Forward

This Guide was prepared by the staff of the Brick Industry Association (BIA) with the assistance of several members and a consultant. Instrumental in the development of the content were:

• Brian Trimble, formerly BIA staff, now International Masonry Institute, Pittsburgh, PA
• Ted Corvey, Pine Hall Brick Company, Winston-Salem, NC
• Steve Jones, Pave Tech, Prior Lake, MN Mark Smallridge,
• Mark Smallridge and Associates, The Colony, TX

Mr. Smallridge is a paving consultant with considerable expertise and experience in segmental paving design and construction. His assistance was especially helpful. The members of the BIA Paving Committee also assisted in the review of this Guide.

Introduction

The concept of segmental pavements (including clay bricks, concrete pavers, stone cobbles, stone and timber sets) is not new. Surviving segmental pavements can be traced back to the Romans and their network of roads, and to even earlier times. The oldest references to brick surfaced streets are believed to date back to Mesopotamian times, 5,000 years ago. Brick roads and streets provide a stable riding surface that is very durable. Brick has been used as a road paving material in this country since the late 1800's and was used through the early 1900's when the first national network of roads was constructed. Although the use of brick as a road surfacing material decreased with increasing vehicle speeds, and the improvement of concrete and asphaltic paving materials and construction methods in the 1930's and 1940's, it is still in use today. The rebirth of central business districts and new festival marketplaces demands that street and pedestrian areasbe made of materials with a more human scale. The appearance, strength, warmth, and flexibility of brick meet this challenge.

Present day practices enable bricks to be set using three basic methods when they are constructed as a pavement surfacing material. These are the sand set, bituminous set and mortar set methods. The former two methods are flexible in nature and are covered in this publication. They are able to accommodate surface applied loads and environmentally induced stresses without the need for discontinuities such as movement joints. Mortar set brick surfaces are considered to be rigid, and require regular placement of movement joints. They are covered in other publications, including BIA Technical Notes on Brick Construction 14 Series.

Pavement design methods consider two different types of pavement. These are "flexible" pavements and "rigid" pavements. Flexible pavements spread the surface applied loads to the underlying layers by load distribution. The materials generally require lower strength properties with increasing depth, because the stresses reduce as the load is spread over a wider area. Rigid pavements spread the surface applied loads by flexure. Rigid pavements include a portland cement concrete slab and brick pavers set in mortar. A rigid pavement's thickness is frequently less than that of an equivalent flexible pavement.

Prior to the publication of the first edition of this Guide in 1993, the structural design of brick pavements had generally been based upon empirical methods. Their use had been primarily for pedestrian areas and streets subject to only light traffic. The Brick Industry Association recognized the need for a rational approach to provide designers with the means of proportioning thicknesses of brick pavements for heavy vehicular applications such as major roads and streets. The American Association of State Highway and Transportation Officials (AASHTO) publication Guide for Design of Pavement Structures was revised in 1993 and provided a nationally accepted method that could be adapted to consider the flexible brick surface. Research on flexible pavements had demonstrated the equivalency of the brick and sand bed construction with other pavement materials considered in the AASHTO Design Guide. Therefore, it was only necessary to develop a suitable layer coefficient for the brick and sand bed in order to apply the AASHTO flexible pavement design methodology to the brick surfaced pavement.

Information in this Guide is based on research and development carried out around the world.

The previous version of this Guide provided instruction that was very comprehensive; however, many users expressed a need for a simplified method of designing flexible brick pavements. This version of the Guide provides such a method based upon specific applications and site conditions. The design solutions were prepared in accordance with the AASHTO method, and the input values are declared in the relevant sections of this Guide. There was also an identifiable need to include consideration of rigid pavement where portland cement concrete was mandated below the brick surfacing, as this was not covered in the previous version. In addition to these revisions related to the AASHTO design method, this edition also discusses an alternative design method for use in those states that have adopted the Caltrans design method or derivations thereof.

This Guide is intended to aid in the proper design, specification, and installation of brick paving systems. The designs presented in this Guide are not intended as a replacement to the advice of an experienced pavement designer. Although the proposed pavement sections may be appropriate in developing preliminary sections and budget costs, it is recommended that an engineer with appropriate pavement experience certify the final design.

Click on images for larger view.

Flexible brick pavements, as defined in this Guide, consist of sand set or bituminous set brick pavers over layers of conventional pavement materials. The flexible brick pavement shown in Figure 1 consists of a compacted subgrade beneath a subbase layer, base layer, and setting bed surfaced with brick pavers and jointing sand. A subbase may not always be necessary between the subgrade and the base. An edge restraint is provided around the flexible brick pavement as part of the system. Sand set brick pavers, and to a lesser extent bituminous set brick pavers, with sand filled joints, develop interlock between adjacent pavers, which distributes the applied loads into the underlying layers. This does not occur with mortar set pavers. Mortared brick paving is only used over a concrete slab and is not covered in this Guide. Although this pavement type has been used successfully, the emphasis in this Guide is on flexible wearing surfaces. Information on other types of brick pavements can be found in BIAA Technical Notes on BrickConstruction 14 Series.

Interlock is a phenomenon that occurs in segmental pavements as a result of the interaction of the pavers and the jointing sand between the pavers. The tight, sand-filled joints transfer loads between adjacent brick pavers through friction. Interlock increases over time as the joint sand becomes thoroughly compacted and debris builds up in the joints. When interlock is present, the wearing surface contributes to the strength of the system. Specially-shaped pavers provide little additional contribution to vertical interlock. However, some bond patterns, such as herringbone, help to distribute horizontal loads. In areas subjected to heavy vehicular traffic, the brick pavers may be required to have a minimum thickness to achieve sufficient interlock.

