If you are an ACPA member, please contact Eric Ferrebee for editing access to the content of this resource.

TB010 - Design and Construction of Joints for Concrete Highways

From ACPA Wiki
Jump to: navigation, search

Joint system designs for concrete highway pavement control cracking and maintain pavement structural capacity and ride quality at low annual costs. In concrete highway pavement, joints:

  • Control transverse and longitudinal cracking from restrained contraction and the combined effects of restrained curling, restrained warping, and traffic loads.
  • Divide the pavement into practical construction increments (eg. traffic lanes).
  • Accommodate slab movements.
  • Provide desired load transfer.
  • Provide a reservoir for joint sealant.

Concrete pavement joint design evolved from:

  • Theoretical studies of concrete slab behavior [1][2][3][4]
  • Model and full-scale laboratory tests [5][6][7]
  • Experimental pavements subject to controlled vehicular traffic [8][9][10][11][12][13].
  • Performance evaluations of special experimental joint designs[14].
  • Long-term performance evaluations of in-service highway pavements[15][16].

This publication presents recommendations on specific items that require consideration for jointing systems in concrete highway pavements.

Suitable and timely construction, as well as adequate design, are key elements in establishing a well perfor­ming joint system. Effective sealing is critical in main­taining proper function of the joint system.

The need for a jointing system in concrete pavements is a result of the desire to control transverse and longitudinal cracking. Cracking results from the com­bined effects of concrete drying shrinkage, temperature and moisture changes, applied traffic loadings, sub­base restraint, and certain material characteristics. In order to design a suitable pavement jointing system, the following considerations must be evaluated.

  • Environmental Conditions: Temperature and moisture changes induce movement of the slab, resulting in stress concentrations, warping, and curling.
  • Slab Thickness: Pavement thickness affects curl­ing stresses and deflections for load transfer.
  • Load Transfer: Load transfer is necessary across any concrete pavement joint. However, the amount of required load transfer varies for each joint type. Where tiebars or dowels are used, the type and size of the bars selected will be a factor in the joint design.
  • Traffic: Traffic is an extremely important con­sideration in joint design. Traffic classification, channelization, vehicle wander and predominance of edge loadings influence the load-transfer re­quirements for long-term performance.
  • Concrete Material Characteristics: Constituent materials affect concrete strength and joint re­quirements. The materials selected for the con­crete determine slab shrinkage. Coarse aggregate influences the concrete thermal coefficient. In ad­dition, inferior materials have a detrimental affect on joint performance. D-cracking and spalling are results of poor constituent materials and are con­centrated along joints.
  • Subbase Type: The support values and interface friction characteristics of different subbase types affect movement and support of the slabs.
  • Sealant Characteristics: The joint spacing will affect the type of joint sealant selected. Other considerations, such as appropriate shape factors and life-cycle costs, also affect the sealant choice.
  • Shoulder Design: The shoulder type (tied con­crete, asphalt, granular, or earth) affects edge support and ability of mainline joints to transfer load. Widened outer lanes have also been effec­tive for maintaining load transfer.
  • Past Performance: Local performance records are an excellent source for establishing a joint design. However, improvements to past designs using current technology and materials can significantly increase performance.

Natural Crack Development

Proper jointing is based on controlling cracks that oc­cur from natural actions on concrete pavement. Joints are placed in the pavement to control the location and geometry of cracking.

Shrinkage

Much of the anticipated concrete shrinkage occurs very early in the pavement life. A major source of early shrinkage is from temperature change. The heat of hydration and temperature of pavement normally peak a short time after final set. After peaking, the temperature of concrete declines due to both reduced hydration activity and lower air temperature during the first night of pavement life.

Another factor contributing to initial shrinkage results from the reduction of volume through loss of mix water. Concrete mixes for roadway applications require more mix water than is required to hydrate the ce­ment. The extra water helps attain adequate workability for placing and finishing operations. During consolida­tion and hardening, most of the excess water bleeds to the surface and evaporates. With the loss of the water the concrete occupies less volume.

Subgrade or subbase friction resists the contraction of the pavement from reduced volume and temperature. This resistance produces tensile stresses within the concrete. If not considered, the tensile stresses cause a transverse crack pattern like that shown in Figure 1.

Figure 1 Initial Cracking in unjointed pavement.

Spacing of initial cracks varies from about 40 to 150 ft. and is dependent on concrete properties, thickness, subbase friction, and climatic conditions during and after placement.

The occurrence and interval of early cracks are impor­tant. Cracking intervals are shorter with stiff stabilized subbases so there is less opening at each individual crack. Crack spacing may be much longer for con­crete placed on granular subbases or natural subgrades. With longer initial crack spacing, much greater opening and movement can be anticipated at each crack.

Gradients

Stresses due to temperature and moisture gradients within concrete can also contribute to cracking. These stresses generally occur after concrete hardening. The exposed top surface undergoes fairly large daily varia­tions in temperature and moisture content. The daily changes in temperature and moisture content are much smaller at and near the bottom of the pavement.

Curling is a result of temperature gradients through the depth of the pavement structure. Temperature gra­dients vary with weather conditions and time of day. Daytime curling arises when the top portion of the slab is at a higher temperature than the bottom. The top of the slab expands more than the bottom portion caus­ing a tendency to curl. The weight of the slab resists the curling action and induces tensile stresses toward the bottom, and compressive stresses toward the top of the slab (Figure 2). At night the stress pattern reverses. Tensile stresses are developed toward the top and compressive stresses toward the bottom of the slab.

Moisture warping is a factor which works to counteract daytime curling. Moisture warping results from a moisture differential from the top to the bottom of a slab. The top of a slab is typically drier than the bot­tom. A decrease in the moisture content causes con­traction and an increase in moisture causes expansion. The differential tends to develop compressive stresses at the base of the slab which counteract load and daytime curl developed tensile stresses.

Evaluating the combined effect of restrained temperature curling and moisture warping is complicated due to their opposing nature. Because of these and other factors, curling stresses computed from formulas which only consider temperature gra­dients are too high when compared to measured values and pavement performance. Combined curling and warping stresses measured on one research pro­ject were only half the values computed on the basis of temperature alone.[17]

Figure 2 Curling action in a concrete pavement slab.

Models will predict crack spacings of 15 to 20 ft. (or less) in the first 24 hours when accounting for stresses resulting from temperature affects only. Field ex­perience indicates that initial crack spacing is not nearly that close [18][19]. In plain pavements (15 to 20 ft. joints) intermediate joints sometimes do not crack for several weeks to months after opening the pavement to traffic. It may take several months to years for in­termediate cracks to occur in pavements with distributed steel (joints spaced 30 ft. or more). However, intermediate cracks are typically spaced from 15 to 20 ft. after they do occur.

Restrained curling and warping in combination with loads will cause additional transverse cracks between initial locations. A longitudinal crack will also form along the approximate centerline of pavements with two lanes of traffic.

Figure 3a shows the resulting natural crack pattern. Proper jointing provides a series of joints spaced to control the formation (location & geometry) of these cracks (Figure 3b).

(A) Crack pattern in unjointed concrete pavement as a result of environmental and load stresses. (B) Proper joint­ing in concrete pavements controls location and geometry of cracks.

Joint Effectiveness

Load transfer is the ability of a joint to transfer a por­tion of an applied load from one side of the joint to the other (Figure 4). It is measured by "joint effectiveness". If a joint is 100 percent effective, it will transfer one-half of the applied load. Zero percent effectiveness means that no load is transferred across the joint. Field evaluation of load transfer is made by measuring the deflections on each side of a joint from an applied load. The equation below is used to rate joint effec­tiveness, E:

2 dU

E = ------------------- 100

dL + dU

where: dL = deflection of the loaded side.

dU= deflection of the unloaded side.

Load transfer is necessary for jointed concrete pavements to perform well. Adequate load transfer lowers deflections, and reduces faulting, spalling, and corner breaks. A joint effectiveness of 75%or more is considered adequate for medium and heavy truck loadings [20].

The following factors contribute to load transfer across joints:

  • Aggregate interlock - the interlocking action be­tween aggregate particles at the face of the joint.
  • Mechanical load transfer devices, i.e. dowel bars.
  • Stabilized subbases.

Aggregate Interlock

Aggregate interlock relies on the shear interaction bet­ween aggregate particles on the joint crack faces that form below a saw cut. It is most effective for slabs built with short joint spacings and non-erodible stabilized or permeable subbases which will experience low truck volumes.

