Skip to contentUnited States Department of Transportation - Federal Highway Administration FHWA Home
Research Home   |   Pavements Home
Report
This report is an archived publication and may contain dated technical, contact, and link information
Publication Number: FHWA-04-122
Date: February 2005

Computer-Based Guidelines for Concrete Pavements Volume II

Previous | Table of Contents | Next

CHAPTER 7. HIPERPAV II USER'S MANUAL

In HIPERPAV II (software version 3.0), several changes were made to the software from the previous release of HIPERPAV I (software version 2.X). This user's manual outlines those changes and discusses the new input needed. For information on the theory behind HIPERPAV II, refer to the guidelines (chapters 2 through 5).

In HIPERPAV II, the user can now perform early-age behavior analyses for jointed concrete pavement, evaluate the impacts of early-age factors on JPCP long-term performance, and perform analysis for CRCP (see figure 82).

Figure 82.  Screen shot.  Project type selection screen for Hiperpav II analysis.  A new project screen with “Early-age JPCP” selected sits on top of the Hiperpav 3 screen.

Figure 82. Project type selection screen for HIPERPAV II analysis.

The input screens for these three project types are described below.

7.1 PROJECT INFORMATION

After the user selects a project type in the new project window (figure 82), the software creates a strategy with default inputs that the user can modify with the inputs for a specific situation.

There are two primary sections in HIPERPAV II: the project information section discussed below, and strategies section discussed later.

Regardless of the project type, project information is needed for all of the HIPERPAV II analyses. Under the project information screen (figure 83), the user can enter general project information, including the project name, project ID, section name, and beginning and ending stations. A comment box is also provided for any additional information.

Figure 83.  Screen shot.  Project information screen for Hiperpav II analysis.  An Untitled Hiperpav 3 screen is shown.  The project information button on top is pushed, and project information is highlighted, which allows the user to enter project information such as Project Name, Project ID, Section Name, Begin Station, End Station, and Comments.

Figure 83. Project information screen for HIPERPAV II analysis.

7.1.1 Geography Screen

In the geography screen, the user can select the position within the 50 U.S. States where the project is located (figure 84). Historical weather data from nearby weather stations to the site can be used during the analysis if desired. Climatic information including air temperatures, windspeed, relative humidity, cloudiness, and annual rainfall conditions will be estimated based on this location from the nearest weather stations. Section 7.3.7 provides additional information on selecting the project location.

To navigate the geography screen, five icons are available for use. The user can:

increase magnification icon Magnify the map.

decrease magnification icon Decrease magnification of the map.

select location of the project icon Select the location of the project.

drag icon Drag the map.

zoom icon Zoom to show the entire map.

Figure 84.  Screen Shot.  Geography screen for Hiperpav II analysis.  Here, geography is highlighted, which allows the user to pick a location of interest, either from the map or from a scroll down menu.

Figure 84. Geography screen for HIPERPAV II analysis.

7.1.2 Monthly Weather Data Screen

Monthly rainfall data is used for the long-term analysis only. If the user has more accurate rainfall data, it can be entered in the monthly weather data screen (figure 85).

Figure 85.  Screen Shot.  Monthly weather data screen for Hiperpav II analysis.  A continuation screen shot from figure 84.  Here, monthly weather data is highlighted, which allows the user to either load data from database by selecting “Estimated Monthly Data from Location,” or manually enter data by selecting “User-Defined.”

Figure 85. Monthly weather data screen for HIPERPAV II analysis.

7.2 EARLY-AGE JPCP ANALYSIS

After the user selects option 1 in the project type selection screen (figure 82), and the general project information is entered, one or more strategies for that specific location can be created for analysis under the strategies section (figure 86).

7.2.1 Strategies

The strategies section contains, in tree view, the specific input data required to perform an early-age JPCP analysis under the categories of:

7.2.1.1 Strategy Toolbar

It is possible to have multiple strategies in one project. Each of the strategies can have different design, materials and mix design, construction, and environment inputs.

The strategy toolbar is provided to manipulate strategies. The strategy toolbar includes the following commands:

add a strategy icon Add a strategy

copy the currently selected strategy icon Copy the currently selected strategy.

delete a strategy icon Delete a strategy.

validate a strategy icon Validate a strategy to verify that inputs are within allowable ranges.

run an analysis icon Run an analysis for the currently selected strategy.

All the above functions can also be performed by right clicking on the strategy name.

7.2.1.2 JPCP Early-Age Strategy Information

To perform an early-age JPCP analysis, a new strategy must be added. With the strategy created, the strategy information including design, materials and mix design, construction, and environment information can be entered. The screens where this information is entered are listed in a menu format on the left side of the HIPERPAV II screen. As shown on the strategy information screen (figure 86), the user can enter his or her name and chosen reliability level. After the strategy is run, the date of analysis also will appear on this screen. A comment box is available for each strategy.

Figure 86.  Screen Shot.  Strategy information screen for Hiperpav II analysis.  A Project 1 dot HP3 – Hiperpav 3 screen is shown.  The strategies button on top is pushed, and strategy information is highlighted, which allows the user to enter information such as User Name, Reliability Level, Data Last Analyzed, and Comments.

Figure 86. Strategy information screen for HIPERPAV II analysis.

7.2.2 Design Inputs

Under the design inputs, the user can enter the pavement design information, including pavement geometry, dowel design, and slab support information.

7.2.2.1 Geometry Screen

Under geometry (figure 87), fields to enter the slab thickness, slab width, and transverse joint spacing are provided. For any input in HIPERPAV II, the user can select from several either metric or English units and can convert between unit systems. To convert value, check the convert value box and click on the desired unit.

Figure 87.  Screen Shot.  Geometry screen.  An Untitled – Hiperpav 3 screen is shown.  The strategies button on top is pushed, and geometry is highlighted under design to the left, which allows the user to enter information such as New Slab Thickness, New Slab Width, and Transverse Joint Spacing.

Figure 87. Geometry screen.

A schematic of the geometry input entered in HIPERPAV II is shown in figure 88. The slab width that should be input in HIPERPAV II is the widest spacing between adjacent longitudinal joints. Similarly, the transverse joint spacing in the case of random joint spacings should be the largest spacing between transverse joints (although the user can run each joint spacing as a separate strategy to compare the results).

Figure 88.  Sketch.  Schematic of the geometry input in Hiperpav II.  Sketch shows a 3-D slab and information such as Transverse Joints, Shoulders, Slab Width, Paint Stripe, Longitudinal Joints, Joint Spacing, and Paving Width.

Figure 88. Schematic of the geometry input in HIPERPAV II.

