U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590

Skip to content U.S. Department of Transportation/Federal Highway AdministrationU.S. Department of Transportation/Federal Highway Administration


Report on the Diagnosis, Prognosis, and Mitigation of Alkali-Silica Reaction (ASR) in Transportation Structures

Appendix B Diagnosis of Alkali-Silica Reaction (ASR)

The Cracking Index (CI) Method

B.1 Introductory Remarks

The Cracking Index (CI) is obtained from a crack mapping process that consists in the measurement and summation of crack widths along a set of lines drawn perpendicularly (i.e., parallel and perpendicular to the main restraint(s)) on the surface of the concrete member investigated (Figure B1). The purpose of the method is to quantify the state of cracking on an area of the concrete member. It is applied to structures exhibiting symptoms of internal deterioration of the concrete (multidirectional cracking, map cracking), especially those suffering from ASR.

When carried out for the first time, the Cracking Index will establish a baseline of the extent of cracking in a structural member, while allowing comparative ratings between structural members. Periodic measurements of the CI will generate data on the evolution of deterioration and could thus be used to establish the frequency of monitoring (based on estimate of the expansion rates) and further actions to be taken. Structures that exhibit crack widths in excess of those tolerable may at some point be subjected to more detailed investigations, which may include a structural analysis to determine their integrity.

B.2 Material and Equipment Required

The method is simple to use and require the following basic material/equipments:

  • Template for drawing the required axes.
  • Pocket size crack comparator (small transparent ruler with width marks ranging from 0.10 to 2 mm) (0.004 to 0.08 in), or graduated micrometer magnifying lens (magnifying power 10 to 20 X, micrometer graduated from 0 to 10/20 mm (0 to 0.4/0.8 in) by increment of 0.05/0.10 mm (0.002 to 0.004 in)).
  • Photo camera.
  • Drawing material (marker) suitable for concrete and resistant to the environment of the structure (humidity, UV).
  • Stainless steel bolts or gage studs with a machined "demec point" at the end and installation equipment required (yard/meter stick, cordless drill, rapid-set cement or epoxy glue, etc.) for marking the corner of the reference grid.

B.3 Laying Out of the Reference Grid

The location(s) of the reference grid(s) is to be chosen so as to represent the cracking pattern present on the structure (or parts of it). In choosing the grid location, one also has to take into account the main cracking system, the shape of structural member, the access ways, the need to insure the reference grid's integrity by protecting it from the potential aggressions of the environment and from vandalism.

The reference grid is made of four axes graduated in tenth of a meter (0.1 m) (4 in): two parallel vertical axis and two parallel horizontal axis of same length (or two axis parallel and two perpendicular to the main restraints in the case of reinforced concrete members) (Figure B1). The laying out and drawing of the reference grid is to be done with caution and with care not to alter the openings of the cracks. The concrete member can be cleaned using appropriate means prior to the drawing of the reference grid. The axes are drawn on the concrete member with a template or a meter (yard) stick (Figure B2-A). It is highly recommended that demec points be set in the concrete in the corner part of the reference grid (Figure B2-A). This has two objectives: 1) ensure that the CI readings be repeated at the exact same location during subsequent field operations, and 2) allow in-situ expansion measurements.

Figure B1. Sketches showing the proposed location of reference grids for cracking index measurements. The number of horizontal and vertical measuring lines should be such that a total of 2 meters of line-measurements should be done in each direction. (Corresponding dimensions in the figure above are as follows: 0.5 m = 20 in., 2m = 80 in.

This figure shows five sketches of various geometric shapes showing schemes for laying out reference grids on concrete surfaces likely to be found in the field. The five structures illustrated in this sketch include a bean, a bridge deck, an abutment wall or pavement, a rectangular or square column and a cylindrical column.

Figure B2. (A and B). Laying out of the reference grid on a concrete median barrier affected by ASR. (C). Four small reference grids are drawn on the exposed surface of the foundation block of reinforced concrete columns of a highway bridge structure affected by ASR. (D). Reference grid on the exposed surface of a reinforced concrete column of a highway bridge structure affected by ASR. (E). Expansion measurement at the surface of the above concrete column. (F). Cracking Index measurements.