A Federal Highway Administration study concluded that the following be considered to increase aggregate interlock load transfer and minimize faulting [21]:

  • Thicker slabs (larger area for interlock provides better load transfer).
  • Shorter joint spacings (e.g. 15 ft or less).
  • Stiff subbases (higher effective k-value).
  • Edge support (e.g. tied concrete shoulders or ex­tended lanes).
  • Coarse grained subgrade soils (drainage).
  • Improved drainage (pipe collection system/permeable subbase).

Figure 4 Effectiveness of load transfer.

Slab length affects joint opening and effectiveness of aggregate interlock. Figure 5 shows how joint effec­tiveness varies with subbase type in laboratory tests [22]. However, the actual load transfer developed on in-service pavements is much higher [23]. Typical field measurements show joint effectiveness of about 58 percent for 15- to 25-year old undoweled pavements [24].

In dry, arid, nonfreeze environments, temperature variations and joint movement (opening) are small. Therefore, load transfer through aggregate interlock can accommodate greater truck volumes. However, short joint spacing is required.

Figure 5 Sensitivity of joint effectiveness to joint opening for laboratory and field data on undoweled joints [25].

Aggregate size is critical to load transfer [26]. Small (1/2 in.) aggregates provide only marginal interlock. Large (greater than 1 in.), durable aggregates are helpful in maintaining load transfer, especially for larger joint openings [27]. Generally crushed stone performs better than natural gravels because crushed aggregates create a rougher joint face. A rough face wears more slowly than a rounded smooth joint face created by natural aggregates. Likewise, early cracking increases the joint face roughness because cracks form around aggregates instead of through them.

Studies have found that, with short joint spacings, ag­gregate interlock provides acceptable joint perfor­mance when truck traffic volume is low (generally less than 80 to 120 trucks per day) [28][29]. This varies depending on aggregate type and support conditions. Other studies indicate that at least four to five million accumulated AASHTO 18,000 pound Equivalent Single Axle Loads (ESAL's) are required to produce objec­tionable faulting [30][31]. (AASHTO is the American Association of State Highway and Transportation Officials).

Mechanical Load Transfer ­Dowel Bars

Aggregate interlock, alone, will not provide enough load transfer for good long-term performance for most highway pavements due to heavy truck traffic. Dowel bars should be used to provide added mechanical load transfer where truck traffic exceeds 120 per day or the accumulated design traffic exceeds four to five million AASHTO ESAL's. Typically, this truck traffic level will require at least an eight in. thick slab. For eight-inch slabs or greater, dowels are recommended for most highway applications.

Dowel bars are smooth round bars placed across joints to transfer loads without restricting horizontal joint movement. They also keep slabs in horizontal and ver­tical alignment. Since dowels span the joint, daily and seasonal joint opening does not affect load transfer across doweled joints as much as it does undoweled joints.

Dowel bars lower deflection and stress in the concrete slab and reduce the potential for faulting, pumping, and corner breaks [32][33]. This is true for short panel lengths and longer joint spacings. Performance evalua­tions of in-service concrete highway pavements show that the use of dowels effectively reduces faulting [34][35][36][37].

Dowels also increase pavement service life by reduc­ing deflections and stresses in the slab by effectively transmitting the load across the joint. For example, a 10-in. doweled slab with 80% load transfer will have the same deflection as a 12-in. undoweled slab with only 40% load transfer. For highway pavements less than 10 in. thick, 1-1/4 in. diameter dowels are recommended. One and one-half inch dowels should be used for pavements 10 in. thick or greater. A minimum dowel diameter of 1-1/4 to 1-1/2 in. is needed to control faulting on highway pavements. Loads on smaller diameter dowels induce higher bearing stresses and cause the concrete matrix around the dowel to deteriorate or elongate (socket). Elongation of the dowel hole reduces load transfer capabilities [38][39][40][41][42][43][44]. Figure 6 shows how dowel diameters 1-1/4 to 1-1/2 in. are subject to con­siderably lower bearing stresses than smaller diameter bars.

Dowel lengths are commonly in the range of 15 to 18 in. with shorter dowels being used in shorter slabs. An embedment length 6 times the dowel diameter is ade­quate for load transfer [45][46]. To account for joint opening and construction tolerances, 2 to 3 in. is add­ed to twice the embedment length to arrive at a prac­tical dowel length. Most agencies specify 18 in. long dowels for typical highway pavements.

Figure 6 Sensitivity of bearing stress to dowel diameter with vary­ing dowel spacing [47].

Dowels are typically placed at mid-depth in the slab and spaced 12 in. apart (center to center) (Figure 7). However, some research shows that nonuniform spac­ing may provide satisfactory performance for larger diameter dowels in short slabs [48][49][50].

It is important to remember that aggregate interlock also helps load transfer where dowels are used. There­fore considerations that improve aggregate interlock should not be entirely neglected when using dowel bars. Large crushed aggregate and stronger subbases will also improve doweled joint performance [51].

Figure 7 Typical dowel placement in highway pavements.

Stabilized Subbases

Stabilized subbases reduce joint deflection, and im­prove and maintain joint effectiveness under repetitive loads. Stabilized subbases also provide an all-weather working platform for the paving contractor and a stable, smooth trackline.

Increased support provides better joint performance. Studies show that subbase strength significantly affects long-term load transfer [52][53][54]. Cement stabilized subgrades and lean concrete subbases produce greater slab support, reduce deflections, and increase fatigue life. Figure 8 shows that stabilized support more than doubles joint effectiveness and that loss of load transfer with a cement-treated subbase occurs slowly and stabilizes around one-half million load applications.

Figure 8 Joint Effectiveness with various subbase types (based on 9 in. slab after 1 million load applications).

Types of Joints

The most common types of joints in concrete pavements are:

  • Transverse Contraction Joints: Joints that are constructed transverse to the centerline and spaced to control cracking from stresses caused by shrinkage, and moisture and thermal changes. Typically transverse contraction joints are oriented at right angles to the centerline and edge of the pavement lanes. Some agencies skew contraction joints to help decrease the dynamic loading across the joint and eliminate simultaneous loading by each wheel.
  • Transverse Construction Joints: Joints in­stalled at the end of a day-long paving operation, or other placement interruption (e.g. bridge ap­proach). These joints are installed at the location of a planned joint whenever possible.
  • Transverse Expansion/Isolation Joints: Joints placed at locations that allow movement of the pavement without damaging adjacent structures (bridges, drainage structures, etc.) or the pave­ment itself.
  • Longitudinal Contraction Joints: Joints that divide lanes of traffic and control cracking where two or more lane widths are placed at one time.
  • Longitudinal Construction Joints: Joints that join adjacent pavement lanes which are paved at different times.


Transverse Contraction Joints

Transverse contraction joints primarily control natural cracking in a pavement. Proper joint design and con­struction is critical to overall pavement performance. Many problems which have resulted in pavement ser­viceability loss have occurred at poorly designed joints.

Design

Spacing — The main function of transverse contraction joints is to control the formation of cracks. Local ser­vice records are an excellent guide for determining transverse joint spacings that will control cracking. Changing nonstructural factors such as coarse ag­gregates, concrete mix design, or curing method may also impact the joint spacing. Some states have used random spacing of joints to eliminate harmonic induced ride quality problems. However, even-spaced, right-angle joints will perform well with adequate load transfer.

Figure 9 shows recommended joint spacings for a range of slab thickness and support conditions. Transverse contraction joint design includes considera­tion of the effect of longitudinal slab movement on sealant and load transfer performance. For concrete highway pavements, spacing of transverse joints should average 12 to 20 ft., depending on slab thickness. No more than 20 ft. is advisable. Figure 10 shows how transverse cracking increases with panel length [55].

Figure 9 Recommended Maximum Joint Spacing for Highway Pavements.

For reinforced (mesh-doweled) concrete slabs the max­imum advisable spacing is 30 ft. Longer slabs have a greater tendency to develop working mid-panel cracks caused by the rupture of the steel reinforcement. Studies also show that the faulting rate increases as the joint spacing increases above 30 ft. [56]. In panels longer than 30 ft., greater joint movements decrease sealant performance.