7.2.2.2 Dowels Screen

The dowels design information is entered only if the user wants to evaluate the bearing stresses at the dowels during the early age. The dowel analysis module can assess how dowel bars affect the early-age performance of the joint in jointed concrete pavements (e.g., how the bearing stress of the dowel on the concrete changes as a function of temperature and time will be calculated). If the bearing stress is greater than the concrete's early-age bearing strength, then it is probable that the dowel will loosen in the pavement at a faster rate than if the bearing stress was designed to be less than the concrete's bearing strength. Cyclic loadings, particularly during the early age, commonly cause crushing of the concrete at the dowel interface due to the high stresses at those locations. Voids at the dowel-concrete interface will eventually cause the dowel to provide reduced load transfer efficiency.

If this module is enabled, then the dowel size, Poisson's ratio, and modulus must be input (figure 89). In addition, the 28-day effective modulus of dowel support is required. The modulus of dowel support is defined as "the bearing stress in the concrete mass developing under unit deflection of the bar."(54)Additional information is required on concrete compressive strength under the PCC properties screen if a dowel analysis is to be performed.

Figure 89.  Screen Shot.  Dowel input screen.  Here, dowels is highlighted under Design, which allows the user to enter information such as Dowel Size (Diameter), Dowel Poisson’s Ratio, Dowel Modulus, and 28-Day Effective Modulus of Dowel Support.  It also has a checkbox for “Includes Dowels in The Early Age Analysis.”

Figure 89. Dowel input screen.

7.2.2.3 Slab Support Screen

The slab support screen requires information on the material type directly underneath the slab (see figure 90). Options are provided for different subbase types or subgrade conditions (see pulldown menu in figure 91). Additionally, information on the subgrade stiffness in terms of modulus of subgrade reaction (k-value) or subgrade resilient modulus, and subbase thickness and subbase elastic modulus are required.

The k-value may be obtained by: 1) correlations with soil types and other subgrade support tests (i.e., CBR, R-value, etc.); 2) determination of elastic modulus of individual layers (e.g., with the use of deflection testing) and backcalculation methods; or 3) plate load testing methods.

AASHTO T 274 may be used to characterize relatively low stiffness materials such as natural soils, unbound granular materials, stabilized materials, and asphalt concrete. Alternatively, the repeated-load indirect tensile test method ASTM D 4123 may be used to test bound or higher stiffness materials such as stabilized bases and asphalt concrete. The elastic modulus of base materials with a high cement content may be obtained with the ASTM C 469 test method.

Figure 90.  Screen Shot.  Slab support input screen.  A continuation screen from figure 89.  Here, slab support is highlighted under design, which allows the user to enter information such as Material Type, Subgrade Stiffness Type (Use lowercase K-Value or Use Subgrade Modulus), Subbase Properties (Thickness, Modulus), Axial Restraint (Estimate from Material Type, User-defined Linear, and User-defined Nonlinear), Critical Restraint Stress, Movement at Sliding, and Curvature.

Figure 90. Slab support input screen.

Figure 91.  Screen Shot.  Available subbase types or subgrade conditions.  A continuation screen from figure 90.   Screen shot shows the available subbase type or subgrade conditions in Hiperpav II, which are Asphalt Concrete Subbase (Rough), Asphalt Concrete Subbase (Smooth), Asphalt Stabilized Subbase (Rough), Asphalt Stabilized Subbase (Smooth), Cement Stabilized Subbase, Lime Treated Clay Subgrade, Untreated Clay Subgrade, and Unbound Aggregate Subbase.

Figure 91. Available subbase types or subgrade conditions.

The axial restraint conditions can be estimated based on the layer directly underneath the slab, or can be user-defined with linear or nonlinear functions that best match the restraint stress-movement curve from field restraint tests. A restraint or pushoff test consists of casting a small slab with an approximate thickness of the projected pavement on top of the same subbase type. A horizontal force is applied on one side of the slab with a hydraulic jack and a load cell is used to measure the applied load. Dial gages are installed at the opposite side to measure the slab movement. The restraint stress is determined from the applied force divided by the area of the test slab. The testing history of restraint stress and slab movement during the pushoff procedure is recorded, and the restraint stress at free sliding is evaluated.(55)

7.2.3 Materials and Mix Design Inputs

The second input category is materials and mix design. This input category includes the cement, PCC mix, PCC properties, and maturity data inputs.

7.2.3.1 Cement Screen

Under the cement input screen shown in figure 92, the user selects the type of cement either by ASTM cement type classification (pulldown in figure 93), user-defined oxides, or user-defined Bogue compounds. If selected, the user-defined oxides or user-defined Bogue compounds are entered in tabular form.

Blaine fineness can be estimated from the cement type or user-defined if this information is available.

Figure 92.  Screen Shot.  Cement input screen.  Here, cement is highlighted under materials and mix design, which allows the user to enter information such as Cement Definition: ASTM Cement Type Classification, User-Defined Oxides, and User-Defined Bogue Compounds (Lime, Silica, Alumina, Ferric oxide, Magnesia, Sulfur Trioxide, and Other); Blain Fineness Index: Estimate from Cement Type and User-Defined.

Figure 92. Cement input screen.

Figure 93.  Screen Shot.  Available cement types in Hiperpav II.  A continuation of figure 92.  Screen shot shows available ASTM Cement Type Classification in Hiperpav II, which are Type I, Type IP, Type IS, Type II, Type III, Type V.

Figure 93. Available cement types in HIPERPAV II.

A help screen that explains the heat of hydration effects of different cementitious materials in HIPERPAV II is available in this screen (see figure 94).

Figure 94.  Screen Shot.  Heat of hydration help screen in Hiperpav II.  The Hiperpav help screen is shown.  A continuation of figure 92.  Heat of Hydration is the main topic with section 1: Hydration of the Cement, subsection 1.1: Default Hiperpav Characterization; and section 2: Effect of Admixtures on Concrete Hydration, subsection 2.1: Fly Ash.

Figure 94. Heat of hydration help screen in HIPERPAV II.

7.2.3.2 PCC Mix Screen

The PCC mix screen is used to enter the mix design related information (figure 95). Under the PCC mix screen, the user selects the type of coarse aggregate used in the mix (see pulldown menu in figure 96), the user can also enter the type of chemical admixtures according to ASTM C 494 (see pulldown in figure 97). Note that Type A and F are not included, because they would not affect the analysis (hover over to see tool tip). If fly ash is used, it is necessary to specify the fly ash class and approximate calcium oxide (lime) content for Class F (see pulldown menu in figure 98). Batch proportions by mass are also entered in this screen.