Figure 1. Flexible Brick Pavement

Adequate design and construction results in three types of interlock: vertical interlock, rotational interlock, and horizontal interlock. See Figure 2. If a vertical load were applied to a single brick in a pavement without vertical interlock, that brick would be forced down between adjacent bricks, transmitting concentrated stresses onto the setting bed. Brick pavers that are compacted into the setting bed and have well consolidated sand in the joints between them provide shear resistance in the wearing surface. Thus, the load is spread over a wide area of setting bed. See Figure 2. Sand set pavers develop greater vertical interlock than bituminous set pavers as they are vibrated to compact the sand bed and densify the joint sand to a higher degree.

Figure 2. Vertical Interlock

If a load is applied asymmetrically to an individual brick, the brick may rotate, displacing the setting bed and adjacent bricks. Rotational interlock holds the brick in place while rigid edge restraints prevent the bricks from moving laterally, thereby eliminating rotation. See Figure 3.

Figure 3. Rotational Interlock

Figure 4. Horizontal Interlock

Horizontal interlock is not achieved if horizontal movement is allowed. In vehicular traffic areas, horizontal braking, cornering and accelerating forces try to move pavers along the road; this is known as creep. Sand filled joints and an interlocking bond pattern transfer these forces within a paving area to rigid edging. See Figure 4. Loads created by turning vehicular traffic are distributed more evenly in all directions by a herringbone pattern than by running bond pattern, which has acceptable horizontal interlock in only one direction. See Figure 5. Basket weave patterns may have continuous joints in two directions, resulting in unacceptable horizontal interlock. Sand set brick pavers initially develop greater horizontal interlock than bituminous set brick pavers as the joint sand is better compacted.

Figure 5. Horizontal Load Interlock

Although a flexible brick pavement provides a durable surface for light or heavy vehicular applications, the surface of a segmental pavement may not provide a smooth ride at high speeds. Brick roads tend to slow traffic as subtle variations in the surface cause decreasing ride comfort as speed increases. These variations help reduce speeds in areas where faster vehicular traffic may be a concern. This is one of many traffic calming measures that cities use to slow traffic in residential neighborhoods. However, segmental pavements are not recommended where vehicle speeds exceed 40 mph (64 kph).

Deciding on the appropriate brick paving system to use is important to ensure proper performance. Since brick can be used in a variety of ways, as shown in Figure 6, Table 1 is provided to assist in selecting the most appropriate system.

Figure 6. Brick Paving Systems

The AASHTO methodology was described in detail in the first edition of this Guide. Direction was given on determining the amount of traffic that would use the pavement, dependent on the many factors identified in the 1993 AASHTO Design Guide. Tables were provided for calculating the estimated traffic in the design lane based upon axle loads, equivalency factors, structural numbers, growth factors and lane distribution factors. The equations used to calculate the required structural number for the pavement, and to proportion the individual pavement layers were set out, along with figures and tables to assess the layer and drainage coefficients necessary to design the pavement. In addition, nomographs and design examples were provided to clarify the procedure.

This version of the Guide provides a method based upon specific applications and site conditions rather than the more complex procedure of the first edition. The design solutions were prepared in accordance with the AASHTO methodology, and the input values are declared in the relevant sections. Although the proposed pavement sections may be appropriate in developing preliminary sections and budget costs, it is recommended that the final design be certified by an engineer with pavement design experience.

In addition to these revisions related to the AASHTO design method, this edition also discusses an alternative design method for use in those states that have adopted the Caltrans design method or derivations thereof. Design solutions are not provided using this method, but gravel equivalent factors are suggested for use with the Caltrans manual, so that it can be adapted to consider brick pavers.

Because of the range of climates and the variability of soils, base and subbase materials, designers must use good engineering judgment in detailing a flexible brick paving system. The designer should be acquainted with site conditions and use all available resources to create a cost-effective solution. Consult references listed at the end of this Guide for more information.

Table 1 Brick Paving System Selection   [top]

The subgrade is classified by the existing soil conditions, the environment and drainage. The more accurate the subgrade classification, the better the performance of the pavement.

Table 2 Subgrade Categories   [top]

Table 3 Subgrade Categories from USCS

Soil Conditions

The existing soil conditions for a project should be determined prior to commencing the design of the pavement sections. A geotechnical engineer who specializes in site investigation work will generally test and classify the soil conditions for the project area. Testing will be carried out at the project site and samples will be recovered for additional testing in the laboratory.

The site investigation should include test pits and borings along the alignment of the road or street, or over the area of the pavement. Samples should be collected for the following laboratory tests considered essential to determine the engineering properties of the soils over which the pavement will be constructed:

• Grain-size distribution: sieve analysis test and hydrometer test to determine the percentage of the individual grain sizes in the sample;
• Atterberg Limits: consistency tests to determine the moisture content at which the sample changes from a semi-solid state to a plastic state (Plastic Limit) and from a plastic state to a liquid state (Liquid Limit);
• Natural moisture content: test to determine the in-place moisture content of the soil;
• Natural Density: test to determine the in-place density of the soil;
• Dry density/optimum moisture content relationship (standard or modified): compaction test to determine the ideal moisture content to achieve the specified state of compaction;
• Strength tests [California Bearing Ratio (CBR) or Resistance Value (R-value)]: mechanical tests to determine the bearing capacity of the soil for use in the design.