The designer should remember that the specified joint spacing has a significant influence on the joint sealant selection and reservoir design. The sealant and reser­voir must be capable of accommodating the an­ticipated panel and joint movement. Joint movement can be estimated by the following equation [57]:

ΔL = C L (αΔT + ε)

where:

ΔL= the expected change in slab length (in.).
C = the subbase/slab frictional restraint fac­tor (0.65 for stabilized material, 0.80 for granular material).
L = the slab length, (in.).
α = the PCC coefficient of thermal expan­sion (see Table 1).
ΔT = the maximum temperature range (generally the maximum concrete temperature at placement minus the minimum ambient temperature in January, °F).
ε = the shrinkage coefficient of the con­crete (see Table 2). Note: this factor should be eliminated on rehabilitation projects, where shrinkage is no longer a factor).

When applying this equation, it is important to allow safety factors for deviations between design and field movement.

Skewing — Skewed joints are a variation of transverse contraction joints often used in plain pavements without dowels. A skewed joint is a transverse contraction joint skewed to 4 ft in 24 ft. Orientation of the skew places the obtuse angle at the outside pavement edge on the leave side of the joint (Figure 11). Each wheel on an axle crosses a skewed joint at a separate time. This alternation of loading reduces stresses and deflections of the concrete slabs. Potential for pumping and faulting is reduced.

Table 1.

Typical Values for PCC Coefficient of Thermal Expansion (α)(26)
Type of Coarse Aggregate PCC Coefficient of Thermal Expansion (10-6/°F)
Quartz 6.6
Sandstone 6.5
Gravel 6.0
Granite 5.3
Basalt 4.8
Limestone 3.8

Table 2.

Typical Values for PCC Coefficient of Shrinkage (ε)(26)
Indirect Tensile Strength (psi) PCC Coefficient of Shrinkage (in./in.)
300 (or less) 0.0008
400 0.0006
500 0.00045
600 0.0003
700 (or greater) 0.0002


Skewing is not a substitute for dowels or mechanical load transfer. Skewing is effective for undoweled joints on low-volume routes. Load transfer dowels are recom­mended for highway pavements with significant truck traffic.

Some agencies specify joints that are skewed and doweled for highway pavements. This design presumes that in conjunction with dowels, skewing will enhance load transfer. However, conclusive evidence of this presumption has not been found in field studies [58].

When dowel bars are specified, skewing the joint should be a contractor option. Time savings is a primary benefit of allowing the contractor to use available dowel insertion equipment. Dowels should always be placed parallel to the centerline of the pavement.

Construction

Good construction practices are required to obtain op­timum load transfer capabilities. Both aggregate in­terlock and mechanical load transfer require attention during construction.

Aggregate interlock relies on proper consolidation and uniformity of the concrete. Since these factors are also vital to other important concrete properties, such as strength and durability, the contractor generally observes these factors very closely. As a result, good aggregate interlock is achieved in most cases.

Placement — A contractor can place dowels using mechanical insertion or load transfer assemblies (dowel baskets). Mechanical insertion requires an attachment to the slipform paver which places dowels by an inserting and vibrating process. Load transfer assemblies are wire frames or cages which hold and support the dowels at the proper depth and alignment.

Dowel placement is important and requires careful verification soon after paving operations begin. Align­ment and location tolerances provide limits beyond which joint performance might be hindered, but allow some latitude for the heavy construction environment. Construction operations may need adjustment for variances due to the field conditions.


Dowel assemblies in place prior to paving.

Dowel bar inserter in operation.

Location and alignment are very important regardless of how dowels are placed. Dowels should be oriented parallel to the pavement centerline and surface of the pavement. Tolerances of ±1/4 in. per 12 in. of dowel length in the horizontal, vertical, or combined alignments are acceptable. Maintaining dowel align­ment within recommended tolerances helps ensure good joint and pavement performance.

Dowel Movement — Dowels require sufficient lubrication to permit movement of the concrete along the dowel surface. An application of a parafin based lubricant, an asphalt emulsion, a concrete form oil or a standard grease will provide excellent lubrication. Some states recommend that lubrication provide a maximum pull-out resistance of 200 pounds per dowel. However, actual field conditions often vary.

A maximum lubrication (bond breaker) thickness of 5 mils or less will provide a tight fit and good consolida­tion of concrete around the dowel. Thicker bond breaking layers may result in dowel looseness and reduced joint effectiveness. Therefore care is necessary, especially when using grease.

Lack of a proper bondbreaker could generate ex­cessive stresses at the joint and lead to joint distress. The manufacturer or contractor should lubricate the en­tire length of the dowel to ensure meeting pull-out resistance specifications.

Dowels should be smooth and free from burrs. A smooth dowel free of deformations will allow the joint to move easily and provide good performance. The dowel should meet American Society of Testing Materials (ASTM) Specification A615. Consolidation of the concrete in the area of the joint and around dowels is also critical. Consolidation affects load transfer, concrete strength and durability.

Dowel bars should be corrosion-resistant. AASHTO Specification M254 is recommended for corrosion-resistant coating. The coating thickness must exceed 5 to 15 mils. Many states require thicker coating for added protection for dowel bars. Epoxy-coating pro­vides a very even outer layer and good resistance to corrosion.

Dowel Assemblies— Welded dowel assemblies are wire frames that support dowels at the proper location. They are also commonly called baskets. The manufac­turer alternates the ends at which each dowel is tack-welded to the wire frame. The other end of each dowel is free to move. A bondbreaker covers the en­tire dowel length. This method creates a controlled condition for expansion and contraction along the joint.

Dowel assemblies must be fastened to the subbase to prevent movement and overturning while paving. A common fastening device is a steel staking pin with a welded hook. Typically, staking pins have a minimum diameter of 0.3 in., and are mechanically driven into the subbase/subgrade. A minimum of six to eight stakes will secure a 12 ft. basket assembly effectively on granular or other non-stabilized materials.

An effective mechanism for securing dowel assemblies to stabilized (hard) subbases is a clip that fastens around the lower runner wires (Figure 12). The clip wraps around the bottom runner wires and is nailed to the subbase. The ability of the subbase material to hold the nail or other fastening member will dictate the number of clips needed to secure the assembly.

Placing the basket assemblies into plastic subbase material is another useful option for lean concrete or econocrete subbases. Laborers lower dowel assemblies into place in the plastic subbase concrete and check for accurate placement. They work from a rolling bridge over the fresh subbase. This procedure can be more efficient than using clips and ensures that the basket assembly will not move during paving.

Temporary spacer wires extend across the joint and stiffen and stabilize the basket. This helps resist move­ment and deflection while paving. Many contractors and specifiers feel the spacer wires can be left intact without adversely affecting joint performance and may actually improve dowel alignment. Other specifications require cutting spacer wires on a basket assembly. The decision to cut spacer wires should be based on past performance and local experience.

Sawing — Sawing is the most reliable method of creating transverse contraction joints. The initial saw cut provides a plane of weakness where cracking will begin. A secondary sawing operation may be needed to provide the proper shape factor for good sealant performance.

The contractor should mark joint locations during pav­ing operations. Where dowels are specified, a laborer should mark the center of the basket, or insertion area. This ensures that the saw cut will be centered over the dowels.

Determining the proper time to begin sawing the pave­ment after placement is critical and requires judgment; sawing too late could lead to uncontrolled cracking in some cases. The quality of the saw cut will vary with concrete strength. Sawing too soon, results in some spalling and raveling along the joint face. Weather con­ditions (temperature, wind, humidity and direct sunlight) have a large influence on concrete strength gain and the optimal time to begin sawing. The concrete mix design and the subbase or subgrade also affect pro­per timing. Mixes with softer limestone aggregates re­quire less strength development than do mixes with harder coarse aggregates.

Sawing should begin as soon as possible after ade­quate strength is obtained. Under most normal condi­tions, contractors begin sawing between 4 to 12 hours after placement, depending on curing conditions and subbase type. Extreme temperature conditions require even greater attention to detail. Hot weather can lead to increased rate of shrinkage requiring sawing to begin sooner than four hours. Colder temperatures can retard concrete strength gain, prohibiting sawing operations for up to 24 hours or more after paving.

The initial saw cut for a weakened plane in hardened concrete should be at least one-third the thickness of the slab (D/3) with a minimum width of 1/8 in. Under most circumstances, each joint should be sawed as soon as possible after paving. On natural subgrade or granular subbases, initial joints may sometimes be sawed at 60- to 80-foot intervals initially, with the in­termediate joints sawed at a later time. Stiff stabilized subbases require that all transverse contraction joints be sawed consecutively to prevent uncontrolled cracking.

Timing of sawing transverse contraction joints is critical.