 

help iconHelp icons are provided in several software screens to provide additional guidance on the proper selection of inputs.

Figure 95.  Screen Shot.  PCC mix input screen.  Here, PCC mix is highlighted under materials and mix design, which allows the user to enter information such as  Aggregate Type, Admixtures (ASTM C494 Type), Fly Ash Class, and Batch Proportions, which include Content Unit (Constituents: Coarse Aggregate, Fine Aggregate, Water, Cement (Type I), Fly Ash, GGBF Slag, and Silica Fume), Water/Cement Ratio, Water/Cementitious Ratio.

Figure 95. PCC mix input screen.

 

Figure 96.  Screen Shot.  Available coarse aggregate types in Hiperpav II.  A continuation of figure 95.  Screen shot shows available coarse aggregate types in Hiperpav II, which are Basalt, Granite/Gneiss, Limestone, Sandstone, and Siliceous Gravel.

Figure 96. Available coarse aggregate types in HIPERPAV II.

 

Figure 97.  Screen Shot.  Available admixture types in Hiperpav II.  A continuation of figure 95.  Screen shot shows available admixture types in Hiperpav II, which are None, Type B–Retarding, Type C–Accelerating, Type D–Water-reducing and retarding, Type E–Water-reducing and accelerating, and Type G–Water-reducing, high range, and retarding.

Figure 97. Available admixture types in HIPERPAV II.

Figure 98.  Screen Shot.  Fly ash class drop-down menu.  A continuation of figure 95.  Screen shot shows available fly ash class from Hiperpav II, which are None, Class C, Class F (CaO less than or equal to 7 percent), and Class F (CaO greater than 7 percent).

Figure 98. Fly ash class drop-down menu.

7.2.3.3 PCC Properties screen

The PCC properties screen includes inputs for the 28-day strength, concrete CTE, 28-day modulus, and ultimate drying shrinkage (figure 99).

The 28-day tensile strength is determined through the ASTM C 496 or AASHTO T 198 splitting tensile test. Alternatively, the user can estimate the splitting tensile strength based on relationships with other strength types (see pulldown menu in figure 100). In the equations shown on the screen, "STR" is the PCC 28-day strength type. The splitting tensile strength (STS) is used to predict the formation of cracks in the early-age JPCP. The compressive strength (CS) is used to assess dowel-bearing stresses at early ages with the dowel analysis module. Default conversion values are provided in HIPERPAV II, however, the user is cautioned to develop project specific coefficients based on laboratory testing for the mix in study (see figure 101).

The PCC modulus of elasticity may be determined with ASTM C 469. AASHTO T 160 test procedure at 40 percent relative humidity may be used to determine PCC ultimate drying shrinkage. This test is equivalent to ASTM C 157. The PCC CTE can be entered or estimated by HIPERPAV II based on the coarse aggregate CTE. The latter can be user-defined or estimated based on default values. AASHTO has recently developed a provisional standard that can be used for testing the concrete CTE: TP-60-00, "Standard Test Method for the Coefficient of Thermal Expansion of Hydraulic Cement Concrete."

Figure 99.  Screen Shot.  PCC properties screen for early-age JPCP analysis.  Here, PCC properties is highlighted under materials and mix design, which allows the user to enter information such as PCC 28-day Strength:  Strength Type, 28-day Strength (STR); Strength Coefficients: Estimate from Strength Type or User-Defined; PCC 28-day Modulus:  Estimate from Strength or User-Defined; PCC Ultimate Shrinkage: Estimate from Mix or User-Defined; Coefficient of Thermal Expansion: Estimated from Aggregate Type, User-Defined Aggregate CTE, or User-Defined PCC CTE.

Figure 99. PCC properties screen for early-age JPCP analysis.

Figure 100.  Screen Shot.  Strength type drop-down menu.  A continuation of figure 99.  Screen shot shows available strength types in Hiperpav II, which are 3rd Point Flexural, Center Point Flexural, Compressive, and Splitting Tensile.

Figure 100. Strength type drop-down menu.

Figure 101.  Screen Shot.  Help icon under PCC properties screen.  A continuation of figure 99.  The 28-day value screen is shown with recommendations for laboratory testing for a specific mix to determine conversion coefficients.

Figure 101. Help icon under PCC properties screen.

7.2.3.4 Maturity Data Screen

The last input under the materials and mix design category is used to enter maturity data, if available. Two methods of maturity are provided: the temperature-time factor with the use of the Nurse-Saul maturity relationships according to ASTM 1074; or the equivalent age method with the use of the Arrhenius relationships. For either method, maturity and strength information are entered in a gridline and automatically plotted on the screen, as shown in figure 102.

Figure 102.  Screen Shot.  Maturity data input screen.  Here, maturity data is highlighted under materials and mix design, which allows the user to enter information such as Default Strength Gain; Use Maturity Data; Number of Points; Strength Type; Maturity Calculation Method: Temperature-Time Factor (Datum Temperature), Equivalent Age; Maturity and Strength Data.

Figure 102. Maturity data input screen.

7.2.4 Construction Inputs

The third input category corresponds to construction information (figure 103). In this screen, the initial temperatures of the concrete mix and support layer are entered. Information on the curing method used is required, as is the age at which curing is applied. The age when curing is applied is defined as the time elapsed from construction to the time when the slab is covered with membrane curing compounds or polyethylene sheeting. If removable curing methods such as polyethylene sheeting are selected, the age when curing is removed is required. Available curing methods are presented in the drop-down menu for curing method shown in figure 104. A help icon is also available in the curing method inputs, which includes information on the effects of curing type selection.

Information on when joints are cut is also entered in this screen. Three sawcut options are available to the user:

In addition, if a given strength for opening to traffic is required, this information can be entered in the construction inputs screen. Figure 104 shows the strength types that can be entered. Also, as shown in figure 104, a help icon for strength for opening to traffic input is provided. After the analysis is run, the strength for opening to traffic is indicated in the analysis screen (later shown in figure 110) if this strength is achieved within the 72-hour analysis period.

Figure 103.  Screen Shot.  Construction screen.  Here, construction is highlighted, which allows the user to enter information such as Initial Conditions: Initial PCC Mix Temperature, Initial Support Layer Temperature; Curing Method: Curing Method, Age Curing Applied, Age Curing Removed; Sawcutting: Saw at Optimum Time, User-Defined Sawing Age, No Sawing, Skip Sawcutting Intervals, Sawing Age of Skip Joints; Strength for Opening to Traffic: Strength, Strength Type.

Figure 103. Construction screen.