Photo A shows a man holding a yard stick drawing out a reference grid on a map-cracked Jersey barrier with rule and marker. Photo B shows a square divided into 4 equal triangles projected onto the surface of a jersey barrier with corners labelled beginning at the bottom left hand corner and continuing clockwise around the square: O, A, B, C. Photo C shows 4 square contiguous reference grids superimposed on the side of a severely cracked foundation block. Photo D shows a square superimposed onto surface of a cylindrical column. The measurement area includes localized longitudinal cracking. Photo E shows a close-up of a worker holding a gauge measurement device and is recording distance measurements between two fixed points on a severely cracked concrete column. Photo F shows two men next to a map-cracked jersey barrier. One man holds a handheld magnifying glass close to the side of the barrier while the second man is recording data.

The size of the reference grid will depend on the size of the structural member as well as on the gauge length of the length-change measuring device available. Ideally, 0.50 m (40 in) long gauge lengths will be used and thus 0.50 m (20 in) reference grids will be drawn on the member (Figure B1 and B2-B). Smaller (e.g., 0.20 m (8 in) long) lengths could also be used for smaller members or as a function of the length comparator available (Figure B2-C to B2-E). In order to generate a statistically representative assessment of the extent of cracking through the CI method, it is recommended that a minimum of two CI reference grids, 0.5 m (20 in) in size, be drawn on the surface of the member to be evaluated. This will provide a total of eight, 0.50 m (20 in)-long measuring lines (or a total of 2 m (80 in) of measuring lines along each direction). A larger number of smaller-size reference grids will however be required to provide the same statistical evaluation of the cracking system.

The demec points to be installed in the corner of the reference grid consist of approximately 5-7 mm (0.2-0.3 in) diameter by approximately 15-20 mm (0.6-0.8 in)-long stainless steel bolt or gage studs with a machined "demec point" at the end (see Appendix D). In order to insert the gage studs into the concrete, holes are dry drilled into the concrete member. The holes are just large enough to insert the gage stud fairly tight with rapid-set cement or epoxy glue. The head of the gage stud should appear flush with the surface of the concrete member.

B.4 Selection of Structural Members

Cracking Index measurements should be done on the member(s) showing the most severe degree of deterioration and/or on the most critical structural members of the structure that show signs of cracking. For each type of member, the CI should be carried out on the portions of the members exposed to moisture as cracking will normally be more widely developed. As mentioned before, cracking is usually most severe in areas of structures where the concrete has a constantly renewable supply of moisture, such as close to the waterline in piers, from the ground behind retaining walls, beneath pavements slabs, or by wick action in piers or columns. Concrete members undergoing AAR and experiencing cyclic exposure to sun, rain and wind, or portions of concrete piles in tidal zones often show severe surface cracking resulting from induced tension cracking in the "less expansive" (due to alkali leaching/dilution processes, variable humidity conditions, etc.) surface layer under the expansive thrust of the inner concrete core. It is consequently in the exposed portions of the members affected by ASR that the surface cracking is likely to provide the best estimate of the expansion reached to date, i.e., the closest to the current volumetric expansion in the affected member (see Section 5.5.2).

As the extent of expansion in a structural member is very much a function of the physical restraint to which it is subjected to, it is thus critical that CI measurements be taken parallel and perpendicular to the main restraints. Figure B1 provides examples of CI readings for different types of structural concrete members. In the case of reinforced concrete beams and columns, CI readings are taken on the exposed surface in the longitudinal and the two transverse axes of the members.

The Cracking Index is used to estimate the expansion to date in the affected member (see Section 5.5.2). This estimate is used as a criteria for the selection of the time left before proceeding with mitigation/remediation work; the evaluation is based on the number of years left before AAR-induced expansion would have reached a level of about 0.20 percent, level at risk for the stability of the reinforcing steel in the structure. This is assuming that the expansion thus estimated is similar to that reached in the reinforcing steel. One other critical factor is thus the quality of the concrete-steel bonding.

In the case of square or rectangular shaped reinforced concrete members (such as columns and beams), the most prominent cracks are often those appearing above the external bars of the reinforcing cage (i.e., close to the edge of the member). Since the steel-to-concrete bond is likely to be most affected at that level, it is recommended that: 1) the reference grid for the CI measurements stay within (i.e., not covering) those external cracking system, or perhaps better, 2) cover the whole section of the member with the reference grid but not accounting for the larger corner cracks in the calculation of the CI (as the crack width might not be representative of the concrete expansion because of the loss in the concrete/steel bond).