A widening cut is used to establish the proper shape factor for the specified sealant material. Contractors use saws equipped with multiple blades and spacer combinations to get the necessary widths. Widening cuts are generally made within 7 days after paving and initial sawing.

Joint reservoir dimensions vary in width from 1/4 to 1/2 in. wide and from 3/4 to 1-3/4 in. deep, depending on joint spacing and sealant material. It is advisable to keep the initial joint reservoir width to less than 3/8 in. This should be adequate for most sealants and allows for future joint resealing.

The selection of saw blade type (abrasive/dry or dia­mond/wet) depends on the aggregates in the concrete. Aggregates vary from soft (easy to cut) to hard (difficult to cut). Table 3 shows some general aggregate hard­ness classifications. Aggregate size can also influence the ease of sawing regardless of overall aggregate hardness.

Table 3.

General Aggregate Hardness Table
Soft Medium Hard
Limestone River Gravel Granite
Dolomite Trap Rock Flint
Coral Chert
Quartz


Diamond saw blades can cut all concrete materials. Abrasive blades are sufficient to economically cut soft aggregates, but are usually not used on harder materials. Regardless of blade type, operators should monitor blade wear by occasionally measuring blade diameter. Measuring the saw cut depth is the most reliable way to ensure the cut meets requirements.

Daily paving production requirements determine the size (horsepower) and quantity of saws needed for a project. Large projects with many joints require high-production equipment. Self-propelled saws from 35-65 horse power or multi-blade span saws are recom­mended to keep up with paving production. It is im­portant that the contractor have backup saws available in case of mechanical breakdowns.

Cleaning — The reservoir faces require a thorough cleaning to be sure of good sealant adhesion and long-term performance. No dust or visible traces of dirt should remain on the joint faces during sealant installa­tion. The following outlines recommended cleaning procedures (with compression sealants, steps b and c are not required):

  1. Immediately after sawing, a water wash (less than 100 psi pressure) should be used to remove the slurry from the sawing operation.
  2. After the joint has sufficiently dried, the joint should be sandblasted to remove any remaining residue. One pass along each reservoir face pro­vides excellent results. This effectively cleans the joint faces, and also enhances sealant adhesion by providing texture to the reservoir faces.
  3. Just prior to sealing, the joint should be air blown to remove sand, as well as any dirt and dust deposited by wind and traffic. Air pressure should be greater than 90 psi. The contractor should be sure that the air compressor is equipped with a filter to remove moisture and oil from the air.

Sealing Operations — After cleaning, the first step is to install the backer rod. The contractor uses a rolling device which inserts the backer rod to the desired depth. The backer rod should not be stretched so it can maintain contact to the reservoir walls.

Liquid sealants require uniform installation. The reser­voir should be filled from the bottom upward to avoid trapping air bubbles. Over-filling or completely filling the reservoir is not desirable. Good practice is to recess the sealant at least one-eighth to one-quarter in. below the surface of the pavement. This allows room for sealant expansion during summer joint closure. If not recessed, the sealant can extrude onto the pave­ment surface where traffic may dislodge it from the reservoir [59].

Low-modulus silicone sealants which are not self-leveling require tooling to provide desired results. After applying the sealant, a laborer shapes the sealant by drawing a tool over the surface of the silicone sealant. This forces the sealant into contact with the sidewall at the top of the joint and produces the proper shape factor [60].

The sealing operation for preformed compression seals requires application of a lubricant/adhesive to the reservoir sidewalls. The compression seal is then in­serted into the reservoir (no backer rod is used). The lubricant/adhesive material eases sealant insertion, and forms a weak adhesive that helps hold the seal in place. Care is required during installation to avoid twisting or stretching the sealant. More than five per­cent stretch is excessive and should be avoided. Most compression seal manufacturers provide installation machines which eliminate stretch [61].

Transverse Construction Joints

Design

Transverse construction joints are installed at the end of a day-long paving operation or other placement interruptions. Common interruptions are for bridges, in­tersections and emergency shutdowns. If possible, the contractor should place a transverse construction joint at the same location as would be used for a transverse contraction joint. If this is not possible, in­stallation should be within the middle third of a plan­ned panel. Transverse construction joints are always oriented perpendicular to the centerline, even where contraction joints are skewed.

Transverse Expansion Joint

Transverse construction joints are butt joints and do not benefit from aggregate interlock. Where the con­struction joint is placed in a planned location or paving is not adjacent to an existing concrete slab, dowels are required to provide load transfer. The same dowel siz­ing and placing recommendations described for doweled contraction joints apply to transverse con­struction joints.

Construction

Paving contractors often refer to transverse construc­tion joints as headers. To a contractor, a header is the location at which paving will resume on the next day.

If the transverse construction joint must be placed within the middle third of a planned panel and paving is adjacent to an existing slab, the construction joint should be tied. Tiebars will prevent movement and vir­tually eliminate the possibility of developing a sympathy crack in the adjacent slab. To provide load transfer, tiebar diameters should be the same as required for dowel bars.

The most common method of header construction is to run out the paving operation ending at a header board. The installation of a header board requires handwork, which can lead to a rough surface. Dowels are placed through the header board in pre-drilled locations. Additional consolidation with hand held vibrators should assure satisfactory encasement of the dowels. Before resuming paving, the header board is removed.

Although transverse construction joints (headers) are generally formed against a header board, they can also be sawed. To construct a sawed header, the slip­form paver operator runs the machine past the desired construction joint location. Generally the last two con­crete batches approaching a sawed header are altered for high-early strength gain. No forms are used. Ex­cess material is removed after sawing at the desired location. The contractor drills holes and grouts dowel bars into the sawed header face. Sawed headers are advantageous because they provide very smooth tran­sitions between paving sections.

Formed transverse construction joint (header).

Sawed transverse construction joint. (Note: sawing location is clearly indi­cated and corresponds to the planned location of a transverse contraction joint.) Transverse construction joints do not require initial saw­ing. Reservoir dimensions and cleaning and sealant in­stallation operations are the same as those used on contraction joints.


Transverse Expansion Joints

Highway pavements do not normally require transverse expansion joints. Surveys of in-service and experimen­tal highway pavements show that expansion joints are only necessary at fixed structures [62][63]. In the past designers placed transverse expansion joints to relieve compressive forces in the pavement and limit blow­ups. However, in many cases the expansion joints allowed too much opening of adjacent transverse con­traction joints which led to loss of aggregate interlock and sealant damage [64]. By eliminating unnecessary expansion joints, adjacent contraction joints will remain tight and provide good load transfer and joint effectiveness.

Transverse Construction Joint

Design

The purpose of an expansion joint determines whether or not load transfer is required. Expansion joints used to isolate an on-line structure, such as a bridge, should have dowels to increase load transfer and joint effec­tiveness. However at unsymmetrical intersections and ramps, load transfer dowels should be omitted so dif­ferential horizontal movements can occur without damaging the abutting pavement.

Expansion or isolation joints should be 3/4 to 1 in. wide. Excessive movement (slab migration) may occur with greater widths. In expansion joints, a preformed joint filler material occupies the gap between the sub­base or subgrade and the joint sealant. The filler is recessed about one inch below the surface and must extend the full depth and width of the slab. Expansion joint filler material should allow 50% compression and be nonshrinking, nonabsorbant, nonreactive, nonex­truding and flexible.

The joint sealant is installed on top of the preformed filler. The sealant inhibits the infiltration of incom­pressibles and keeps the filler in place. It is essential to recess the sealant about 1/4 in. to protect it from the damaging effects of traffic. The sealant and preformed filler material should be compatible. Some sealant manufacturers recommend a tape bond breaker bet­ween incompatible sealant and filler materials. Regular maintenance inspections will be necessary in order to evaluate the performance of the expansion joint sealing materials [65].

Contraction joints within 60 to 100 ft. of expansion joints should be doweled [66]. The expansion joint may allow adjacent contraction joints to open more than other contraction joints. If not doweled, adjacent contraction joints would lose load transfer [67]. Pressure relief joints serve the same purpose as ex­pansion or isolation joints. However, installation occurs years after original pavement construction. Pressure relief joints relieve pressure against structures and alleviate potential blowups. Although for most pavements they are not needed, pressure relief joints may be appropriate to relieve imminent structure damage from excessive compressive stresses [68].