Figure 104.  Screen Shot.  Drop-down menus and help screens under construction inputs.  Screen shot is a continuation of figure 103 with the drop-down menus for curing method and strength type shown.  Also shown are the Hiperpav Help—Effect of Curing Type Selection screen and the Strength for Opening to Traffic screen.

Figure 104. Drop-down menus and help screens under construction inputs.

7.2.5 Environment

The last input category corresponds to environment. Under this input category, the user can enter climatic conditions for the 72 hours following construction, as shown in figure 105.

Using the HIPERPAV II environmental database, it is possible to estimate temperature, windspeed, humidity, and cloud cover for any location selected in the geography screen (figure 84). Or, if the user wishes, specific hourly weather data can be entered in the gridlines provided. The information displayed in this screen corresponds to the climatic input selected in the data to display drop-down menu shown in figure 106.

Figure 105.  Screen Shot.  Environment screen.  Here, environment is highlighted, which allows the user to enter information such as Data To Display; Construction Date; Construction Time; a checkbox for “Prompt to update hourly data;” Display Tools: Unit, Elapsed Time, Time of Day, and a checkbox for Autoscale; Hour, Time, and Temperature inputs; Weather Data Tools: Estimate (Point, High-Low, Cold Front) or Estimate All.

Figure 105. Environment screen.

Figure 106.  Screen Shot.  Drop-down menu for selection of climatic input screen to display.  A continuation of figure 105.  Screen shot shows available data to display types in Hiperpav II, which are Temperature, Wind Speed, Humidity, and Cloud Cover.

Figure 106. Drop-down menu for selection of climatic input screen to display.

If the construction date or time are changed, it is important that all of the temperature, windspeed, humidity, and cloud cover inputs be estimated again using the "estimate" or "estimate all" icons to match the selected point in time based on the geographic location selected. "Estimate" updates only the data displayed on the current screen. "Estimate all" updates all the weather data under temperature, windspeed, humidity, and cloud cover. The "prompt to update hourly data" checkbox can be unchecked to automatically update weather data as a function of changes in construction time or construction date. Otherwise, a prompt asking to update weather data will appear whenever a change in construction time or date is made.

The user also has the options to:

Options for plotting weather data in terms of elapsed time from construction or for time of day are provided. In addition, an autoscale feature can be selected to cover a wider or narrower weather data range.

Figure 107.  Screen Shot.  Time-temperature distribution modified with the point feature.  Screen shot shows a plot of Temperature (Y-axis) versus Time of Day (X-axis) modified with the point feature within the weather data tools.  A point that does not fall on the wavy line is circled for emphasis.

Figure 107. Time-temperature distribution modified with the point feature.

Figure 108.  Screen Shot.  Time-temperature distribution modified with the high-low feature.  Screen shot shows a plot of Temperature (Y-axis) versus Time of Day (X-axis) modified with the high-low feature within the weather data tools.  Local high and low peaks on the wavy line are changed to diamond shapes for emphasis.  A local high that is quite lower than the other high peaks is also circled.

Figure 108. Time-temperature distribution modified with the high-low feature.

 

Figure 109.  Screen Shot.  Time-temperature distribution modified subjected to a cold front.  Screen shot shows a plot of Temperature (Y-axis) versus Time of Day (X-axis) modified with the cold front feature within the weather data tools.  The lowest point on the wavy line is circled for emphasis.

Figure 109. Time-temperature distribution modified subjected to a cold front.

7.2.6 Early-Age JPCP Analysis

With the input assigned, the user should save the inputs before performing an early-age JPCP analysis.

After the analysis is run, a window appears showing the analysis status, and the stresses at the top and at the bottom of the slab are plotted along with the development of the tensile strength in the concrete, as shown in figure 110.

Figure 110.  Screen Shot.  Analysis screen for early-age JPCP.  Here, analysis is highlighted, which allows the user to enter information such as Display Units (Y-axis): U.S. Customary (PSI), Metric (kilopascals); Time Display (X-axis): Elapsed Time, Time of Day; checkboxes for Autoscale and Hiperpav 2X Chart; Inputs such as Hour, Time, PCC Strength, Critical Stress, Critical Stress at Slab Top, and Critical Stress at Slab Bottom.  The information inputted resulted in opening strength of 351 PSI reached at 36 hours.

Figure 110. Analysis screen for early-age JPCP.

The numerical stress and strength values also are presented. The critical stress is the largest tensile stress, whether it is at the top or at the bottom of the slab. The numerical value of tensile stress and strength can also be obtained by hovering the mouse over the point of interest. The point in time when the strength for opening to traffic is reached is indicated here in green (highlighted with an arrow and circle for clarity).

Also, because more than one strategy can be analyzed with HIPERPAV II, comparisons can be made between the analysis results more easily. It is possible to toggle between the analysis screens by selecting one strategy and then the other one (for this, the user must have at least two strategies previously analyzed for comparison and toggle between them).

7.2.6.1 Interpretation of Analysis Results

HIPERPAV executes a series of powerful algorithms that calculate the PCC pavement stress and strength development continuously for the first 72 hours following placement. The user is presented with a graphical screen, which plots the results of the analysis as they are computed in real time. The user can observe the trend of the strength and stress development and assess the behavior of the pavement based on the specific user inputs. Following processing, HIPERPAV will then identify possible problem areas and inform the user that the potential for early-age damage is present with the given set of inputs.

Figure 110 shows an output screen from a typical run that demonstrates mix design, pavement design, and construction during PCC pavement placement that lead to a high probability of good performance. The strength curve is the top curve, which remains above the stress curve at all times during the first 72 hours. Note the cyclical manner in which the critical stress occurs. Peaks in the stress curve occur at critical periods when either the axial stresses are dominant or when curling stresses are dominant; the former is in the early morning hours, and the latter is just after midday. In this scenario, the probability for PCC pavement distress (random transverse cracking) is low, because the stress does not exceed the strength during the first 72 hours after placement.

Figure 111 shows an output screen from a typical run where the combination of mix design, pavement design, construction, and the conditions during placement are such that the probability of a poorly performing PCCP is significant. The inputs for this particular run differ from that in figure 110 only by the minimum ambient temperature. A lower value of minimum ambient temperature was used for the scenario presented in figure 111, thus resulting in a thermal shock, which can lead to premature cracking.

The objective of the HIPERPAV guidelines and the accompanying computer software module is to minimize damage to the PCCP during early ages (first 72 hours after placement). By following the general recommendations made in previous chapters of this document in conjunction with the software described in this manual, the chances for achieving this objective are increased substantially.