The CBR test method is set out in ASTM D 1883 Test Method for CBR of Laboratory Compacted Soils (AASHTO T193). It can be conducted on treated and untreated base, subbase and subgrade materials. This test is a comparative measure of the load-bearing capacity of a soil. It measures the load required to drive a standard plunger a set depth into a sample of soil at a standard rate of penetration. The CBR value is the ratio of the load measured in the test and the load used to achieve the same penetration in a standard sample of crushed stone. The test is generally undertaken in the laboratory, but in-place tests can also be carried out. The values are greatly affected by the degree of compaction and the moisture content of the specimen. The test should be conducted on a specimen compacted to a density representative of the material to be used in the pavement, or alternatively at a range of densities likely to be encountered. To represent the materialÕs potential moisture condition in the pavement, the test should be conducted on specimens that have been soaked after compaction for a period of four days.

The R-value test method is set out in ASTM D 2844 Test Method for Resistance R-Value and Expansion Pressure of Compacted Soils (AASHTO T190). It can be conducted on treated and untreated base, subbase and subgrade materials, but the test can only be undertaken in the laboratory. For base, subbase and non-expansive granular soils, the R-value is determined at a density equivalent to the density used during construction. For cohesive soils and expansive granular materials, the R-value test involves two separate procedures. One procedure calculates the estimated thickness of the overlying pavement layers required to maintain the state of compaction of the material. The other procedure estimates the thickness of the overlying pavement layers required to prevent plastic deformation in the material. The R-value is determined at the moisture content and density at which the thickness of overlying materials is similar in the two procedures.

The AASHTO design method uses the resilient modulus (MR) as the design input for the subgrade properties. The mean value of all test results for each pavement section or soil type should be used for design. The test to directly determine MR is not widely used, so AASHTO has proposed the following relationships between the CBR and R-value test results. These relationships are as follows:

MR (MPa) = 10.3 X CBR                  (Eq. 1)
Where: MR is the resilient modulus and CBR is the California Bearing Ratio

MR (MPa) = 6.9 + 3.8 X R-value       (Eq. 2)

The Caltrans design method uses the R-value as the design input for the subgrade properties. The lowest R-value should be used for design over a section of pavement. However, if there are one or two significantly lower R-values in a localized section, consideration should be given to replacing these areas with better material and using the next lowest value.

Soils or subgrades are typically classified into different groups to represent their engineering properties. There are several systems used in the United States, but the two most common are the Unified Soil Classification System (USCS), used for general engineering purposes, and the AASHTO System, used for highway engineering purposes. The USCS is set out in ASTM D 2487 and the AASHTO system is set out in AASHTO Standard M145, Classification of Soils and Soil-Aggregate Mixtures for Highway Construction.

In the USCS, the soils are classified in twenty-five groups by two letter designations dependent on the soil type and physical properties. The first letter represents the main soil type (gravel, sand, silt, clay or organic), and the second modifies the first letter based upon the grain-size distribution for granular soils or the Atterberg Limits for cohesive soils. Eleven groups have paired designations. In the AASHTO system, the soils are classified in seven main groups (and in twelve sub-groups) based upon their grain-size distribution and Atterberg Limits. The two systems are given in Tables 3 and 4, in a manner that correlates each group with the suggested subgrade categories (U -unsatisfactory, P - poor, F - fair, G - good, E - excellent).

A geotechnical engineer’s report will include a description of the soils encountered at the project site and will set out the test results. In addition, it will generally provide recommendations on the strength properties of the soils to be used for design and may contain some design options for the pavement section. If a recommended value is given in the report, this should be used to select the subgrade category (see Table 2). If there is no recommended value, but CBR or R-value test results are given, the average of these values should be used for the AASHTO design methodology. When the design is to be undertaken using the Caltrans adaptation, the minimum R-value should be used for selecting the subgrade category. If only the subgrade’s USCS or AASHTO classification is known, Table 3 or 4 can be used to estimate the subgrade category from the column titled “Average”. The remaining columns are addressed in the following section.

Environmental conditions and the quality of subgrade drainage can have a major effect on the support offered by the subgrade. In wet climates, poorly drained areas, or those that experience freezing conditions, the subgrade support is likely to be reduced during certain periods of the pavement's life. Conversely, in arid climates or well-drained areas, it is likely that a higher degree of subgrade support will be experienced during part of the pavement's life. These factors can have a significant effect on the performance of the pavement.

Saturation of the subgrade, and the materials in the pavement section, can lead to premature distress as this condition reduces the strength of these materials. Water can enter the pavement through the joints between bricks, through cracks and joints in the bound base materials, or from a high ground water condition. The amount of water penetrating from the surface depends on the regional climate. Fluctuations in moisture content can also be problematic, leading to changes in volume and load support. Rapid removal of water from the pavement is therefore an important design objective, and a positive drainage system should be considered. Design of such a system is beyond the scope of this Guide.