Construction

Doweled Expansion Joints — In transverse expan­sion joints one end of each dowel is equipped with an expansion cap. The expansion cap allows the dowel to move freely as the joint expands and contracts. The cap must be long enough to cover at least 2 in. of the dowel and should provide a watertight fit. The cap should be equipped with a stop which prevents the cap from slipping off of the dowel during placement. A good stop location will provide a minimum dowel coverage equal to 1/4 in. more than the expansion joint width (typically 1-1/4 in.). The capped end of the dowel is also lubricated to prevent bond.

The same dowel placement and alignment re­quirements used for doweled contraction joints apply to doweled expansion joints. The dowels are typically placed at mid-depth, spaced 12 in. apart (center to center), and have a diameter of 1-1/4 in. for 8 to 9-1/2 in. slabs, and 1-1/2 in. diameter dowels for thicknesses of 10 in. or greater. Epoxy coating for corrosion resistance is recommended for harsh climates.

An expansion basket supports and aligns the dowel bars while also supporting the preformed filler material. The filler must extend the entire width of the slab and fit snugly into the basket frame.

Typical doweled expansion joint. Note dowel caps and filler material.

Undoweled Expansion Joints (Isolation Joints) — For undoweled expansion joints, the joint faces are thickened to reduce load stresses developed along the slab bottom. The abutting slabs should be thickened by 20% along the expansion joint. The thickness transition is tapered at a slope of 6 to 10 times the pavement thickness. For example, a thickness transition from 10 in. to a 12 in. thickened edge, would occur over 60 to 100 in. (6 to 8 ft).

Longitudinal Contraction Joints

Longitudinal Contraction Joint

Design

Longitudinal contraction joints divide lanes of traffic and control cracking where two or more lane widths are placed at one time. Longitudinal joints are necessary when slab widths exceed 15 ft.

Longitudinal joints should be cut one-third the depth of the slab. An initial saw cut 1/8 to 3/8 in. wide will ac­commodate most sealant materials. The reservoir shape factor is not critical due to small movements at these joints. Typical reservoir dimensions range from 1/4 to 3/8 in. wide by 1-1/4 in. deep.

Load transfer at longitudinal joints is usually achieved through aggregate interlock. To maintain the interlock, steel tiebars are often used to hold the longitudinal joints tight. Tiebar spacing varies with the thickness of the pavement and the distance of the joint to the nearest free edge. Table 4 provides recommended tiebar spacings [69]. Recommended tiebar lengths are based on the allowable working strength of the steel. However, concrete pavement movements rarely develop stresses which approach the strength of the steel tiebars. Bar sizes and embedment lengths should reflect the actual forces acting in the pavement, not only the working strength of the steel. Specifiers should choose standard manufactured tiebar lengths.

Tiebars should not be placed within 15 in. of transverse joints or they can interfere with joint move­ment. If tiebars longer than 32 in. are being used with skewed joints, they should not be closer than 18 in. from a transverse joint.

Tiebars should be protected from corrosion. AASHTO Specification M284 and ASTM Specification D3963 provide guidelines for protection. The coating should not exceed 5-12 mils thickness. Thicker coating could result in a reduction in the gripping ability of the deformed bars.

Highway loop and slip ramps which are 16 ft. wide may require special attention. Some agencies do not specify an interior longitudinal joint in the ramp width. However, many states have experienced uncontrolled longitudinal cracking in their ramp pavements. These agencies place a longitudinal joint at the lane center or quarter point. Local performance experience and ease of construction should be considered.

Construction

Longitudinal contraction joints are sawed similarly to transverse contraction joints. With slipform equipment, tiebars may placed manually or mechanically by an in­sertion device. The bars are inserted mid-depth at a specified spacing.

Timing of sawing operations for longitudinal joints is not as critical as for transverse contraction joints. On stiff stabilized subbases, longitudinal joint sawing should start as soon as possible, with the longitudinal sawing proceeding after transverse joints are sawed in­itially. During periods with large temperature fluctua­tions (spring, fall, or sudden thunderstorms), sawing should start as early as possible.

Typically, a joint reservoir is not required for a longitudinal contraction joint. A single saw cut of 1/8 to 3/8 in. is sufficient. The initial saw cut depth in hard­ened concrete should be a minimum of one-third the thickness of the slab (D/3).

Sawing longitudinal contraction joints.

Where a sealant reservoir is desired, an initial saw cut is made with a special multi-blade centerline saw, or with two saws to give the desired step-cut. The width and depth of the saw cut should correspond to the desired reservoir dimensions.

Plastic inserts are not recommended for longitudinal joint construction [70]. Although several states have had success with plastic inserts, inconsistent perfor­mance has been widespread [71].

It is considered good practice to optimize joint sealant cost. Many states specify different sealant materials in longitudinal contraction joints than in transverse joints because longitudinal joint movement is insignificant.


Longitudinal Construction Joints

Longitudinal Construction Joint

Design

Longitudinal construction joints join lanes which are paved in separate passes. This includes concrete shoulders and traffic lanes. Keyways and/or tiebars can provide load transfer.

Keyway designs are either trapezoidal or half-round. Some agencies eliminate keyways when a pavement is Figure 13 shows standard dimensions for both basic ­designs. The keyway should be located at mid-depth of the slab to provide maximum strength.

Some agencies eliminate keyways when a pavement is less than 10 in. thick [72]. This is based on field experience where keys failed in shear and resulted in spalling along the joint. However, these failures generally occur when the keyways are built too large or are above mid-depth of the slab. Good design and construction practice should eliminate these problems. The specifier should use local experience in making the decision to use keyways in slabs less than 10 in.

To maintain load transfer, tiebars are always necessary when using keyways. Tiebars hold the male and female portions of the key together. Tiebar requirements for keyed longitudinal construction joints are similar to those for longitudinal contraction joints Figure 13 Standard dimensions for basic keyway designs. (see Table 4).

Maximum Recommended Tiebar Spacings (in.)
Bar Size #4 Bar #5 Bar
Grade Steel Grade 40** Grade 60** Grade 40** Grade 60**
Length of Bar* 24 in. 32 in. 30 in. 40 in.
Dist. To Free Edge (ft.) 10 12 16 22 24 10 12 16 22 24 10 12 16 22 24 10 12 16 22 24
Thickness Type of Joint
9" Warp 37 31 23 17 16 48 47 35 25 23 48 48 36 26 24 48 48 48 40 36
Butt 26 22 16 12 11 40 34 25 18 16 42 35 26 19 17 48 48 39 29 26
10" Warp 34 28 22 16 14 48 42 32 23 20 48 44 33 24 22 48 48 48 36 32
Butt 24 20 16 11 10 36 30 23 16 14 38 31 24 17 16 48 47 35 26 23
11" Warp 31 25 20 15 13 47 38 29 21 19 48 40 30 22 20 48 40 30 22 20
Butt 22 18 14 11 9 34 27 21 15 14 34 29 21 16 14 48 43 31 23 21
12" Warp 28 23 18 13 12 42 35 27 19 18 44 36 28 20 13 48 48 41 30 28
Butt 20 16 13 9 9 30 25 19 14 13 31 26 20 14 13 47 39 29 21 20
Warp joint: A sawed or construction joint with a keyway.
Butt joint: a construction joint with no keyway.
* Length to develop full yield strength steel. Agencies should specify standard manufactured lengths (24, 30,36, 42, 48 in.).
** Steel grades based on ASTM A615.


Tiebars must provide all of the necessary load transfer where a keyway is not used. Because traffic is not constantly crossing these smooth longitudinal joints, dowels are not necessary to provide structural edge support. Small diameter (#4 or #5) tiebars are ade­quate, but the spacing between bars must be reduced to 12 to 24 in. to effectively transfer load and reduce slab stresses and deflections [73].

Construction

There are three basic methods for tiebar placement for longitudinal construction joints. A common method is to use ninety-degree bent tiebars. The bent tiebars are in­serted into the side of the slab while paving (Figure 14). The bars are straightened before paving adjacent lanes. Another method is to drill holes into the longitudinal joint face. Tiebars are inserted into the holes and secured with an epoxy grout. Some states use a two-part threaded tiebar and splice coupler system. The female couplers are cast in the longitudinal joint. Before paving an adjacent lane the threaded bars are screwed into the in-place couplers.

Regardless of the method of tiebar placement, it is im­portant that the tiebars are securely anchored. Tiebars need to provide adequate pull-out resistance to function well. Table 5 provides pull-out resistance criteria [74].