Figure 111.  Screen Shot.  Hiperpav II early-age analysis showing poor performance.  Same as figure 110 with different inputs, resulting in opening strength not reached within the first 72 hours, and stress exceeding strength after 9 hours.

Figure 111. HIPERPAV II early-age analysis showing poor performance.

7.2.6.2 Evaporation Rate Analysis

An analysis screen also is provided for evaporation rate (figure 112). The effect of curing methods is not considered on the evaporation rate analysis. The evaporation rate analysis is a function of the windspeed, relative humidity, air temperature, and HIPERPAV II predicted concrete temperatures. The predicted evaporation rate is presented in both graphic and tabular form in this screen.

Figure 112.  Screen Shot.  Evaporation rate analysis for early-age JPCP analysis.  Here, evaporation rate analysis is highlighted, which allows the user to enter information such as Display Units (Y-axis): U.S. Customary (pounds per squared foot per hour), Metric (kilograms per squared meter per hour); Time Display (X-axis): Elapsed Time, Time of Day; Inputs such as Hour, Time, Predicted Evaporation Rate; legend of color corresponding to windspeed for graph.

Figure 112. Evaporation rate analysis for early-age JPCP analysis.

7.2.6.3 Dowel Analysis

The dowel analysis screen compares the concrete bearing strength to the bearing stress generated in proximity to the dowels due to slab curling as shown in figure 113. Because slabs in the field may show different curling conditions from one slab to the other, two scenarios are considered: 1) two slabs with similar curling conditions (curl-curl stress); and 2) one slab flat and the adjacent slab in a curled shape (flat-curl stress). It is believed that the actual stress to which a concrete/dowel may be subjected would be intermediate between these two conditions. Like the analysis screen, the analysis results are presented in both a graphical and a tabular form.

Figure 113.  Screen Shot.  Dowel analysis screen for early-age JPCP.  Here, dowel analysis is highlighted, which allows the user to enter information such as Display Units (Y-axis): U.S. Customary (PSI), Metric (kilopascals); Time Display (X-axis):  Elapsed Time, Time of Day; Inputs such as Hour, Time, Bearing Strength, Curl-Curl Stress, and Flat-Curl Stress.

Figure 113. Dowel analysis screen for early-age JPCP.

7.3 JPCP LONG-TERM COMPARATIVE ANALYSIS

The main focus of the software is in the early-age behavior of JPCP. The long-term performance module should be used to further optimize early-age pavement design, materials selection, and construction procedures. The long-term performance modeling should examine relative (strategy comparison) rather than absolute performance, based on early-age factors. In this way, the long-term module can reinforce good paving practices. It must not be used for pavement (structural) design purposes.

In this section, the long-term comparative analysis of early-age strategies is demonstrated. The user should be familiar with the early-age JPCP analysis screens before proceeding with this section.

After the user selects option 2 in the project type selection screen (figure 82), the software creates two early-age JPCP strategies with default inputs that the user can modify with the inputs for his or her specific situation.

In the example in figure 114, a limestone coarse aggregate is selected for the second strategy to observe the impact of this change in the long term as compared to the siliceous coarse aggregate used in the first strategy. The strategies are automatically sorted alphabetically. After both strategies have been analyzed, the user can switch to the long-term analysis module by selecting "LT JPCP" under the strategy type pulldown menu.

Figure 114.  Screen Shot.  Strategy type pulldown menu.  An untitled Hiperpav 3 screen is shown with the strategy type menu pulled down to allow a user to pick EA JPCP or LT JPCP.  LT JPCP is highlighted here.

Figure 114. Strategy type pulldown menu.

After generating at least two early-age JPCP strategies and performing analyses on them, it is possible to run a long-term JPCP analysis.

This long-term analysis is designed to be comparative, so an early-age base strategy is selected and compared to the early-age comparison strategy. This allows the user to see how early-age factors affect long-term pavement performance. In this way, long-term performance effects can be reduced with the optimum selection of early-age factors.

7.3.1 Long-Term Strategy Information

Under strategy information, the user selects the base strategy and the comparison strategy (figure 115). Fields for user name and comments also are provided. After the strategy is run, the date of analysis will appear on this screen.

Figure 115.  Screen Shot.  Strategy information for long-term JPCP analysis.  A project 1 dot HP3 Hiperpav 3 screen is shown.  The strategies button on top is pushed, and strategy information is highlighted, which allows the user to enter information such as User Name, Base Strategy, Comparison Strategy, Date Last Analyzed, and Comments.

Figure 115. Strategy information for long-term JPCP analysis.

7.3.2 Performance Parameters

In the performance parameters screen, the user is able to enter the long-term analysis period, the reliability level, and the initial roughness index (figure 116). An option is also provided to assess the effect of built-in curling. Built-in curling is a term used for the curling state that develops at set and that later influences the curled shape of the slab as the thermal gradient is subsequently modified by the hydration process and climatic conditions (see section 2.4.2.2). HIPERPAV II considers two slab curling conditions:

In addition to the above inputs, the user can set the maximum allowable joint faulting, transverse cracking, longitudinal cracking, IRI, and serviceability in this input screen. These set limits are compared to the analysis results in the long-term analysis screen. If the analysis has predicted a level of distress that the user denotes as "terminating" (beyond the predefined distress threshold), HIPERPAV II reports it as the end of the service life for that analysis. However, the analysis continues until the analysis period is reached.

Figure 116.  Screen Shot.  Performance criteria screen for long-term JPCP analysis.  An untitled Hiperpav 3 screen is shown.  The strategies button on top is pushed, and performance parameters is highlighted, which allows the user to enter information such as Analysis Period, Reliability Level, and Initial Ride (IRI); a checkbox for “Considered Built-in Curling Effect;” Distress Thresholds: Maximum Joint Faulting, Maximum Transverse Cracking, Maximum Longitudinal Cracking, Maximum Ride (IRI), and Minimum Serviceability.

Figure 116. Performance criteria screen for long-term JPCP analysis.

7.3.3 Joint Design

Under the joint design the user selects weather dowels and tie bars are provided. Dowels reinforce the transverse joint while tie bars reinforce the longitudinal joint (figure 117).

Note: The inputs in the early-age dowel analysis module do not apply to the long-term analysis. If dowels are selected as the load-transfer device in the long-term analysis, then for every strategy, dowels are used in the analysis, regardless if they were used or not used in the early-age analysis.

The user can select from no transfer devices, only dowels, only tie-bars, or both dowels and tie bars. If dowels are selected, dowel spacing and length are required in addition to dowel size and modulus. If tie bars are selected, their size and spacing must be input. The transverse sawcut depth is also entered as part of the joint design inputs.

