They come in all configurations, shapes, and widths: from gigantic to nearly invisible, from invisible to the naked eye but visible under a microscope to invisible even under a microscope—that is, the nonstructurally induced cracks. With little fear of ever being wrong, we can say “all concrete cracks.” All cracks are the result of stress relief due to either positive or negative strains that exceed the concrete's strength. They can be produced by differential volume changes where there is restraint to surface drying and carbonation shrinkage by the concrete below, or expansion of a concrete “body” but not its “skin” such as due to alkali-aggregate reactions. Cracks can occur at a wall that shrinks without impediment except for restraint imposed to its base such as by friction or anchors, so it cracks vertically and not horizontally.
Control joints simply tell slabs where to crack. As a good rule of thumb, joints should be centered (in feet) about two to three times the slab thickness (in inches). So, a 4-inch-thick garage slab should have joints that create minimum 8x8-foot sections or maximum 12xl2-foot sections. Whether to use the two- or three-multiplier rule depends upon the anticipated magnitude of drying shrinkage as dictated by water-cement ratio and aggregate properties. Not all concrete shrinks the same amount.
Joints can be full depth, saw-cut to a depth of about one quarter the slab thickness, or formed by embedding a thin piece of wood or metal to that depth (such as zip strips). Full-depth joints may be in harm's way if the edges of slabs move up and down independent of each other when loads roll across them. This happens when subbases are not fully compacted, so load transfer dowels (smooth bars) are used to span across joints to provide continuity that accommodates the rolling loads. Cracks below saw-cut or tooled joints usually go around aggregate particles, providing an inter-fingering known as “aggregate interlock,” which restrains differential vertical movements of adjacent slabs and allows them to vertically move together.
Post-tensioning of slabs “pulls” concrete “together” as it shrinks, and often prevents visible cracks. But shrinkage usually begins on Day One, and slabs may crack before tensioning.
Shrinkage compensating concrete still shrinks, but the shrinkage is restrained by placing reinforcing steel such that it is put in tension by a slight early concrete expansion. That tension in the steel results in concrete that is in compression, which accommodates and eliminates cracks. Recently, admixtures have been produced that reduce drying shrinkage so that larger joint spacings can be used.
Hidden from view are microcracks. An optical light microscope or a scanning electron microscope can be used to see these “invisible” cracks. When these microcracks join up, damage may occur.
A major problem with cracks is that they allow aggressive chemicals, such as chloride salts and carbon dioxide, access to reinforcing steel where corrosion of the steel may occur. Unfortunately, the “better” the concrete, the more it will crack because it is the cement paste that shrinks. So the higher the cement content at a given water-cement ratio, the greater the drying shrinkage. Some bridge decks are being built today with high cement contents to make concrete denser and thus less permeable to corrosion-causing chloride salts and carbon dioxide. But they then crack. On some days it doesn't pay to get up.
Cracks are also caused by expansive chemical reactions within concrete such as sulfate attack, alkali-silica and alkali-carbonate reactions, corrosion of embedded metals, unsound cement, or unstable aggregates. Unstable aggregates are those that are “soft” and do little to restrain normal concrete drying shrinkage, or those that shrink upon drying or expand upon wetting. Interestingly, sometimes it is difficult to determine if cracks resulted because concrete expanded or shrank!
If you want to unravel the mystery of why your nonstructural concrete cracked, your best option is to consult your competent, friendly petrographer.
Bernard Erlin is president of The Erlin Company (TEC), Latrobe, Pa., and has been involved with all aspects of concrete for over 47 years.
William Hime is a principal with Wiss, Janney, Elstner Associates and began working as a chemist at PCA 53 years ago.