Pull-Out Resistance Criteria
Tied Width of Pavement (Distance from Joint To Nearest Free Edge) Average Pull-out Resistance of Tiebars, lbs./L.F. of joint, min.
12 ft. or less 2200
Over 12 ft. to 17 ft. 3200
Over 17 ft. to 24 ft. 4500
Over 24 ft. to 28 ft. 5200
Over 28 ft. to 36 ft. 6800
Over 36 ft. 9000


Keyed longitudinal construction joint. Keyway adds constructability by keep­ing tiebars out of the way until straightened just prior to paving.

Ninety-degree bent tiebars should be manufactured with ASTM Grade 40 or a proprietary steel specifica­tion. Grade 40 steel tolerates straightening better than Grade 60 steel. Any pull-out tests conducted for this installation method should be performed on bars which have been bent and straightened to simulate actual field installation.

Longitudinal construction joints require a wider reser­voir (up to 1/2 in.) to accommodate variations in edge location due to the paving process. Sawing and seal­ing operations are similar to those used for longitudinal contraction joints.

Sealants

The role of the joint sealant is to minimize surface water infiltration into the pavement structure and pre­vent incompressibles from entering the joint. Incom­pressibles cause point bearing pressures which can lead to spalling and "blow ups" in extreme cases.

The designer should first select a sealant material that meets performance and cost criteria, then size a reser­voir that will permit the sealant to function properly. Joint movement and environmental conditions are fac­tors that influence reservoir dimensions. Proper reser­voir design helps ensure good joint sealant performance.

Table 6.

Common Joint Sealant Materials
Sealant Type Specification Properties
Hot Pour Sealants
Polymeric Asphalt Based AASHTO MO173 Self Leveling
ASTM D3405 Self Leveling
SS-S-1401 C Self Leveling
ASTM D1190 Self Leveling
Polymeric Sealant ASTM D3405 Self Leveling
Low Modulus Modified Self-leveling
Elastomeric Sealant SS-S-1614 Self Leveling
Coal Tar, PVC ASTM D3406 Self Leveling
Cold Pour Sealants / Single Components
Silicone Sealant N.A. Non sag, toolable, low modulus
Silicone Sealant N.A. Self Leveling (no tooling), low modulus
Silicone Sealant N.A. Self-leveling (no tooling), ultra-low modulus
Nitrile Rubber Sealant N.A. Self Leveling (no tooling), non sag
Polysulfide Sealant N.A. Self Leveling(not tooling), Low Modulus
Preformed Polychloroprene Elostomoeric (Compression Seals)
Preformed Compression Seals ASTM D2628-81 20-50% allowable Strain
Lubricant Adhesive ASTM D2835
Preformed Expansion Joint Filler Material
Preformed Filler Material ASTM D1751
AASHTO M213
Bituminous, Nonextruding, resilient
Preformed Filler Material ASTM D1752
AASHTO M153
Sponge Rubber, Cork
Preformed Filler Material ASTM D994
AASHTO M33
Bituminous


The expected joint movement should determine the selection of sealant material. Little movement is ex­pected at transverse contraction joints with short spac­ing (less than 20 ft.), longitudinal joints, or shoulder joints. Tied joints have virtually no movement. However, transverse contraction joints in reinforced panels will move significantly.

Materials

There are many acceptable materials available for seal­ing joints in concrete pavements. Sealants are typically classified as liquid or preformed. Liquid sealants may be hot or cold poured, single or two component, and self-leveling or toolable. All liquid sealants depend on long-term adhesion to the joint face for successful seal­ing. Preformed sealants (compression sealants) depend on long-term compression recovery for successful seal­ing. Table 6 gives descriptions and specifications of most available sealants.

While many agencies specify single-component cold-pour sealants, there are no standard national specifica­tions for these materials. An agency must use the manufacturer's recommendations or develop their own specification.

Sealant properties necessary for long-term performance depend on the specific application and the climatic en­vironment of the installation. Properties to consider include:

  • Elasticity: The ability of a sealant to return to its original size when stretched or compressed.
  • Low Modulus: The change in internal stresses in a sealant while being stretched and compressed over a range of temperatures. A low modulus is desirable and is particularly important in cold weather climates.
  • Adhesion: Ability of a sealant to adhere to con­crete. Initial adhesion and long-term adhesion are equally important. (Not applicable to compression seals.)
  • Cohesion: Ability of a sealant to resist tearing from tensile stresses. (Not applicable to compres­sion seals.)
  • Compatibility: Relative reaction of the sealant to materials which it contacts (such as backer rods and other sealants).
  • Weatherability: Ability of a sealant to resist deterioration when exposed to the elements (primarily ultra violet sun rays and ozone).


Preformed Compression Seals

The reservoir design and compression seal selection should ensure that the seal remains compressed at a level between 20 and 50% at all times [75]. The reser­voir depth must exceed the depth of the compressed seal, but is not related directly to the width of the reservoir. In general, the width of the selected com­pression seal can be about twice the width of the joint reservoir. If the sealant is undersized, joint opening may be too wide and compression will be lost.

A successful compression seal installation depends solely on the compressive recovery of the selected compression seal. Unlike liquid sealants which ex­perience both compression and tension, preformed compressive seals are designed to be in tension throughout their life. While the lubricant/adhesive used during installation has some adhesive qualities, its primary function is to provide lubrication during installa­tion. No consideration of adhesive qualities should be included.

The best long-term performance of compression seals has been with those having at least five cells. Figure 15 shows a cross-section of a typical five cell seal.


Backer Rods

Backer rods are an important component for liquid sealant installation. Backer rods prevent the sealant from flowing out of the bottom of the joint and prevent sealant adhesion to the bottom of the joint reservoir. The backer rod also helps define the shape factor and optimize the quantity of sealant used.

Backer rods are installed in the joint reservoirs before liquid sealants are installed. The contractor uses a tool that depresses the backer rod to the depth needed to form the desired shape factor. The backer rod diameter should be 25% greater than the reservoir width to ensure a tight fit.

Installing backer rod in transverse contraction joint.

There are no national specifications for backer rods; however, important considerations for various materials include:

  • Polyethylene Foam: Polyethylene foam is a closed-cell foam that does not absorb water and is moderately compressible. Since polyethylene foam may melt with hot-pour materials it is better suited for cold pour sealants.
  • Crosslinked Polyethylene Foam: Crosslinked polyethylene foam is compatible with hot-pour sealants. It is a closed-cell foam that does not ab­sorb water and is moderately compressible, but will not melt in contact with hot-pour sealant.
  • Polyurethane Foam: This open cell foam ab­sorbs water, but does not melt when used with hot-pour materials. It is very compressible, and commonly used with hot-pour sealants.


Joint Sealant Reservoirs

The shape factor is critical to long-term success of a sealant installation. As a cross section of the joint sealant changes during the expansion and contraction of the concrete pavement, stresses develop within the sealant and along the sealant/reservoir bond line. These stresses can be excessive if the shape factor is not appropriate for the sealant material. Figure 16 shows typical shape factors for liquid and compression sealants. A joint sealant reservoir with a shape factor of one or less develops lower stresses on the joint sealant than a shape factor greater than one. The lower or reduced internal stresses resulting from pro­per shape factors minimize adhesive or cohesive failures. Shape factor design should include recessing the sealant from 1/4 - 3/8 in. Recessing is important to avoid extrusion problems.

Fast Track Considerations

There are no limitations on joint sawing equipment for Fast Track concrete pavements. Both wet-sawing, with diamond blades, and dry-sawing, with silicon carbide or carborundum blades, have been used. Joints on Fast Track projects created with either saw blade type have performed well.

The accelerated strength gain and low water-cement ratio of Fast Track concrete reduce excess moisture on the sidewalls of the joint reservoirs. This will help allow sealing earlier than with standard mix designs and pavement construction procedures. Experience has shown that low-modulus rubber sealants adhere to reservoir faces as early as eight hours after paving.

Silicone sealants have also been used for Fast Track operations. These sealants have provided good perfor­mance and have not shown adhesion problems.

Preformed neoprene compression seals may be ideal. These seals are not highly sensitive to dirt or moisture and may allow sealing at an earlier stage.

To gain the full potential of the Fast Track concept, joint sealing should follow paving as soon as possible. Sealing joints quickly with hot pour sealants has pro­vided good performance. However, installation pro­cedures should follow the sealant manufacturer's recommendations for Fast Track concrete paving.