Figure 117.  Screen Shot.  Load-transfer design for long-term JPCP analysis.  Here, joint design is highlighted, which allows the user to enter information such as Load Transfer Device; Dowel Properties: Size (Diameter), Spacing, and Modulus; Tie-Bar Layout: Size and Spacing; Transverse Sawcut Depth.

Figure 117. Load-transfer design for long-term JPCP analysis.

7.3.4 Traffic Inputs

Traffic loading also must be input to the long-term analysis, in addition to the strategy information, distress thresholds, and load-transfer design input. Five input screens are provided for this category: one for general traffic information, three for traffic load spectra, and one for traffic growth rate.

7.3.4.1 General Traffic Input Screen

In the general traffic input screen (figure 118), the mean truck tire pressure is required. The user can enter a user-defined traffic load spectra or have HIPERPAV II estimate the load spectra based on the input annual ESALs for the design lane. The default traffic load spectra generated by HIPERPAV II is based on typical averages of rural and urban highway traffic (see section B.2.5.4 in appendix B, volume III of this report series). If the user-defined option is selected, the user can enter specific single axle, tandem axle, or tridem axle load values in the gridlines for traffic loading.

Figure 118.  Screen Shot.  General traffic loading screen for long-term JPCP analysis.  Here, general is highlighted under traffic loading, which allows the user to enter information such as Mean Truck Tire Pressure and Traffic Load Spectra: User-Defined Spectra or Estimate Spectra.

Figure 118. General traffic loading screen for long-term JPCP analysis.

7.3.4.2 Axle Load Spectra Input Screens

If the user enters the load spectra information, this information is entered in the single axle load, the tandem axle load, and the tridem axle load screens (figure 119).

For each traffic load spectra category, the number of single tire and dual tire axle loads per year are entered against the axle load group. The single tire, dual tire, or total load spectra per axle type is shown on the plot, depending on the column selected. The total number of axles per axle load group is the sum of the respective single tire and dual tire values.

Figure 119.  Screen Shot.  Single axle load screen for long-term JPCP analysis.  Here, single axle load is highlighted under traffic loading, which allows the user to enter information such as Axle Load Group, Single Tire, Dual Tire, and Total Single Axles.

Figure 119. Single axle load screen for long-term JPCP analysis.

7.3.4.3 Traffic Growth Rate Input Screen

Finally, the traffic growth rate information is entered in the growth rate screen (figure 120). Three different growth functions can be specified: linear, exponential, or logistic. Depending on the function selected, the input required varies. For any function shown, the traffic loading is divided by the base-year traffic loading and plotted against time. Two plot methods may be selected: annual or cumulative.

Figure 120.  Screen Shot.  Traffic loading growth rate screen for long-term JPCP analysis.  Here, growth rate is highlighted under traffic loading, which allows the user to enter information such as Growth Function and Annual Growth Rate (lowercase R in percent).  Linear was chosen in this screen shot, resulting in an equation ADT parenthesis lowercase T equals 100 percent times the sum of 1 plus lowercase R times the difference of lowercase T minus 1.

Figure 120. Traffic loading growth rate screen for long-term JPCP analysis.

7.3.5 Long-Term Analysis

With the input assigned, the user should save the project file before performing a long-term JPCP analysis. Icons to manipulate strategies similar to those available for early-age strategies can be used for long-term analyses (see section 7.2.1.1).

After the analysis is run, a window appears showing the analysis status. The base strategy first is analyzed followed by the comparison strategy (figure 121). For each strategy, several structural and functional distresses are predicted, including:

After the analysis is complete, users can now select the type of distress to plot by selecting the distress drop-down window (figure 122).

Each distress type is plotted as a function of time. The distress magnitudes are presented in tabular format at the end of the prediction lifetime or at the time the threshold value is reached, whichever occurs first.

Using the data values plot method, absolute values of the distress are provided (see plot method drop-down in figure 123. If the change plot method is chosen, the change in predicted distresses between the base strategy and the comparison strategy is plotted.

Figure 121.  Screen Shot.  Analysis screen for long-term JPCP.  Here, analysis is highlighted, which allows the user to enter information such as Distress: Joint Faulting, Transverse Cracking, Longitudinal Cracking, Ride (IRI), and Serviceability; and Plot Method.

Figure 121. Analysis screen for long-term JPCP.

Figure 122. Screen Shot. Drop-down menu for distress to plot.  A continuation of figure 121.  Here, the distress drop-down menu is shown with options the user can enter for type of distress to plot, such as Joint Faulting, Transverse Cracking, Longitudinal Cracking, Ride (IRI) and Serviceability.

Figure 122. Drop-down menu for distress to plot.

Figure 123. Screen Shot. Plot method drop-down menu. A continuation of figure 121.  Here, the plot method drop-down menu is shown with options the user may enter for the type of plotting method to use, such as Data Values and Change.

Figure 123. Plot method drop-down menu.

7.3.6 Multiple Long-Term Comparative Strategies

In the long-term module, if more than two early-age strategies are to be compared, this can be done by creating additional strategies with the strategy toolbar. Once the strategy is created, the additional strategies to compare are selected in the strategy information window, as previously described.

7.3.7 Early-Age and Long-Term JPCP Analysis-Sample Scenario

In this section, step-by-step instructions are provided to show the effect of aggregate selection on early-age thermal cracking potential and on long-term transverse cracking. The early-age inputs used for this exercise are presented in table 1. Input values in bold represent the inputs that are different for these two early-age strategies.

 

Table 1. Inputs for 11 a.m. and 7 p.m. times of placement strategies.

Input Category

Input Screen

Input Value

Project Information

Geography 

Austin, TX 

Monthly Weather Data

Estimate Monthly Data from Location

Strategy Information

Reliability Level

90%

Design

Geometry

New Slab Thickness = 250 mm

New Slab Width = 3.65 m

Transverse Joint Spacing = 4.5 m

Dowels Analysis

Disabled

Slab Support

Material Type = HMA (rough)

k-value = 55 MPa/m

Subbase Thickness = 15 mm

Subbase Modulus = 3500 MPa

Axial Restraint = Estimate from Material Type

Materials and Mix Design

Cement

ASTM Cement Type Classification = Cement Type I

Blaine Fineness Index = Estimate from Cement Type

PCC Mix

Aggregate Type = Siliceous River Gravel (Baseline Strategy)

Limestone (Comparison Strategy)

Admixtures = None

Fly Ash Class = Class F (CaO <= 7.0%

Coarse Aggregate = 1100 kg/m3

Fine Aggregate = 800 kg/m3

Water = 128 kg/m3

Cement = 284 kg/m3

Fly Ash = 60 kg/m3

GGBFS = 0 kg/m3

Silica Fume = 0 kg/m3

PCC Properties

28-Day Splitting Tensile Strength = 3.2 MPa

PCC 28-Day Modulus = 35,000 MPa

PCC Ultimate Drying Shrinkage = Estimate from Mix

PCC CTE = Estimate from Aggregate Type

Maturity Data

Default Strength Gain

Construction

Construction

Initial PCC Mix Temperature = 32.2 °C

Initial Support Layer Temperature = 31.7 °C

Curing Method = Single Coat Curing Liquid Compound

Age Curing Applied = 0 Hours

Sawcutting: Saw at Optimum Time

Skip Sawcutting: Not Used

Strength for Opening to Traffic = Not Relevant for This Analysis

Environment

Environment

Construction Date: July 18, 2002 (year irrelevant for analysis since temperatures are historical averages)

Construction Time: 11:00 a.m.

