chitectural appearance. They can be the cause of leaks from wind-driven rain. Or they can be the first telltale signs of a po- tential wall collapse. Thus no crack should go unex- amined. For every crack, a trained investigator should re c o rd, with photographs and sketches, the following i n f o r m a t i o n : Location on the building Pattern (horizontal, vertical, straight diagonal, or stepped diagonal) L e n g t h Width (uniform or tapered; if tapered note how) Depth (through paint, plaster, or entire wall) Age (clean crack indicates new; coated with paint or dirt indicates old) Moving crack or dormant crack Based on this information, an investigator can assess the cause of the crack, how harmful it is, and how best to repair it, if repair is deemed necessary. The location and pattern of cracks are the most important in deter- mining what caused them. In the drawings shown h e re, the causes of some typical cracks in brick and block masonry are explained, based on the location and pattern of the cracks. Measuring crack width To measure the width of cracks, most investigators use a crack comparator although a graduated magni- fying device is more accurate. Both of these devices are shown in Figure 1. The crack comparator is easy to use, sufficiently accurate for most jobs, and usually is supplied free by firms specializing in failure investiga- tions. The graduated magnifiers cost $50 to $100. When is a crack too wide? Whether a crack is too wide depends on the materi- al, the type of building, the climate, and the type of crack. If the owner spared no cost on the buildings ar- chitectural design, cracks that are 0.01 inch wide may be totally unacceptable. But a homeowner with a 20- y e a r-old brick house would be happyand luckyto have cracks that are only 0.01 inch wide. A c c o rding to the Portland Cement Association (PCA), a crack that is 0.010 to 0.015 inch wide neither hurts the surface appearance nor alarms the viewer Evaluating cracks To repair cracks in masonry properly, you must first determine what caused them By Bruce A. Suprenant Figure 1. To measure the width of cracks, most investigators use a crack comparator. A graduated magnifying device is more accurate, but costs $50 to $100. The crack comparator usually is free from firms that investigate failures. (Ref. 1). Norwegian tests show that wind-driven rain doesnt enter cracks narrower than 0.004 inch (Ref. 2). Even cracks wider than this may not allow enough water penetration to damage masonry or the building. So how does one evaluate the width of cracks? Ta b l e 1 lists typical tolerable crack widths for re i n f o rced con- c rete, modified from Ref. 3. These tolerable crack widths also should be reasonable for re i n f o rced ma- sonry stru c t u res. Based on these values, a way to clas- sify and evaluate crack widths in masonry is given in Table 2. Using Table 2, once the investigator classifies the Cause: Vertical deflection of concrete beam Cause: Shrinkage of mortar and concrete masonry units TABLE 1 TOLERANCE CRACK WIDTHS, INCHES Interior exposure Dry atmosphere 0.016 Wet or moist atmosphere 0.012 Exterior exposure 0.008 Where watertightness is required 0.004 Source: Modified from Ref. 3. TABLE 2 CLASSIFICATION OF CRACK WIDTHS Classification Crack Width, CW Very fine (Watertight) CW 0.004 Fine (Exterior exposure) 0.004 <CW 0.008 Medium (Interior exposure, wet) 0.008 <CW 0.012 Extensive (Interior exposure, dry) 0.012 <CW 0.016 Severe CW > 0.016 crack, he or she also has identified the allowable expo- s u re for that crack. For example, a crack that is 0.014 inch wide would be classified as extensive and usu- ally allowed only in a dry interior wall. If a 0.014-inch- wide crack occurs in an exterior wall it pro b a b l y should be re p a i re d . Beware of the moving crack The diff e rence between an active or working crack and a dormant crack is important. An active crack may open or close, but a dormant crack has stopped moving. Dormant cracks tend to be caused by a tem- porary overload. Live cracks are created by re p e a t e d overloading or by temperature or moisture changes that continually cause the crack to open and close. If the crack is dormant, then it can be filled with a