Table 4: Subgrade Categories from AASHTO

Figure 7 presents the six climatic regions experienced in the United States. It depicts two regions of hard freezes where spring thaw conditions are likely to affect the subgrade, two regions where there is a potential effect from frost, but only if the pavement is thin, and two regions not susceptible to frost. Average depths of frost penetration are indicated for the eastern and central states, although frost depths can vary locally based on many factors. Local data should be used in its place if this is available for a specific project site. This is particularly true in the western states, and no frost depth data is therefore included in Figure 7 for this part of the country. If the depth of frost penetration is greater than the pavement thickness determined in this Guide based on one of the first three columns of Tables 3 and 4, it will be necessary to revise the pavement construction. Either non-frost susceptible material should be added to the thickness of the pavement section so that it is thicker than the depth of penetration, or a revised depth should be determined based upon the anticipated loss of strength in the subgrade. In this case, if the project is located in Regions III or VI on Figure 7, it will be necessary to use the subgrade category from the column for Frost Environmental/ Drainage Conditions. If the project is located in Regions II or V, and the project site drains poorly, it will also be necessary to use the Frost Environmental/ Drainage Conditions column to determine the subgrade category, since, once again, pavement thickness is a factor.

The AASHTO design method utilizes an effective resilient modulus that is derived from the seasonal resilient moduli. These vary depending on the moisture conditions of the subgrade during each season. Such an analysis is beyond the scope of this Guide. It is recommended that the design be undertaken using a subgrade category as described earlier or from the geotechnical report, as the geotechnical consultants will consider these factors when providing their recommendations. If these values are not available, but the USCS or AASHTO designation is known, Tables 3 and 4 should be used to develop the appropriate subgrade category. Most pavements should be designed using the subgrade category from the column for Average Environmental/Drainage Conditions. However, if the project is located in Regions I, II or III of Figure 7, and the project site drains poorly, such that the subgrade is frequently saturated, it will be necessary to use the subgrade category from the column for Wet Environmental/Drainage Conditions. If the project is located in Regions IV, V or VI and the project site drains well, such that the subgrade is rarely saturated, it may be appropriate to use the column for Dry Environmental/ Drainage Conditions.

The traffic analysis for the project should be undertaken before commencing design of the pavement sections. A traffic engineer is typically contracted for this work. When undertaking a design it is necessary to determine the existing (or initial) traffic volume using the road, and to estimate the future traffic volumes over the analysis period. Based upon these data and local experience, it is necessary to establish the traffic flow in each direction and in the design lane. Most of the damage to a pavement is caused by truck traffic; passenger cars, pick-ups and light two axle trucks generally have a negligible effect. Using local data on the anticipated types of vehicles that will use the road, the number of load applications of each axle group can be calculated. Next, all of the repetitions of each axle group are converted into the equivalent number of repetitions of one axle load condition.

In the AASHTO design methodology, the traffic is represented as the equivalent number of load applications of an 18-kip axle load that represents the mixed traffic using the pavement. This is known as an equivalent single axle load (ESAL). Tables in the AASHTO Design Guide provide values for converting different axle loads into ESALs.

In the Caltrans method, this number of ESALs is further converted into a traffic index (TI). This varies in accordance with Equation 3 below, except that the TI is rounded to the nearest 0.5:

TI = 9.0 X (ESAL/106)0.119           (Eq. 3)

In this Guide, a simplified approach is adopted using pavement classes. Nine pavement classes are identified in Table 5, together with a description of their anticipated use and an indication of the total number of ESALs and TI for each class.

The life of the pavement can be expressed in a number of ways depending on local policy. Two terms are used in the AASHTO Design Guide covering different long-term strategies. These are the performance period and the analysis period. The performance period is the length of time that a pavement will remain serviceable before it requires rehabilitation such as an overlay. The analysis period is the amount of time over which the pavement life is to be considered, including any rehabilitation work. For high volume roads (collectors and above), an analysis period of at least thirty years is frequently considered. For low volume roads (locals and below) and all other pavements, a twenty-year life is generally acceptable. As rehabilitation of flexible brick pavements cannot be achieved by strengthening measures without lifting the bricks, this Guide considers the analysis period as the design life of the pavement.

The AASHTO Design Guide is based upon empirical test results from full-scale road tests conducted in the 1960’s. As a result of the measured behavior of the test sections, the researchers developed a performance equation upon which design of new pavement sections can be achieved. The equation relates the design life, in terms of 18 kip ESALs, to a number of different input parameters. These include the reliability of the pavement, the acceptable level of loss in serviceability, the variability of the traffic predictions and performance, the structural number and the average subgrade resilient modulus.

The reliability of the designed pavement section is an important feature in the AASHTO design process. The reliability of the pavement is the probability that it will perform satisfactorily over its design life for the traffic and environmental conditions experienced. The reliability level adopted in this Guide is taken as an 85 percent likelihood that the pavement will reach its design life (analysis period) for pavements with high traffic volumes (collectors and above); and a 75 percent likelihood that the pavement will reach its intended design life for pavements with low traffic volumes (locals and below). Non-highway pavements are also considered to have a reliability of 75 percent. This means that 85 percent or 75 percent of the pavements for each respective use would achieve or exceed the design life.

The AASHTO design method uses a subjective measure of the loss of serviceability and failure of the pavement. It was developed as an interpretation of the quality of the ride experienced by the average road user. A scale from 0 to 5 represents the quality of the ride and is known as the Present Serviceability Index (PSI). A PSI of 0 represents an impassable road while a PSI of 5 represents a perfect road. The change between the initial and final (terminal) PSI, known as the Serviceability Loss, used in this Guide is taken as 1.7 for high traffic volumes and 2.2 for low traffic volumes. This is based upon an initial value of 4.2 and terminal values of 2.5 and 2.0 respectively. This compares favorably with those used for typical flexible pavements. An exception is in pedestrian areas where the potential for trip hazards is an important consideration. Therefore, a serviceability loss of 1.7 is recommended in these locations. Other AASHTO reliability parameters are presented in Table 6.

