars. At this point adding reinforcement would result in a change order, which is less than desirable.

Without adding reinforcement, if control joints are omitted it seems that random cracking in the slab is a certainty. However, with proper curing the slab curling at the control joints seems like it could be controlled (or at least minimized).

 

[Ron]

WWF does little or nothing to control cracking or curling. Don't worry that it was not specified. Using Fibermesh or similar polypropylene fibers is purely a mix enhancement, not a reinforcement. It will help to move the shrinkage cracking to further spacing but outside that, it does little.

I'm not a fan of WWF. It is usually not placed properly and does little to help the concrete. When I specify WWF, I specify it in sheets, not rolls and I specify that it be placed in the second of two lifts of concrete. Almost never gets done, but that's the only way it actually works.

 

[oldestguy]

The main cause for curling is the difference between top and bottom of slab for water-cement-ratio. High W/C/R section shrinks more than the lower W/C/R. That means use as low a slump as possible and not working of the wet concrete any more than absolutely needed. Also wait until enough setting has taken place before finishing to prevent that migration of water up. That barrier underneath also helps avoiding loss into the soil below.

 

[KHoff]

I understand the issues with WWF and that it is often ineffective due to improper placement. If the initial intent with this slab had been to use reinforcing rather than control joints to control cracking I would have specified standard rebar. However, in this case the intent was to use fiber mesh in combination with properly placed control joints and dowels to control cracking in the slab which is essentially unreinforced. I typically specify lower slump for slabs on grade as well.

 

[KHoff]

I was looking for information online about fiber mesh and curling and found the following paragraph:

Curling is caused by differences in moisture content and/or temperature between the top and bottom of the slab. When the top surface dries and ultimately shrinks, while the bottom dries at a slower rate, the slab can exhibit deformations at the edges and corners. Nighttime temperatures can cool the top surface of the slab while the mass or bottom is not cooled, creating or adding to the curling or deformation. The benefit of Fibermesh fiber reinforcement is the internal support provided in the critical green state. Uniform bleeding limiting the formation, scope, and migration of the bleed water hinders the conditions conducive to curling.

This was a publication from a company selling fiber mesh, so they aren't exactly unbiased. While I don't agree with their reference to the product as fiber "reinforcement", I wonder if the fiber mesh does help prevent curling. Does this seem reasonable?

 

[JAE]

So will the slab on grade slab be covered with a wood floor or something? In some sense, the control joints simply organize the cracking into nice straight lines.
If you omit the joints, the slab will crack more randomly. At these cracks, and more so where the cracks intersect (forming corners) you will still get curling unless you
keep the shrinkage down and cure very well as mentioned above.

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

 

[darthsoilsguy2]

i've always heard "NOT having slab control joints close enough CAUSES curling"..... NRMCA agrees.

http://www.nrmca.org/aboutconcrete/cips/19p.pdf

proper curing methods at the proper time is the best thing the Contractor can do to prevent curling. i say proper time meaning they have to finish that PORTION of the slab immediately after bleed water goes and start curing the finished parts. many finishers don't want to start curing until they have all their finishing equipment loaded up in the truck... meanwhile at least half of the slab could have already started curing while they were finishing the other side. There is a diminishing return on curing... more evaporation occurs earlier than later. Contractors will scratch their head about curling because they know they eventually followed the curing procedures when they were done, but don't realize that the pump breaking down 3 or 4 times during the pour made one area sit for too long before they finished it and then again so from curing.....

 

[KHoff]

Yes, the slab will be covered with a wood floor. I suspect this could be the reason the contractor asked about the joints in the gymnasium area specifically and not the other addition to the building that used the same slab design. Also, there was not any curling observed in the slab at the other addition.

 

[hawkaz]

Have them water cure the slab.

 

[BUGGAR]

I used to have big problems with visqueen vapor barriers below the slab - lots of cracking, etc. See other comments for what happens when the water in the mix has only one way to escape.

 

[JAE]

But with a wood floor you do want a vapor barrier.

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

 

[msquared48]

Since the slab is being covered with a wood floor, it is obviously not a curling rink.

Mike McCann, PE, SE (WA)

 

[KHoff]

Well played sir...

Thanks for your comments everyone. It seems that the control joints bring more positive effects to the slab than negative.

 

[doka1]

Have you considered using dowel baskets? Browse PNA-inc.com...curling is a problem for any slab, so by that logic we should never cut slabs???? A good wet cure and cutting right behind the finish operations helps. I really like the mention of larger aggregate.

As a concrete contractor, on a gym floor I am most corn earned with flatness. I would use a laser screed and pans on the largest ride on machine in the fleet.

 

[Ron]

doka1...floor is too thin for dowel baskets, even plate dowels. Could possibly use plate dowels alone, but 4 inches is below major manufacturer's recommendations, even for isolated plate dowels.

Agree with you on laser screed.

 

[doka1]

Good catch Ron, I wasn't thinking about the slab thickness.