Additional Information

Additional information on the design and construction of joints for concrete highways and other applications is available. Contact the American Concrete Pavement Association or the Portland Cement Association.

Summary

The objective of joint system design for concrete highway pavements is to maintain pavement structural capacity and ride quality at low annual costs. Joints control cracking and provide a sealant reservoir that will minimize water and incompressible infiltration into the pavement to prevent spalling, pumping and faulting.

Proper joint design and construction are critical to achieving good performance of concrete highway pavements. Important considerations in joint design and construction include:

  • Transverse contraction joints are the most com­mon and critical to good performance. The three most important design factors for transverse con­traction joints are spacing, load transfer, and joint sealing requirements.
  • Proper construction of transverse contraction joints is also critical. Dowel placement and preparation is important. Excellent placement tolerances have been achieved with dowel assemblies and insertion equipment.
  • Sawing must occur in a timely manner to avoid random cracking. Under most normal conditions, this is usually within 4 to 12 hours of paving, depending on curing conditions and subbase type. Extreme weather conditions require even greater attention to detail and will significantly af­fect the start of sawing operations.
  • Proper joint sealing contributes to good perfor­mance. Hot poured, cold poured, or compression seals are acceptable for highway pavements. The joint spacing influences the choice of sealant materials and reservoir design.
  • Proper placement and consolidation of concrete are essential at transverse construction joints. A "sawed header" can achieve a very smooth transition.
  • Only fixed objects such as bridges, tunnels, re­taining walls, abutments, etc. require isolation (ex­pansion) joints.
  • Where keyways are included in the joint design, tiebars are recommended to hold the key together. Keyway height should not exceed one-fifth slab thickness. For trapezoidal keyways, the width of the keyway should not exceed one-tenth slab thickness.
  • Current sealing materials and methods allow pro­per joint sealing in Fast Track concrete paving operations.