The variability of the traffic prediction and pavement performance is taken as 0.35. This is comparable with the figures used for flexible pavements using asphaltic concrete as a surface course.
The structural number is the only parameter directly related to the pavement section. It is derived from the layer coefficient of each layer, the thickness of each layer, and the drainage coefficient for each layer. This Guide is produced under the assumption that adequate drainage will be provided to the pavement materials such that the latter coefficient can be taken as 1.0. Typical layer coefficients are presented in Table 7, with the default values used in subsequent design tables in this Guide.

The values presented in Table 7 can be used to adjust the layer thicknesses derived from the design tables. The ratio of the layer coefficients can be used to determine the equivalent thickness of an alternative material. For example, to include a 6-in. (150 mm) thick lime stabilized subbase it is possible to reduce the aggregate base by 0.11/0.14 times the 6-in. (150 mm) thickness, i.e. by 4.5 in. (114 mm). Similarly, to replace 8.5 in. (216 mm) of graded crushed aggregate (CBR 100) with graded aggregate (CBR 60) multiply 8.5 by 0.14/0.12, i.e. replace with 10 in. (254 mm) of graded aggregate. However, it is recommended that the top of the aggregate base directly under the paver setting bed always be constructed with graded, crushed material that is 4 in. (100 mm) thick when the ESALs are below 500,000 or 6 in. (150 mm) thick when the ESALs are at 500,000 and above respectively.

Tables 8 through 11 of this Guide have been prepared to provide design solutions to each pavement class and subgrade category. The wearing surface is always a 2-5/8 in. (67 mm) paver on a 1 in. (25 mm) setting bed. The bituminous setting bed is usually specified at 3/4 in. in thickness, and an adjustment to the developed thicknesses is required as discussed in the bituminous setting bed section. The base is as indicated in the tables.

Table 8 presents the thickness of graded aggregate subbase course required for each application. The resultant pavement section will be comprised of 2-5/8 in. (67 mm) thick flexible brick surface on 1 in. (25 mm) of bedding sand, over 4 in. (100 mm) of crushed, graded aggregate base on top of the thickness of graded aggregate subbase determined from the table. Note that the thickness of crushed, graded aggregate base is increased from 4 in. to 6 in. (100 to 150 mm) for traffic levels over 500,000 ESALs. An unbound base course is not considered appropriate for traffic levels above 2,000,000 ESALs.

Similarly, Table 9 provides the required thickness of graded aggregate subbase when a cement treated base is used under the bedding sand. Note that in this case the thickness of cement treated base is increased from 4 in. to 6 in. (100 to 150 mm) for traffic levels over 2,000,000 ESALs. Table 10 can be used when an asphalt treated base is provided.

Table 11 provides typical portland cement concrete slab thicknesses, with a 4 in. (100 mm) aggregate subbase below and a wearing surface of flexible brick paving. This table is for guidance if the bricks are to be used over such a substrate. Little structural benefit is provided by the bricks in this pavement section, and care needs to be exercised in ensuring that detailing allows for thermal and moisture induced movement in the concrete, and for egress of moisture penetrating the brick surface.

CALTRANS Design Concepts   [top]

The Caltrans design method is also based upon a wide range of information including: theory, test track studies, experimental pavement sections, observations of pavement performance, and research on materials. Pavements are generally designed for a twenty year life, but it is accepted that asphalt concrete surfaced pavements will require maintenance at ten to fifteen years if they are to achieve this life. This is generally a surface material issue, and this Guide assumes that the brick pavers will provide a twenty year life.

The Caltrans design method considers the various pavement materials in terms of a gravel factor (Gf). Tables are included in the Caltrans design manual setting out the gravel factors for various materials dependent on the materials properties and the TI. The gravel factor is a representation of the relative ability of the materials to resist the effects of traffic loading, when compared to an equivalent thickness of gravel.

The design of the pavement section is based upon a relationship between the R-Value (R), and the Traffic Index (TI) to develop the Gravel Equivalent (GE) for the pavement. The relationship is represented by Equation 4:

GE (mm) = 0.975 X (TI) X         (100-R) (Eq. 4)

The procedure is carried out from the top of the pavement to the bottom. Treated base layers generally have an R-value greater that 100 and so the equation is typically applied to the highest layer in the pavement section with an R-value less than 100. The thickness of the overlying layers is determined. The process is then used for the underlying layer, and so on down to the subgrade. The thickness of each layer is calculated by dividing the GE by the appropriate Gf. The thickness of each layer is generally rounded to the next 0.05 ft (15 mm) increment.

To allow for deviations from the specified thickness as a result of construction procedures the Caltrans method uses a safety factor procedure. This involves adding 0.2 ft (60 mm) to the GE of the asphalt concrete surface material and subtracting 0.2 ft (60 mm) from the GE of the subbase, or if no subbase is used, from the thickness of the base. The thickness of the brick pavers cannot be changed, and so this practice needs to be undertaken between the base and the subbase, if used.

Design solutions using the Caltrans procedure are not presented in this Guide, however, a Gf of 2.0 is proposed for the brick paver and sand setting bed, and 1.8 is proposed for the brick pavers on a bituminous setting bed, based upon the above noted typical relationship with equivalency to asphalt concrete. This value can be used when the Caltrans manual is appropriate for the design, rather than the AASHTO methodology.