Temperature, Windspeed, Relative Humidity, and Cloud Cover = Estimate from Location

To perform this analysis, a new long-term JPCP analysis is selected in the project type selection screen (see figure 82). The first step is to select the project location under the geography screen of the project information tab (see section 7.1). If not yet chosen, use the following icons in the geography toolbar to select Austin, TX, in the U.S. map (see figure 84):

pin icon pin

zoom icon zoom

drag icon drag

Next, select "Estimate monthly data from location" under the monthly weather data screen (see figure 85).

To enter the early-age inputs for the siliceous river gravel strategy, select the strategies tab (see figure 86). To rename this strategy "Siliceous River Gravel," right click on it and select "Rename," or select "Rename" from the strategy option on the HIPERPAV II top menu. Under "Strategy Information," the reliability level is set to 90 percent. In the same way, all other inputs shown in table 1 are entered under the design, materials and mix design, construction, and environment input screens. For this exercise, data for temperature, windspeed, relative humidity, and cloud cover are set to the default values estimated for that location.

After all the above inputs are entered, the user may perform the analysis by clicking on the analysis icon analysis icon of the strategy toolbar (see section 7.2.1.1) or by selecting "Run Analysis" from the strategy option on the HIPERPAV II top menu. The results from this analysis are presented in figure 124. The results show a high risk of thermal cracking for the siliceous river gravel strategy.

Figure 124. Screen Shot. Analysis results for the siliceous river gravel strategy.  Here, the analysis results for a siliceous river gravel strategy is shown, which allows the user to enter information, such as Display Units: U.S. Customary (PSI), Metric (kilpascals); Time Display: Elapsed Time, Time of Day. Also shown is a table illustrating PCC Strength, Critical Stress, Critical Stress at Slab Top, and Critical Stress at Slab Bottom values for certain times during the day. Also shown is a plot of Tensile Stress and Strength versus Time of Day.

Figure 124. Analysis results for the siliceous river gravel strategy.

To proceed with the second strategy (using limestone coarse aggregate), the user may either enter the information on a new strategy or create a copy of the siliceous river gravel strategy. The later is recommended, since all previously entered data that is identical for the second strategy will be automatically copied to the second strategy. To do this, select the siliceous river gravel strategy and click on the copy icon copy strategy icon. "Copy of Siliceous River Gravel Strategy" appears under the strategy list. To rename this strategy "Limestone," right click on it and select "Rename," or select "Rename" from the strategy option on the HIPERPAV II top menu.

For the limestone strategy, change the aggregate type to limestone in the aggregate type drop-down menu under the PCC mix input screen (see figure 96).

After completing the inputs for the limestone strategy, perform the analysis by clicking on the analysis icon of the strategy toolbar. The result from this analysis is presented in figure 125. Less potential for thermal cracking is observed from the results of the limestone strategy as compared to the siliceous river gravel strategy.

Figure 125. Screen Shot. Analysis screen for limestone strategy.  Here, the analysis results for a limestone strategy is shown, which allows the user to enter information, such as Display Units: U.S. Customary (PSI), Metric (kilopascals); Time Display: Elapsed Time, Time of Day. Also shown is a table illustrating PCC Strength, Critical Stress, Critical Stress at Slab Top, and Critical Stress at Slab Bottom values for certain times during the day. Also shown is a plot of Tensile Stress and Strength versus Time of Day.

Figure 125. Analysis screen for limestone strategy.

After the two early-age strategies have been analyzed, the long-term analysis is performed using the inputs shown in table 2.

To begin the long-term analysis, select "LT JPCP" under the analysis type drop-down (see figure 114). Then, under "Strategy Information," select the siliceous river gravel early-age strategy as the base strategy and the limestone early-age strategy as the comparison strategy (see figure 126). Complete the rest of the input screens with the information provided in table 2.

Table 2. Long-term inputs for aggregate selection sample scenario.

Input Category

Input Screen

Input Value

Strategy Information

Strategy Information

Base Strategy: Siliceous River Gravel

Comparison Strategy: Limestone

Performance Parameters

Performance Parameters

Analysis Period: 30 Years

Reliability Level: 90%

Initial Ride: 0.8 m/km

Consider Built-in Curling Effect: Yes

Distress Thresholds:

Maximum Joint Faulting: 3 mm

Maximum Transverse Cracking: 25%

Maximum Longitudinal Cracking: 25%

Maximum Ride (IRI): 2.7 m/km

Minimum Serviceability: 2.0

Joint Design

Joint Design

Load Transfer Devices: Dowels and Tie-bars

Dowel Properties:

Size (Diameter): 32 mm

Spacing: 300 mm

Modulus: 200,000 MPa

Tiebar Layout:

Size: 19.1 mm

Spacing: 0.9 m

Transverse Sawcut Depth: 33% of Thickness

Traffic Loading

General

Mean Truck Tire Pressure: 0.84 MPa

Traffic Load Spectra: Estimate Spectra from ESAL

Annual ESAL (Design Lane): 250,000

Single Axle Load

Estimate from ESAL

Tandem Axle Load

Estimate from ESAL

Tridem Axle Load

Estimate from ESAL

Growth Rate

Growth Function: Linear

Annual Growth Rate: 3%

Plot Method: Annual

After all the inputs have been entered, the user may perform the analysis by clicking on the analysis icon analysis icon of the strategy toolbar (see section 7.2.1.1) or by selecting "Run Analysis" from the strategy option on the HIPERPAV II top menu. The joint faulting analysis is presented after the analysis is run. Transverse cracking may be selected from the distress to plot drop-down menu (see figure 122). The transverse cracking results for the siliceous river gravel and limestone early-age strategies are shown in figure 127. The strategy with siliceous river gravel coarse aggregate shows a greater development of transverse cracking compared to the strategy with limestone coarse aggregate. From this analysis, selecting the limestone coarse aggregate source with a lower CTE appears to result in better performance, both in the early age and in the long term.