B.5 Measurement of the Cracks

Each crack of width greater than 0.05 mm (0.002 in) is identified and measured with the crack comparator or the graduated micrometer magnifying lens (Figure B2-F). The values obtained for each 0.1 m (4 in) segment of each axis are reported in the corresponding cell of the measurement table (example given in Figure B3). Main cracks can be reported on a sketch if desired.

Each crack is to be measured at least with 0.10 mm (0.004 in) accuracy starting at 0.05 or 0.10 mm (0.002 to 0.004 in) depending on the precision of the measuring device used) where it crosses the axis of the reference grid or at the nearest adequate location if the orientation of the crack differs too much from its general direction at the point of intersection. If the opening of the crack is sealed, swelled or awkwardly shaped, the measurements shall be taken at the nearest adequate location as well. The opening is always measured perpendicularly to its general direction, not following the length crossing the axis. A crack that crosses the axes of the reference grid several times is noted and measured as many times as it crosses an axis. However, one will try to avoid this type of configuration as much as possible when laying out the reference grid. "Crazing" cracking (very fine map-cracking) is not counted and measured, but its presence is reported.

If the measurements taken are part of a periodical inspection, the values obtained are immediately compared to the previous data to avoid gross errors in readings and in transcription. The measurement sheet is to be completed with the date, climatic conditions, and any useful information. Finally, a picture of the reference grid is taken.

The values of C.I. are given separately for the vertical and horizontal measurements. This figure shows a square with the following letters on each corner (starting top right, going clockwise): A, C, O, B. The left and bottom part of the square have incremental lines with numbers alongside it, representing sections where cracks are monitored. Beneath the square is a table showing an example of the measurements taken while performing this method.
Figure B3. Example of measurement of the Cracking Index. The values of C.I. are given separately for the vertical and horizontal measurements.

B.6 Timing / Frequency of Measurements

The frequency of the measurements is adapted to the particular needs of each case. The period can go from a few months in the case of recent and heavily damaged structures to a few years for older and well preserved ones. As a general guideline, bi-yearly (i.e., twice a year) measurements should be taken for the first 3 to 5 years and then every five years if the evolution of the damage is slow or nil.

Because of the effect of temperature and humidity on crack widths, CI readings should be carried out and repeated under very similar conditions of sun exposure, outdoor temperature and humidity conditions. Consequently, it is important that weather conditions at the time of reading be recorded as precisely as possible to allow selecting similar conditions for the next reading(s). Ideally, CI measurements will be best performed during cloudy conditions (i.e. limiting extended exposure to sun) with temperatures ideally ranging between 20 and 25 °C (68 to 77 °F). CI readings should also be taken following 24 to 48 hours of dry conditions, i.e., 24 to 48 hours following exposure to rain; this is proposed to allow stabilization of hygrometric conditions in the surface portion of the concrete member. The establishment of such "standard" weather conditions for the crack measurements will also allow better comparison between results obtained by different organizations (e.g. DOT's).

B.7 Handling of the Results

For each of the four axes of the reference grids, one calculates the total width of the openings, the average width of the openings, and the average opening width per meter (40 in) of axis. These values can be completed with a column graphic showing the width distribution of the openings.

B.8 Calculation of the Cracking Index

As mentioned before, for fair size structural members, at least two 0.50 m (20 in) reference grids will be drawn on the member from which crack mapping will be carried out. This will provide a minimum of 4, 0.50 m (20 in)-long measuring lines for each of the two directions of the grid (for a minimum of 2 m (80 in) of measuring lines). Smaller (e.g., 0.20 m (8 in) long) lengths could also be used for smaller members, but the number of reference grids required will be increased accordingly.

Since the extent of cracking (and of AAR expansion as well) greatly depends on the direction of the main restraints, the Cracking Index is calculated separately for each of the two directions of the reference grid, i.e., by taking the average of the four average opening widths measured in each of the two directions, and normalizing to a meter.

B.9 Estimate of the Expansion Reached to Date

The expansion reached to date by the concrete for each of the two or three dimensions (in the case of concrete columns or beams for instance) is estimated from the opening of all cracks intersected along each direction of the axis of the reference grid, as long as the concrete/steel bond remains appropriate or by eliminating the width of cracks above corner reinforcement where there are doubts about the quality of the bond.

PDF files can be viewed with the Acrobat® Reader®
Updated: 11/09/2015
Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000