Sawcutting Control Joints 2

[Canuck67]

I am looking at the design of a concrete slab-on-grade that is approximately 40'x40' in plan dimension. The final finish is specified by the architect to be polished concrete, therefore final appearance, cracking, etc... must be controlled as much as possible. In order for the control joints to be successful, the mesh must either be cut at time of sawcutting, or placed in such a manner that it is not continuous across the control joint locations.

The difficulty comes from the embedded plastic hydronic piping for the infloor heating. In my opinion, the hydronic piping must also be laid in a pattern to avoid crossing the control joint locations. Can anyone confirm whether this makes sense? I am of the opinion that even the tensile capacity of hydronic piping crossing a sawcut location may negate any 'weakness' induced at the sawcuts allowing cracks to possibly form elsewhere. (Not to mention the risk of cutting into the hydronic piping if depth is not controlled precisely!).

 

[JAE]

I guess for 40 feet - I'd be inclined to look at variations in the design mix to reduce or eliminate shrinkage cracks and avoid joints altogether.

We did a 40 ft. x 80 ft. floor area a while back that required very low curling and approached super-flat floor status (for large format tiles). The mix was a combination of larger aggregate, shrinkage reducing admixtures, water reducers, etc. No joints, no curling, etc. Worked out great.

Try calling your local concrete supplier to see what they can do for you.

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

 

[bmrossetti]

Canuck67,

In my experience, cutting or discontinuing the mesh at the contraction joint is not necessary. It is more important to use a low-shrinkage mix design with low total paste content and well-graded aggregates (see the high-performance "athletic" concrete articles by Bill Phalen of Euclid Chemical), and to cut the contraction joints as early as possible (using the Soff-Cut early entry cutting system, or similar).

By making the cut early you are introducing a weak point into the slab while the thermal and moisture gradients are pretty large; the internal stresses will be concentrated at the cut. The mesh will likely yield, but will maintain the aggregate interlock at the cut and help keep the joint from opening. For the same reasons, I would not worry overly about the hydronic piping (if it is PEX or similar will the concrete even bond enough to develop the tensile stresses?).

Another option is to place a topping slab over a base slab (with the hydronic piping in the base slab), and apply the diamond polish finish to the topping slab. This helps alleviate concerns with the hydronic piping, and has the added bonus of making the polished slab easier to protect from surface damage that would detract from the final results while construction is ongoing, if the topping slab can be placed towards the end of the overall construction sequence.

Regards,

Brandon

 

[wannabeSE]

JAE,
Was the project with the 40'x80' super flat slab a design-build project? Or if it was a design-bid-build project, how were the slab requirements specified in the construction documents? Was a set of performance requirements specified, or did the specifications get into the specifics of the mix design (specifying admixtures types and quantities, etcetera)?

 

[JAE]

The initial specifications were pretty standard until we started getting submittals on the tile which came with mandates that no (or minimal) curling be allowed and any slab joints would have to be aligned with tile joints, etc....a difficult proposition. We did mandate stricter F numbers for the slab in the original specifications. After getting the crazy tile demands, we started communicating with the contractor and came up with a mix design to avoid the joints (see attached).

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

 

[JAE]

Mix design:

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

 

[wannabeSE]

JAE,
Thanks for the follow-up.

 

[Ron]

Ok....let's think about this one.

First, wire mesh will not and does not prevent shrinkage cracking. It only serves to hold the cracks together when they occur and only then if placed in the middle third of the slab (I have eliminated the top third as a practical possibility since in 38 years of taking hundreds of cores of concrete and observing the placement of thousands of cubic yards of concrete, not once do I remember seeing wire mesh in the upper third of the cross section).

Then we have the mix design. Low shrinkage mix designs must have a high percentage of larger coarse aggregate, typically at least as large as #57 stone, preferably even larger. The fine aggregate percentage should be in the range of 38 to 42 percent of the total aggregate content. Having a low water-cement ratio (0.50 or lower) certainly helps; however, the total amount of cementitious material makes a difference (a low water cement ratio can be obtained with a high total cement content...which increases shrinkage), so if the fine aggregate ratio is high or if the coarse aggregate is smaller in nominal size, the amount of cement paste must increase to cover the increased specific area of the aggregate.....thus a larger aggregate mitigates shrinkage.

Third, the mix design that JAE provided might be fine to mitigate large cracks under tile; however, for exposed, polished concrete cracks must be prevented completely. In order to do so, there are several things that must be done with the concrete.....

First, the thickness of the slab MUST be controlled carefully to prevent inordinate subgrade restraint which leads to random cracking. The slab thickness must be maintained such that a tolerance of -1/4 inch, +3/8 inch is maintained; however, even such a transition should not take place abruptly....at least a 5 foot separation for such a transition is necessary.

Second, there must be a plan to sawcut control joints in the slab as soon as the concrete can be cut without ravelling at the cut. This means sawcuts MUST be made the same day as concrete placement. Don't cave on this. It's important.

Third, the sawcut depth MUST be 1/4 of the slab thickness. For a 5-inch slab, that means a 1-1/4 inch sawcut depth.

Fourth, the sawcut joints must be on a regular pattern. For your 40x40 slab, the joints should be 10 feet apart each way. Anything larger may result in uncontrolled cracking, which is not acceptable.


Good luck. It's not a simple as it appears.

 

[hokie66]

If I wanted to build a 40' x 40' polished concrete floor with the maximum amount of surety not to crack, and without joints, I would specify a 6" slab on as smooth and frictionless surface as possible, and provide 400psi of bonded post-tensioning.

 

[Guastavino]

For a 40'x40' slab have you considered putting in enough reinforcing to avoid joints altogether? It's a good amount (I think 0.5% of total area if I remember right), see ACI 360, but if it's really important, I'd consider doing that. And I would use a special mix design.

 

[BUGGAR]

I like a long wet cure to minimize immediate cracking and curling. We write that into our specs, including a mandatory wet curing time so there are no questions. We also make concrete curing a contract pay item which makes it a little more enforceable.

 

[KootK]

@Canuck67: while you've received several suggestions for improving your SOG design, I don't know that anyone's really addressed your original question regarding the tubing. So I'll stick to that.

I don't think that the tubing needs to be oriented so as not to cross the control joints. There are two reasons for my opinion:

1) Like WWM, I don't think that the tubing would have much effect until after the slab cracks form.

2) the presence of the tubing should strengthen and stiffen all sections more or less uniformly, both at the control joints and away from them. As such, I don't feel that they will compromise the intentional weakness that you're trying to design into your control joints.


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.

 

[XR250]

In my house I did 20'x20' squares and used metal keyway at the joints instead of cutting. Made each 20x20 area a separate radiant zone so no tubing crossed the joints.
Also used #4 @ 24" o.c. each way. No cracks yet.