Masonry Magazine March 2003 Page. 47
Causes and Prevention of Shrinkage
Shrinkage Cracks in Concrete Masonry
SHRINKAGE CRACKS OCCUR to alleviate internal wall stresses. Over time, concrete materials have a natural tendency to shrink and to move in response to changing temperature and moisture conditions. These movements tend to decrease the length of the wall. The degree of movement is small, typically less than an inch in a one-hundred-foot long wall, vertical cracks form at regular spacings to accommodate this movement if the ends of the wall are restrained from moving. The cracked wall then acts as a series of separate panels, each with the ability to move slightly without impacting the next panel.
To prevent shrinkage cracks, stresses within the wall must be reduced through the use of control joints. Conversely, horizontal reinforcement can be used to distribute the stresses, which results in more, but smaller, cracks held tightly together.
Control Joints
CONTROL JOINTS are planned vertical wall separations. They basically divide a wall into separate panels, similar to what happens naturally after shrinkage cracks occur.
There are several common methods to incorporate control joints (see Figure 1), and all of them share certain characteristics. First and foremost, the control joint provides a vertical bond break within the wall, which allows longitudinal movement. This is typically accomplished by eliminating mortar between adjacent units or using building paper to break the bond if units are grouted (detail D). Next, if the wall is subject to lateral loads, the joint should allow some out-of-plane load transfer under wind or other lateral loads. This can occur by using a preformed joint filler (detail A), filling the space between "flanged" units with grout and providing a bond breaker to allow longitudinal movement (detail D), by using smooth dowels with a bond breaker of grease or plastic sleeve across the joint (detail E), or by using special control block units (detail F). In addition, for exterior walls, the joints must be weather-tight. Using backer rod and sealant in the mortar joint spaces accomplishes this.
Nonstructural reinforcement, such as horizontal joint reinforcement used to control shrinkage cracking, should be discontinued at the control joint. Any reinforcement crossing the joint effectively restrains the wall and may negate the benefit of the joint. Structural horizontal reinforcement, such as bond beam reinforcement at floor and roof diaphragms, however, may be required to be continuous across the control joint to maintain the structural integrity of the wall.
Control Joint Location
FOR WALLS THAT WILL include control joints, they should be placed where stress concentrations (and, therefore, cracks) are expected, such as:
* at changes in wall height or thickness (including at pilasters);
* at movement joints in adjacent foundations, floors, and roofs;
* near door and window openings if adequate vertical and horizontal reinforcement is not provided around the opening (at one side of an opening less than six feet wide, and at both sides of openings wider than six feet); and
* near corners and intersecting walls (locate control joint within half the typical control joint spacing for the wall).
For walls without openings or other stress concentration points, control joint spacing has traditionally relied on specifying low moisture content units and on using various combinations of horizontal reinforcement. Prior to the 2000 edition of ASTM C 90 Standard Specifications for Loadbearing Concrete Masonry Units, low moisture content was specified by requiring a Type I moisture controlled unit. The intent was to provide designers an assurance of units with lower moisture content to minimize potential shrinkage cracking. However, there are several limitations to relying on moisture content alone since there are other factors that influence shrinkage which are not accounted for by specifying a Type I unit, as will be discussed later. Additionally, Type 1 units were not always inventoried by concrete