Reference

  1. loannides, A.M. and Korovesis, G.T. "Aggregate Interlock: A Pure-Shear Load Transfer Mechanism" Paper No. 89-0189, presented to the 1990 Annual Meeting of the Transportation Research Board, Washington, D.C., March 1990.
  2. Colley, B.E., and Humphrey, H.A., "Aggregate Interlock at Joints in Concrete Pavements," Highway Research Record No. 189, Highway Research Board, National Research Council, Washington D.C., 1967.
  3. Nowlen, W.J., "Influence of Aggregate Properties on the Effectiveness of Interlock Joints in Concrete pavements," Journal of the Portland Cement Association, Research and Development Laboratories, Vol. 10, No. 2, May 1968, pp. 2-8.
  4. Darter, M.I., "Design of Zero-Maintenance Plain Jointed Pavements," FHWA-RD-77-111, Federal Highway Administration, Washington D.C., June 1977.
  5. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. I, Evaluation of Concrete Pavement Performance and Design Features, Federal Highway Administration, FHWA-RD-89-136, March 1990.
  6. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. IV Appendix A, Project Summary Reports and Summary Tables, Federal Highway Administration, FHWA-RD-89-139, March 1990.
  7. Kelleher, K. and Larson, R.M., "The Design of Plain Doweled Jointed Concrete Pavement," Proceedings from the 4th International Conference on Concrete Pavement Design and Rehabilitation, Purdue University, West Lafayette, April, 1989.
  8. "NCHRP Synthesis of Highway Practice #19: Design, Construction and Maintenance of PCC Pavement Joints,", Transportation Research Board, National Research Council, Washington D.C., 1973.
  9. Tabatabaie, A.M., and Barenberg, E.J. "Structural Analysis of Concrete Pavements Systems,"Transportation Engineering Journal, ASCE, Vol 106, No. TE5, Proceedings Paper 15671, pp 493-506.
  10. Tabatabaie, A.M., Barenberg, E.J., and Smith, R.E. "Longitudinal Joint Systems in Slip-Formed Rigid Pavements: Vol II. - Analysis of Load Transfer Systems for Concrete Pavements," Federal Highway Administration, FHWA/RD-86/042, Washington D.C., November, 1979.
  11. loannides, A.M. and Korovesis, G.T. "Analysis and Design of Doweled Slab-On-Grade Pavement Systems," Submitted for publication in the Journal of Transportation Engineering, ASCE, March 1990.
  12. Kilareski, W.P., Ozxbeki, M.A., Anderson, D.A., "Fourth Cycle of Pavement Research at the Pennsylvania Transportation Research Facility - Vol. 4: Rigid Pavement Joint Evaluation and Full-Depth Patch Designs," Report No. FHWA/PA-84-026, Pennsylvania State University, University Park, PA, December 1984.
  13. Kilareski, W.P., Ozbeki, M.A., Anderson, D.A., "Computer Simulation and Field Evaluation of Transverse Joints in Rigid Pavements," Proceedings from the 3rd International Conference on Concrete Pavement Design and Rehabilitation, Purdue University, West Lafayette, April, 1985.
  14. Kilareski, W.P., Ozxbeki, M.A., Anderson, D.A., "Fourth Cycle of Pavement Research at the Pennsylvania Transportation Research Facility - Vol. 4: Rigid Pavement Joint Evaluation and Full-Depth Patch Designs," Report No. FHWA/PA-84-026, Pennsylvania State University, University Park, PA, December 1984.
  15. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. I, Evaluation of Concrete Pavement Performance and Design Features, Federal Highway Administration, FHWA-RD-89-136, March 1990.
  16. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. IV Appendix A, Project Summary Reports and Summary Tables, Federal Highway Administration, FHWA-RD-89-139, March 1990.
  17. Teller, L.W. and Cashel, H.D. "Performance of Doweled Joints under Repetitive Loadings," Bulletin 217, Highway Research Board, Washington, D.C., 1968.
  18. Kelleher, K., Larson, R.M., "The Design of Plain Doweled Jointed Concrete Pavement," Proceedings from the 4th International Conference on Concrete Pavement Design and Rehabilitation, Purdue University, West Lafayette, IN, April 1989.
  19. Anderson, D.A., Kilareski, W.P., Luhr, D.R.A. "Fourth Cycle of Pavement Research at the Pennsylvania Transportation Research Facility - Vol. 7," Fourth Cycle of Pavement Research Summary Report, FHWA/PAS-84-029, Pennsylvania Department of Transportation, Harrisburg, PA, December 1984.
  20. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. I, Evaluation of Concrete Pavement Performance and Design Features, Federal Highway Administration, FHWA-RD-89-136, March 1990.
  21. Henrichs, K.W, Liu, M.J., Darter, M.I., Carpenter S.H., and loannides, A.M. "Rigid Pavement Analysis and Design," FHWA-RD-88-068, Federal Highway Administration, Washington D.C., June 1989.
  22. Colley, B.E., and Humphrey, H.A., "Aggregate Interlock at Joints in Concrete Pavements," Highway Research Record No. 189, Highway Research Board, National Research Council, Washington D.C., 1967.
  23. Colley, B.E., and Humphrey, H.A., "Aggregate Interlock at Joints in Concrete Pavements," Highway Research Record No. 189, Highway Research Board, National Research Council, Washington D.C., 1967.
  24. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. IV Appendix A, Project Summary Reports and Summary Tables, Federal Highway Administration, FHWA-RD-89-139, March 1990.
  25. Colley, B.E., and Humphrey, H.A., "Aggregate Interlock at Joints in Concrete Pavements," Highway Research Record No. 189, Highway Research Board, National Research Council, Washington D.C., 1967.
  26. Colley, B.E., and Humphrey, H.A., "Aggregate Interlock at Joints in Concrete Pavements," Highway Research Record No. 189, Highway Research Board, National Research Council, Washington D.C., 1967.
  27. Colley, B.E., and Humphrey, H.A., "Aggregate Interlock at Joints in Concrete Pavements," Highway Research Record No. 189, Highway Research Board, National Research Council, Washington D.C., 1967.
  28. Kelleher, K. and Larson, R.M., "The Design of Plain Doweled Jointed Concrete Pavement," Proceedings from the 4th International Conference on Concrete Pavement Design and Rehabilitation, Purdue University, West Lafayette, April, 1989.
  29. Packard, R.G., "Design Considerations for Control of Joint Faulting of Undoweled Pavements," Proceedings from the First International Conference on Concrete Pavement Design, Purdue University, West Lafayette, IN, February 1977.
  30. Hudson, et al. Aggregate and Paved Surface Design and Rehabilitation Manual for Low-Volume Roads, Federal Highway Administration.
  31. Darter, M.I., Becker, J.M., Snyder, M.B., and Smith, R.E. "Portland Cement Concrete Pavement Evaluation (COPES)," National Cooperative Highway Research Program Report 277, Transportation Research Board, Washington D.C., 1985.
  32. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. IV Appendix A, Project Summary Reports and Summary Tables, Federal Highway Administration, FHWA-RD-89-139, March 1990.
  33. Construction and Maintenance of Rigid Pavements General Report, XVIII Brussels World Road Congress, PIARC, September 1987.
  34. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. IV Appendix A, Project Summary Reports and Summary Tables, Federal Highway Administration, FHWA-RD-89-139, March 1990.
  35. Construction and Maintenance of Rigid Pavements General Report, XVIII Brussels World Road Congress, PIARC, September 1987.
  36. Smith, L.L. "Load Transfer at Contraction Joints in Plain Portland Cement Concrete Pavements," Florida Research Bulletin No. 90, 1965.
  37. Gulden, W. "Pavement Faulting Study Investigation Into the Causes of Pavement Faulting on the Georgia Interstate System," Interim Report 2, Project 7104, Georgia Department of Transportation, Atlanta, GA, 1974.
  38. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. I, Evaluation of Concrete Pavement Performance and Design Features, Federal Highway Administration, FHWA-RD-89-136, March 1990.
  39. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. IV Appendix A, Project Summary Reports and Summary Tables, Federal Highway Administration, FHWA-RD-89-139, March 1990.
  40. Tabatabaie, A.M., and Barenberg, E.J. "Structural Analysis of Concrete Pavements Systems,"Transportation Engineering Journal, ASCE, Vol 106, No. TE5, Proceedings Paper 15671, pp 493-506.
  41. Tabatabaie, A.M., Barenberg, E.J., and Smith, R.E. "Longitudinal Joint Systems in Slip-Formed Rigid Pavements: Vol II. - Analysis of Load Transfer Systems for Concrete Pavements," Federal Highway Administration, FHWA/RD-86/042, Washington D.C., November, 1979.
  42. Kilareski, W.P., Ozxbeki, M.A., Anderson, D.A., "Fourth Cycle of Pavement Research at the Pennsylvania Transportation Research Facility - Vol. 4: Rigid Pavement Joint Evaluation and Full-Depth Patch Designs," Report No. FHWA/PA-84-026, Pennsylvania State University, University Park, PA, December 1984.
  43. Kilareski, W.P., Ozbeki, M.A., Anderson, D.A., "Computer Simulation and Field Evaluation of Transverse Joints in Rigid Pavements," Proceedings from the 3rd International Conference on Concrete Pavement Design and Rehabilitation, Purdue University, West Lafayette, April, 1985.
  44. Construction and Maintenance of Rigid Pavements General Report, XVIII Brussels World Road Congress, PIARC, September 1987.
  45. Teller, L.W. and Cashel, H.D. "Performance of Doweled Joints under Repetitive Loadings," Bulletin 217, Highway Research Board, Washington, D.C., 1968.
  46. Kelleher, K., Larson, R.M., "The Design of Plain Doweled Jointed Concrete Pavement," Proceedings from the 4th International Conference on Concrete Pavement Design and Rehabilitation, Purdue University, West Lafayette, IN, April 1989.
  47. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. I, Evaluation of Concrete Pavement Performance and Design Features, Federal Highway Administration, FHWA-RD-89-136, March 1990.
  48. Kelleher, K., Larson, R.M., "The Design of Plain Doweled Jointed Concrete Pavement," Proceedings from the 4th International Conference on Concrete Pavement Design and Rehabilitation, Purdue University, West Lafayette, IN, April 1989.
  49. Anderson, D.A., Kilareski, W.P., Luhr, D.R.A. "Fourth Cycle of Pavement Research at the Pennsylvania Transportation Research Facility - Vol. 7," Fourth Cycle of Pavement Research Summary Report, FHWA/PAS-84-029, Pennsylvania Department of Transportation, Harrisburg, PA, December 1984.
  50. Black, K.N., Larson, R.M., Staunton, L.R. "Evaluation of Stainless-Steel Pipes for Use as Dowel Bars," Public Roads, Vol. 52 No. 2, Federal Highway Administration, Washington, D.C., September 1988.
  51. Colley, B.E., and Humphrey, H.A., "Aggregate Interlock at Joints in Concrete Pavements," Highway Research Record No. 189, Highway Research Board, National Research Council, Washington D.C., 1967.
  52. loannides, A.M. and Korovesis, G.T. "Aggregate Interlock: A Pure-Shear Load Transfer Mechanism" Paper No. 89-0189, presented to the 1990 Annual Meeting of the Transportation Research Board, Washington, D.C., March 1990.
  53. Colley, B.E., and Humphrey, H.A., "Aggregate Interlock at Joints in Concrete Pavements," Highway Research Record No. 189, Highway Research Board, National Research Council, Washington D.C., 1967.
  54. Henrichs, K.W, Liu, M.J., Darter, M.I., Carpenter S.H., and loannides, A.M. "Rigid Pavement Analysis and Design," FHWA-RD-88-068, Federal Highway Administration, Washington D.C., June 1989.
  55. Darter, M.I., "Design of Zero-Maintenance Plain Jointed Pavements," FHWA-RD-77-111, Federal Highway Administration, Washington D.C., June 1977
  56. Construction and Maintenance of Rigid Pavements General Report, XVIII Brussels World Road Congress, PIARC, September 1987.
  57. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. IV Appendix A, Project Summary Reports and Summary Tables, Federal Highway Administration, FHWA-RD-89-139, March 1990.
  58. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. IV Appendix A, Project Summary Reports and Summary Tables, Federal Highway Administration, FHWA-RD-89-139, March 1990.
  59. "Techniques for Pavement Rehabilitation," prepared by ERES Inc. for National Highway Institute, Federal Highway Administration, Third Edition, 1987.
  60. "Techniques for Pavement Rehabilitation," prepared by ERES Inc. for National Highway Institute, Federal Highway Administration, Third Edition, 1987.
  61. "Techniques for Pavement Rehabilitation," prepared by ERES Inc. for National Highway Institute, Federal Highway Administration, Third Edition, 1987.
  62. "FHWA Pavement Rehabilitation Manual," FHWA-ED-88-025, Federal Highway Administration, Washington D.C., March 1988, Chapter 9.
  63. "Joint Spacing in Concrete Pavements," Highway Research Board, Research Report 17-B, Washington D.C., 1956.
  64. "Concrete Pavement Joints," Technical Advisory 5040.30, Federal Highway Administration, Washington D.C., November, 1990.
  65. DeYoung, C., "Spacing of Undoweled Joints in Plain Concrete Pavement," Highway Research Record 112, Highway Research Board, Washington D.C., 1966.
  66. Packard, R.G. "Design of Concrete Airport Pavement," Portland Cement Association, EB050.03P, Skokie, IL., 1973, pg. 27.
  67. Packard, R.G. "Design of Concrete Airport Pavement," Portland Cement Association, EB050.03P, Skokie, IL., 1973, pg. 27.
  68. "Concrete Pavement Joints," Technical Advisory 5040.30, Federal Highway Administration, Washington D.C., November, 1990.
  69. "Concrete Pavement Joints," Technical Advisory 5040.30, Federal Highway Administration, Washington D.C., November, 1990.
  70. "Concrete Pavement Joints," Technical Advisory 5040.30, Federal Highway Administration, Washington D.C., November, 1990.
  71. Smith, K.D., Peshkin, D.G., Darter, M.I., Mueller, A.L., Carpenter, S.H. Performance of Jointed Pavements, Vol. IV Appendix A, Project Summary Reports and Summary Tables, Federal Highway Administration, FHWA-RD-89-139, March 1990.
  72. "Concrete Pavement Joints," Technical Advisory 5040.30, Federal Highway Administration, Washington D.C., November, 1990.
  73. Sawan, J.S., Darter, M.I., and Dempsey, B.J., "Structural Analysis and Design of PCC Shoulders," Report No. FHWA/RD-81/122, Federal Highway Administration, 1982.
  74. "Concrete Pavement Joints," Technical Advisory 5040.30, Federal Highway Administration, Washington D.C., November, 1990.
  75. "Concrete Pavement Joints," Technical Advisory 5040.30, Federal Highway Administration, Washington D.C., November, 1990.