Controlling Cracks in Slabs with Unusual Panel Geometries

We have all heard the saying “concrete can be counted on to do two things — get hard and crack.” Without thoughtful human intervention that saying might be true, but today slabs-on-ground can be engineered to resist unplanned cracking. Some slab shapes, however, are very susceptible to cracking, and that’s the subject of this article.

The shape and size of a slab or panel has much to do with susceptibility to cracking. Square shapes, for example, resist cracking the best, while rectangles with long, narrow sides and odd-shaped panels with interior corners are the most at risk.

This is an example of poorly laid out joints for a parking lot. The contractor was lucky to not have cracks crossing the many acute angles.In terms of slab panel sizes, the American Concrete Pavement Association (ACPA) says a safe formula to use for laying out crack control joints is to multiply the thickness of a slab by 2 when placed directly on the subgrade or on unstabilized subbase, or by 1.75 when placed on a stabilized subbase, expressed in feet, to determine the maximum joint spacing. For instance, if a slab on subgrade is 6 inches thick and placed directly on grade, joints should be located every 12 feet or less. But regardless of slab thickness, panels shouldn’t exceed 15 feet, unless special conditions are present. The ACPA publishes two booklets which provide the guidance for joint layout. These books can be purchased on ACPA’s website www.acpa.org/bookstore. Use the search box to locate documents IS006 and EB237.

There are some joint configurations and panel shapes that require special attention. These include joints coming together in the shape of a "T," long and narrow rectangular panels, panels with an inside corner, and locations where two or more joints intersect at odd angles. The focus here is to provide information about how to lay out joints to redirect stresses that develop in concrete as it shrinks during the drying process. Some suggestions for designing concrete mixtures with reduced shrinkage are also included.

T-shaped joints

Fig. 1 If T-shaped joints can't be avoided, drill a hole at the intersection of the joint.As a general rule, T-shaped joints should be avoided because cracks that form along the stem of the T tend to continue across the intersection of the T. There are times, however, when this type of joint can’t be avoided. In this case there are a couple of strategies that can be used individually, or together, to mitigate the potential for a crack to migrate across the cross-bar of the T. One option involves drilling, coring or forming a 3/8-inch-diameter or larger hole (see Figure 1) at the intersection at least twice the depth of the saw cut, but preferably the full thickness of the slab.

Fig. 2 If T-shaped joints cannot be avoided, provide supplemental reinforcing perpendicular to the terminating joint.When the location of a T-joint can be predetermined, placing supplemental reinforcement just above the intersection point of the T, and parallel to the cross-bar, can reduce the potential for a crack to migrate across the bar and into the adjacent slab. Follow these suggestions:

• Cut two lengths of #4 (1/2 inch diameter) rebar 3 feet long.
• Locate the first bar 1 inch below the paving surface and place it 1 inch from the intersecting joint, parallel to the top of the T.
• Place the second bar parallel to the first bar and 1 inch beyond (see Figure 2).

Narrow panels

Fig. 3 For long, narrow panels, place rebar at 0.5% of the cross sectional area of a slab.It’s best to avoid panel shapes that exceed ratios of 1.5-to-1; 1-to-1 ratios are the best for crack control — without preventative measures cracking tends to form a square pattern. There are times, however, when narrow panels are unavoidable, so providing additional rebar reinforcement can help reduce the opening width of intermediate cracks that tend to form across the narrow dimension. If shrinkage in narrow panels isn’t controlled with rebar reinforcement, the cracking that develops is most likely to form square panels within the rectangular shape. The recommended amount of reinforcement to provide between joints is 0.5% of the cross sectional area of a slab. For example, if a slab is 5 or 6 inches thick, place #4 rebar 1 inch below the surface at 6 inch on-center intervals parallel to the longest sides of the panel. If a concrete section is 7 to 8 inches thick, place #4 bars at 4 inch on-center intervals (see Figure 3). This won’t stop cracks from occurring but instead of one wide crack the reinforcement will distribute the shrinkage across many hairline cracks that are hardly visible. The supplemental reinforcing should be discontinued at joints. Narrow panels shouldn’t exceed the length of panels around them.