Design of Bituminous Setting Bed Systems   [top]

Bituminous setting bed systems as shown in Figure 6b can be used in most of the same applications as sand setting bed systems. Higher speed apllications are less desirable as the interlock between pavers is reduced. Although it has reduced structural benefits, the system can provide better moisture protection to the underlying layers. In addition, the bonding action of the system enables the use of pavers with a lower standard of dimensional tolerances where wider joints would lead to reduced “lock-up” in a sand set system. Joint widths of up to 1/4 inch (6mm) can be tolerated, especially in low traffic applications. As there is no vibration used to compact the pavers, chipping is less of a problem, particularly with pavers that do not have chamfers or lugs. This system may also have advantages where edge restraints are less reliable, or where movement may be encountered. This typically occurs where pavers are placed against steel rails for light rail applications.

The bituminous setting bed system can be used as an alternative to a sand setting bed system. The 1 in. (25 mm) thick sand setting bed is replaced by a 3/4 in. (19 mm) thick asphalt coated sand mixture that is “bonded” to the underlying pavement layer using a tack coat. The brick pavers are bonded onto this layer with a rubberized asphalt adhesive. The joints are filled with stabilized sand, but no vibration is used. Consequently, “lock-up” is not as well established and the load spreading is reduced. The base thicknesses presented in this Guide can be used for this system; however, the thickness of the underlying layers needs to be increased. This can be achieved by adding an additional 1 in. (25 mm) to the thickness of the cement treated base layer, 3/4 in. (19 mm) to the thickness of the asphalt treated base layer, or 1-1/2 in. (38 mm) to the thickness of the graded, crushed aggregate base. No revision is necessary for the portland cement concrete sub-slab option.

Mortared Brick Paving   [top]

The structural design of mortared brick paving follows the design of rigid pavements. The brick pavers and the mortar setting bed are not taken into account in the thickness design. The design of mortared brick paving is not the aim of this Guide. Refer to the AASHTO Design Guide or BIA Technical Notes 14 Series.

Detailing
Surface Profile  [top]

Satisfactory slopes for flexible paving must be provided to avoid ponding water. A minimum slope of 2 percent, (1/4 in. per foot or 1 mm per 50 mm), is suggested for all exterior brick paving. Crowns on roads usually provide adequate slope. A maximum slope of 10 percent is recommended for flexible brick streets and roads, since larger slopes will cause washout of the jointing sand and braking vehicles will increase the creep of the pavement. Surface grades of up to 15 percent, or even 20 percent, can be used on pavement areas subject to slow moving traffic or pedestrians. However, joint sand stabilization, as well as a high level of installation quality, is desirable to reduce creep that occurs.

Drainage  [top]

Drainage is one of the most important design requirements, since improper drainage may cause failure of the pavement, erosion of the base or subbase, possible deterioration of the pavers, or slippery pavements. Drainage needs to be considered at three levels in the pavement. These are at the surface, to the setting materials and to the pavement structure and subgrade. Surface drainage is undertaken in accordance with standard design concepts for pavement areas. The pavement surface should be finished 1/8 to 1/4 in. (3 to 6 mm) above drainage gratings to allow for potential secondary compaction of the setting bed under trafficking. Surface profiles are covered in the previous paragraph. The bond pattern may affect the flow rate of water over the surface of the paving, as water tends to flow along the joint lines. Surface runoff will increase with time as the joints become filled with debris, however, some water will penetrate the brick surface layer.

Water that penetrates the wearing surface should be drained away from the setting bed and base when the underlying layers are not free-draining. This is particularly the case when the setting bed is placed over a portland cement concrete slab or a cement or asphalt treated base. Weep holes placed vertically through the portland cement concrete slab may be necessary depending on the environmental conditions. Drainage is less of a concern with bituminous setting beds as some water will percolate through them. It may be necessary to provide a sub-surface drainage system. Sub-surface drainage weeps should be provided at low points and at the edge restraints to drain water to the pavement edge or storm drains. A perforated pipe wrapped with an appropriate geotextile material may be used. The geotextile is necessary to keep small particles from washing out of the setting bed into sub-surface drains. A drainage layer of open graded aggregate may also be used, but requires proper planning, designing and specifications. Design and detailing of such systems are not included in this Guide, and the reader is directed to the references for several manuals on this subject.

Bond Patterns  [top]

Many different bond patterns exist, providing different aesthetic effects, a few of which are shown in Figure 8. Herringbone provides the best resistance to the horizontal forces from accelerating, braking and turning of wheels, and should be used in areas subjected to heavy vehicular traffic. This is required for sand setting beds and recommended for bituminous setting beds. The pattern can be oriented at 45 degrees or 90 degrees to the direction of traffic. It is not necessary to turn the pattern at corners and bends, as the horizontal interlock is good in all directions.

Many of the pavements laid in the 19th and 20th Century were laid in running bond, either directly across the streets, or occasionally at up to 45 degrees across them. Running bond patterns have continuous joints in one direction. They do not transfer loads well along the continuous joints, and so careful consideration is necessary with their use. They require smaller joints between pavers in order to minimize creep. Running bond pattern is not recommended for high volume roads and streets. For low traffic volume paved areas the traffic should run perpendicular to the direction of the continuous joints.

Other bond patterns, such as basketweave and stack bond, can be used in pedestrian areas. This also applies to derivations of these patterns, such as Spanish bond. In all of these arrangements, the patterns include continuous joints in two perpendicular directions. For most basket weave patterns laid in a flexible pavement the bonding ratio of the paver is 2:1 or 3:1 (length: width) for proper alignment of the pattern. Herringbone bond, running bond and stack bond do not have to follow this rule, although joints will not align with a herringbone bond using an irregular module, which may affect installation quality. Basketweave and stack bond patterns tend to show irregularities of the pavers and misalignment of the bond pattern more than other bond patterns. Great care needs to be exercised with such patterns if there is any likelihood of in-place wheel turning.