rigid filler, such as epoxy or cement. Working cracks, on the other hand, should be filled with flexible sealants. If theyre re p a i red with a rigid material, theyll only crack again. If the filler is stronger than the m a s o n r y, a new crack parallel to the re p a i red crack p robably will occur. How do you tell if a crack is active or dormant? T h e re are three ways. The easiest is to measure the crack width with a crack comparator at regular time intervals, every day or every week. Record both crack width and date of re c o rding. Always measure the crack width at the same location. Draw a line acro s s the crack to mark where you measured it. Do this at t h ree or four places along the crack. On exterior walls, use a waterproof marker. Figure 2. Caused by lateral loads, this diagonal shear crack is plastered to determine if the crack is active or dormant. If the crack is active, the patch will crack at a later date. Dormant cracks can be filled with epoxy or cement, but active cracks that are still moving should be filled with a flexible sealant. Cause: Foundation settlement or soil heave Another inexpensive method is to spot patch the crack with plaster (Figure 2). Use hot water to speed the set of the patch. Note the date the patch was placed and inspect it at regular intervals to see if it has cracked. A cracked patch shows that the crack is ac- tive. Be sure the patching material doesnt crack fro m drying shrinkage; use a nonshrink patching material. A two-piece crack monitor that sells for less than $15 also can be used to detect crack movement (Figure 3). One of the two plastic pieces has red cross hairs; the other piece has a grid system with a zero mark at the c e n t e r. The two pieces come taped together so that the intersection of the red cross hairs on the one piece co- incide with the zero mark on the grid of the other p i e c e . The monitor is laid across a crack and each end is at- tached to the wall with epoxy or a fast-setting glue. Af- ter the adhesive cures (about 15 minutes for fast-set- ting glue; about 24 hours for epoxy), the tape holding the two pieces together is cut. Now, if either side of the crack moves, one of the plastic pieces moves. The re d c ross hairs, originally at zero, slide over the grid sys- tem indicating the amount of horizontal and vertical crack movement. Both the amount and direction of movement can be observed. Usually the monitor is checked at the same time each day. Evaluate cracks early If cracks arent examined, evaluated, andif neces- s a r y re p a i red, they can grow from an eyesore into a costly headache. The larger cracks get, the more wind- driven rain they let into the wall. Water in turn can c o r rode metal components, cause eff l o rescence, and if it freezes it can spall the masonry. Some cracks can even lead to veneer or wall failures. Cracks are a warn- ing that shouldnt be ignored. References 1. Building Movements and Joints, 1982, Portland Cement Association, 5420 Old Orchard Rd., Skokie, IL 60077. 2. O. Birkeland and S. D. Sevendsen, Norwegian Test Methods for Rain Penetration through Masonry Walls, Sym- posium on Masonry Testing, STP 320, 1963, ASTM, 1916 Race St., Philadelphia, PA 19103. 3. Control of Cracking in Concrete Structures, ACI 224, ACI Manual of Concrete Practice, Part 3, 1989, American Concrete Institute, P.O. Box 19150, Detroit, MI 48219. 4. Clayford T. Grimm, Masonry Cracks: A Review of the Lit- erature, Masonry: Materials, Design, Construction, and Maintenance, STP 992, 1988, ASTM. 5. R. E. Copeland, Shrinkage and Temperature Stresses in Masonry, Journal of the American Concrete Institute, Feb- ruary 1957. Bruce A. Suprenant is a consulting engineer, an Adjunct As- sociate Professor at the University of Colorado at Boulder, and a contributing writer to this magazine. He also took the photographs for this article. Figure 3. Laid across a crack and adhered to the wall on both sides of the crack, this two-piece crack monitor detects crack movement. If the crack moves, the red cross hairs on one of the two pieces becomes offset from the center of the grid on the other piece. P U B L I C AT I O N# M 9 0 0 0 7 6 Copyright 1990, The Aberdeen Gro u p All rights re s e r v e d