Inside corners

Fig. 4 With inside corners, if there are no joints going to the corner, place two #4 rebars 3 ft. long diagonal to the corner an inch away.Cracks often develop diagonally off inside corners. The approach to controlling these cracks depends on whether there are crack control joints ending at the corner. If there are no joints then you can place two pieces of #4 bar 3 feet long at 1 inch below the surface at 45 degree angles to the sides of the corner — the first bar being an inch from the corner and the second bar an inch beyond the first bar (see Figure 4). When there is one joint ending at the inside corner, place one bar parallel to the joint and the leg of the corner (see Figure 5).

Fig. 5 With an inside corner, when there is one control joint meeting the corner, place reinforcement as shown.If there are two joints ending at the corner, place an “L” shaped #4 bar 3 feet long at 1 inch away from both joints. Include a second bar 1 inch beyond the first one (see Figure 6). For all applications it’s important to place the rebar close to the corner, where cracks first form. In order to maintain the sawcut depth at the corner for sawed contraction joints, it may be necessary to saw into the formwork or to plunge the saw slightly at the corner.

Fig. 6 When two control joints meet at an inside corner, place reinforcement as shown.

When control joints intersect at an angle

Fig. 7 When two or more control joints meet at a point, no angle should be less than 60 degrees, but 90 degrees is better.There are conditions when panel or curb geometry requires joints to meet at intersecting angles. This can involve two or more intersecting joints. Lines shouldn’t meet at angles less than 60 degrees (90 degrees is considered to be the best) because concrete tends to crack across acute angles (see Figure 7). Plunge cut each saw joint at the meeting point to ensure the required sawcut depth at the joint intersection. It’s also helpful to drill a 3/8-inch-diameter hole twice the depth of the joint at the intersection, following the same guidelines used for T-shaped joints above.

When control joints meet curving form lines the joint should meet the formed edge at a 90 degree angle. Intersecting joints should be located a minimum of 18 inches from the formed edge of the slab or curb (see Figure 8). The full depth of the sawcut should be maintained until meeting the formed edge.

Fig. 8 An example of three control joints coming together at one point and meeting a formed edge.

Additional considerations

There are additional considerations that can affect whether concrete develops cracks. The condition and preparation of the sub-grade is important. If there is ground settlement or frost heaving due to lack of water drainage under slabs, the resulting concrete stresses can cause cracks.

Some contractors include high dosage levels of fiber reinforcement, steel or synthetic macro fiber, in their concrete to help prevent cracking in all panel shapes. Steel fibers placed in the range of 70 pounds per cubic yard or synthetic macro-fibers at 7-1/2 pounds per cubic yard of concrete reduce both shrinkage and curl in panels. Experimentation by some contractors suggests that greater sized panels also are possible with the use of steel or macro fibers.

This graphic demonstrates how to lay out joints under unusual circumstances.

Using well-designed concrete mixes reduces shrinkage also. Kevin MacDonald, president of Beton Consulting Engineers, Mendota Heights, Minn., says the amount of water is key when designing concrete mixes for floors and other slab-on-grade applications — not the water-cement ratio but the total amount of water added per cubic yard of concrete. Water has a direct relationship to both shrinkage and finishability. “The shrinkage factor when 250 pounds of water is added for a cubic yard of concrete is 1 and that’s acceptable. Adding 300 pounds of water yields shrinkage factors of 1.5 to 1.8 and that isn’t,” he adds. Rather than adding additional water to the mix in the field to enhance workability, both mid-range water reducers and polycarboxylate super plasticizers can be used to enhance placability while helping to control total water amounts for interior slabs and exterior slabs.

Shrinkage can also be reduced by increasing aggregate sizes to 1-1/2 inches or more and optimizing aggregate gradations because there is less total aggregate surface area to coat with cement paste. But local aggregate sources must be considered as well as slab thicknesses when making these decisions.

The amount of sand or fine aggregate has a large influence on finishability — too much sand makes concrete difficult to finish. Cellulose-based viscosity modified admixtures (VMA) can be used to replace some sand in a mix to make finishing easier.

Finally, don’t base decisions about concrete mixes by their strength alone — high strength concrete slabs usually shrink more if the higher strength is obtained by increasing the cement content.