Some pedestrian plazas and other facilities have been installed using the brick pavers in modules that coincide with the building grids, tree pits, or other features. Other designs have been created in herringbone bond or bands using different color pavers. When setting out a pattern using fixed dimension modules or different colors, it is important to remember that the individual bricks are not exact sizes. In a modular pattern a consistent 1/16 in. (2 mm) under or over sizing of the bricks, allowable in standard manufacturing tolerances, will soon lead to a shortfall or overrun of several inches in a grid module. This should only be overcome by cutting the bricks to fit, as spacing the pavers with wider joints can affect the structural integrity of the pavement. When mixing different colors of pavers in a pattern area, it is necessary to have an understanding of the potential for different sizes so that the desired appearance is achieved.

Edge Restraints  [top]

Edge restraints are necessary along the perimeter of the pavement to prevent lateral movement of the pavers and loss of the setting bed. The edge restraints should be able to resist anticipated loads with minimal movement in order to maintain interlock. Edge restraints can be placed before laying the setting bed, and those incorporating concrete should be cured before compaction of the brick pavement begins. All edge restraints should be placed to a depth of at least the bottom of the setting bed. Edge restraints are required for both sand set and bituminous setting methods, although the former requires more robust construction. It is important that the inside face of the edge restraint is vertical so that the pavers can be laid against it without a tapered joint that will reduce the integrity.

Concrete Dividers and Inlays  [top]

In many pavement areas, the brick pavers are laid between concrete elements that divide the pavement into sections. These are typically 8 to 16 in. (200 to 400 mm) wide. The perception is frequently that this will provide the opportunity to change the pattern orientation or that by incorporating such fixed features it will be possible to prevent creep of the brick pavers. In actuality, it introduces a discontinuity into the pavement that creates a weakness within the traffic area. Placement of small cut pieces, opening of the joints, and settlement of the pavers often occurs at these locations. Therefore concrete dividers are not recommended. Changes in the pattern orientation can be formed by incorporating a header or sailor course of bricks if a band type feature is required, by a single sawcut joint line, or by a carefully introduced series of staggered sawcut joints to maintain continuity.

Inlays are also frequently used where a panel of brick pavers is incorporated as an entrance feature. Such inlays are usually surrounded by concrete dividers, or portland cement concrete pavement. When detailing such an inlay it is important that the sides of the dividers are vertical so that interlock can be generated between the bricks and the concrete. The brick pavers should be finished approximately 1/8 in. (3 mm) high against the concrete so that there is accommodation for secondary settlement of the setting bed under traffic. It is also advisable to keep the cut pieces of brick against the edge divider as large as possible, with no pieces less than a quarter of a paver. Thin slivers are particularly vulnerable to damage at these positions. Some benefit can be gained by incorporating a header, sailor, or string course adjacent to the concrete edge.

Traffic Buttons, Reflectors and Paint Markings  [top]

Traffic buttons are frequently used as lane markings in streets and roads where snow clearance is not an issue. These buttons and reflectors are generally secured on the pavement surface with an epoxy adhesive. Setting onto individual brick pavers can be detrimental to the pavement. The impact from vehicle tires can loosen and even dislodge the bricks. It is recommended that alternative means of lane markings are adopted for brick pavers, or that larger, low-profile buttons are used that fix to more than one paver. One effective solution has been to inlay concrete pavers that have a specialized top surface finish. As concrete pavers are typically 3-1/8 (80 mm) thick, they protrude above the pavement surface sufficiently to create tire feedback to indicate their presence.

There are several different types of traffic marking methods for the pavement surface. These include adhesive strips, paints, and thermo-plastics. When in service, the individual brick pavers continue to move independently of each other to a slight degree. As such, the joints open and close a small amount when a wheel passes over them. This movement is often sufficient to cause cracking of the adhesive strip and thermo-plastic markings. Although thinner paint markings frequently have a shorter service life on other pavement materials, they may be more cost effective on brick paved surfaces. Where the visibility of traffic markings is not a legislated requirement, inlaying contrasting colored pavers can provide the most durable option.

Movement Joints  [top]

Pavement materials expand and contract as a result of temperature and moisture changes. Flexible pavement materials, such as aggregate, asphalt concrete and cement treated base materials, distribute this movement over the entire pavement areas such that localized strains are very small. As a result the pavers are unaffected by the underlying movement. Brick pavers behave similarly, with any movement taken up in the joints between pavers without putting stress on the brick or edging. Expansion joints are therefore not typically required in such flexible brick pavements. However, when a portland cement concrete slab, used under the flexible brick paving wearing surface, expands and contracts, the movement is concentrated at the joints between the slabs. This movement will be reflected into the overlying pavers, which can be detrimental to the pavement. When the concrete contracts, it causes the overlying joints to open. This results in a loss of interlock, settling of the pavers and a loss of integrity. When the concrete expands it causes the joints to close. This can impose large horizontal pressure into the brick paver layer, possibly causing paver movement that may result in chipping and spalling of the pavers, and in extreme conditions heaving of the surface. It is therefore necessary to continue expansion joints through the brick and setting bed layer by incorporating edge restraints on either side of the joint. This should be applied to both sand set and bituminous set pavers.

Portland cement concrete slabs are also provided with contraction joints to control cracking during curing. Movement at these locations is less than at expansion joints, and it is not normally necessary to reflect the joints into the surfacing if the underlying contraction joints are at less than 10 ft (3 m) centers. In order to distribute the movement over a wider area it is beneficial to cover the control joints with a strip of geotextiles under the sand bedding course. This is not done under bituminous setting materials.

Design Examples

Examples are available in the Adobe® Acobat version of this file. To download the file click here.

Part II: Materials
Introduction  [top]

The performance of any pavement is only as good as the base, subbase and soil on which it is laid. Quality materials for every layer in the pavement system are vitally important to good performance. Materials should conform to state or local department of transportation (DOT) specifications, ASTM standards, or other applicable industry standards. Project specifications typically require submission of qualifying tests set by the standards.

Base and Subbase Materials  [top]

Some pavement systems contain only a base, while others contain a base and a subbase. See Figure 1 for location of layers. The quality or type of base and subbase materials is usually dictated by the design requirements. The design tables are based on various CBR values for the aggregate layers, and compressive strength or Marshall stability for the cement and asphalt treated layers.

Base materials may consist of unbound granular materials, such as crushed aggregate; cement-treated or asphalt-treated aggregates; or concrete and asphalt bases. Subbases are usually composed of aggregate materials. Aggregate base and subbase materials, including cement- and lime-stabilized materials, are commonly specified in local state and municipal standards for highway construction. All materials should conform to state or local DOT specifications. Materials conforming to ASTM or other industry standards can be used as alternates. The choice and quality of base and subbase materials influences the performance of the pavement. Typically, each layer material can resist progressively higher stresses from the subgrade upward to the wearing course.

Aggregates  [top]

The National Stone Association has provided gradation limits for base and subbase aggregate materials, see Table 12. This table is similar to requirements found in ASTM D 2940 Specification for Graded Aggregate Material for Bases or Subbases for Highways or Airports.

Crushed, quarry processed aggregate is preferred because of its ease of construction. The maximum size of aggregate to be used in construction may depend on the size of the project and the size of equipment being used. Proper gradation of materials is required to achieve adequate compaction. Layers consisting of single-size aggregate will not consolidate during compaction and should not be used.

For flexible brick pavements subjected to pedestrian and light vehicular traffic, aggregate graded to “3/4 minus”, similar to the gradations in Table 12, is usually sufficient as a base material because it is easy to work with and is readily available. Smaller graded aggregates or rounded aggregates may not be sufficient to achieve interlock within the aggregate layer and will not transfer loads properly. Open graded (gap graded) aggregates can be used to promote water drainage in areas subjected to frost heave to minimize damage. Geotextiles may be needed to prevent intrusion of smaller material into the open graded aggregates.

Asphalt Bases

New or existing asphalt bases can be used for flexible brick pavements. Specification of asphalt concrete should follow industry standards or local DOT requirements. The adequacy of existing asphalt and the materials beneath should be verified.

Concrete Bases  [top]

New and existing concrete bases can be used for flexible brick pavements. New concrete slabs should be specified, with reinforcement as needed, and constructed according to industry practice. Concrete bases should be properly cured before installing the flexible brick paving. High early strength cement may be used to reduce the time before the wearing surface is placed. Existing concrete slabs should be checked for appropriate strength and repaired or reinforced as necessary. A geotextile can be used where there is a possibility of loss of sand through cracks or holes in the existing slab. The adequacy of the materials beneath the existing concrete slab should be verified.

Soil and Base Stabilization  [top]

Subgrade soils or granular material unsuitable for use alone may be treated to produce a stronger layer. The subgrade soil may be stabilized by adding portland cement or lime, depending on the quality of the soil. Subbase and base materials may be improved by adding portland cement, lime, asphalt or pozzolanic materials. Modifying unsuitable materials is considered when economically feasible or where suitable untreated materials are in short supply, although caution should be used in specifying treated soils. Their use should be based on local availability and experience.

Subgrade soils or granular material unsuitable for use alone may be treated to produce a stronger layer. The subgrade soil may be stabilized by adding portland cement or lime, depending on the quality of the soil. Subbase and base materials may be improved by adding portland cement, lime, asphalt or pozzolanic materials. Modifying unsuitable materials is considered when economically feasible or where suitable untreated materials are in short supply, although caution should be used in specifying treated soils. Their use should be based on local availability and experience.

Aggregate subbase materials, as well as cement- and lime- stabilized materials, are commonly specified in local state and municipal standards for highway construction.

Setting Beds
Sand Setting Bed   [top]

Sand used as the setting bed should be a washed, well-graded, sand with a maximum size of about 3/4 in. (4.8 mm). Sand conforming to ASTM C 33 Specification for Concrete Aggregates is acceptable. Table 13 shows the gradation limitations taken from ASTM C 33. In addition, the amount of material passing the 75 µm (No. 200) sieve should be limited to no more than 3 percent. The sand particles should be sub-angular. For pavements subjected to heavy channelized traffic, experience has shown that only naturally occurring, washed silica sand with no silt content should be used. The gradation for the silica sand in channelized traffic should be as shown in Table 14 and no more than 0.3% passing the 75 µm (No. 200) sieve.

An excess of fine particles can increase the moisture sensitivity of the bedding sand. Bedding sands with high fines content can lead to rutting and movement forms of distress. It is not only important to ensure that the fine content is satisfactory on the selected bedding sand, but also that it will not break down under heavy traffic. A degradation test can be undertaken on the bedding sand to compare different bedding sand options. The test involves rotating sand samples