Figure 126. Screen Shot. Selection of early-age strategies to compare.  Here, strategy information is highlighted, which allows the user to enter information such as User Name, Base Strategy, Comparison Strategy, Date Last Analyzed, and Comments.

Figure 126. Selection of early-age strategies to compare.

Figure 127. Screen Shot. Long-term strategy transverse cracking results.  Here, analysis is highlighted, which allows the user to enter information, such as Distress:  Joint Faulting, Transverse Cracking, Longitudinal Cracking, Ride (IRI), Serviceability; and Plot Method.

Figure 127. Long-term strategy transverse cracking results.

7.4 EARLY-AGE CRCP ANALYSIS

In this section, the analysis for an early-age CRCP project is demonstrated.

After the user selects option 3 in the project type selection screen (figure 82), the software creates an early-age CRCP strategy with default inputs that the user can modify with the inputs for the specific situation. The input screens are similar to those required for the early-age JPCP analysis, with some differences.

After the general project information is entered, as shown in section 7.1, one or more strategies for that specific location can be created for analysis under the strategies section.

7.4.1 Early-Age CRCP Strategies Section

The strategies section contains, in tree view, the specific input data required to perform an early-age CRCP analysis under the categories of:

7.4.1.1 Strategy Toolbar

For information on the strategy toolbar, see section 7.2.1.1.

7.4.1.2 CRCP Strategy Information

Under the strategy information screen, the user can enter his or her name. After the strategy is run, the date of analysis also will appear on this screen. A comment box is available for each strategy. This screen is similar to the strategy information screen for early-age JPCP analyses, although no reliability information is required (see figure 86).

7.4.2 CRCP Design Inputs

Under the design inputs, the user can enter the pavement design information, including pavement geometry, steel design, and slab support information.

7.4.2.1 CRCP Geometry

The geometry screen (figure 128) provides fields to enter the slab thickness and slab width. The slab width that should be input in HIPERPAV II is the widest spacing between adjacent longitudinal joints.

Figure 128. Screen Shot. Geometry screen for early-age CRCP analysis.  Here, geometry is highlighted, and the user is allowed to enter information such as New Slab Thickness; and Desirable CRCP Cracking Characteristics: Minimum Crack Spacing, Maximum Crack Spacing, and Maximum Crack Width.

Figure 128. Geometry screen for early-age CRCP analysis.

In addition, long-term desirable CRCP cracking characteristics (after 1 year), including minimum and maximum crack spacing and average crack width, can be entered; these later will be shown by the dashed lines in the early-age CRCP analysis plot.

7.4.2.2 CRCP Steel Design

The longitudinal steel information can be input to the program in the steel design screen (figure 129). Information about the longitudinal steel layout and steel properties is required. For the steel layout, the bar spacing and size are needed, as well as the number of steel mats. The steel properties needed are the steel yield strength, steel modulus, and CTE. HIPERPAV II then calculates, based on the input, the percent steel for the current slab thickness.

Figure 129. Screen Shot. Steel design screen for early-age CRCP analysis.  Here, steel design is highlighted, which allows the users to enter information, such as CRCP Longitudinal Steel Layout: Bar Spacing, Bar Size, and Number of Bar Mats; and CRCP Steel Properties: Yield Strength (Grade), Steel Modulus, and Steel CTE. Also shown is the calculated “Percent Steel for Current Slab Thickness” for the information entered by the user.

Figure 129. Steel design screen for early-age CRCP analysis.

7.4.2.3 Slab Support

For information about this screen, see section 7.2.2.3.

7.4.3 Materials and Mix Design Inputs

The second input category is materials and mix design (see section 7.2.3)

7.4.4 Construction

The third input category corresponds to construction information. The construction screen for the early-age CRCP analysis is shown in figure 130. In this screen, the initial temperatures of the concrete mix and support layer are entered. Information on the curing method used (pulldown) is required along with the time of application, and if removable curing methods are selected, the age when curing is removed also is required.

In addition, if the effect of construction traffic is enabled, loading conditions are required. The user can assess the influence of construction traffic on the performance of the CRCP by inputting the days before the wheel load is applied, the wheel load, and the radius of the circular loaded area.

Figure 130. Screen Shot. Early-age construction screen for early-age CRCP analysis.  Here, construction is highlighted, which allows users to enter information such as Initial Conditions: Initial PCC Mix Temperature and Initial Support Layer Temperature; Curing Method: Curing Method, Age Curing Applied, and Age Curing Removed; Construction Traffic: Option to Enable/Disable Effects of Construction Traffic, Days Before Wheel Load Applied, Wheel Load, and Radius of Circular Loaded Areas.

Figure 130. Early-age construction screen for early-age CRCP analysis.

7.4.5 Environment

The environment screen is the same as the early-age JPCP screen seen in section .

7.4.6 Post 72-Hour Temperatures

For the early-age CRCP analysis, the post 72-hour temperatures are required (figure 131). The user may estimate these inputs using the HIPERPAV II environmental database, or the user can define them. The critical analysis period corresponds to the time when construction traffic is applied. For this period, the minimum and maximum air temperatures are also required. In addition, after the critical analysis period, the date of the minimum temperature for the year is needed to estimate the pavement age for the largest temperature drop to which the pavement will be subjected. This date can be estimated by HIPERPAV II, or it can be defined by the user. Similarly, HIPERPAV II can either estimate the minimum and maximum temperatures, or these can be input.

Early age in HIPERPAV II includes the first 72 hours after construction for both JPCP and CRCP. Cracking in CRCP typically progresses until approximately 1 year after construction. After 1 year, cracking typically remains constant. The HIPERPAV II analysis extends beyond early age up to 1 year (early life) to realistically assess the behavior of CRC pavements inservice.

Figure 131. Screen Shot. Post 72-hour air temperatures screen for early-age CRCP analysis.  Here, post 72-hour temperatures is highlighted.  An estimate button is shown to estimate the air temperatures from location. The user is also allowed to enter information such as Critical Analysis Period: Minimum Air Temperature and Maximum Air Temperature; Air Temperatures After Critical Analysis Period: Estimate Date of Minimum Temperature, User-Defined Date of Minimum Temperature, Minimum Air Temperature, and Maximum Air Temperature.

Figure 131. Post 72-hour air temperatures screen for early-age CRCP analysis.

7.4.7 CRCP Analysis

With the input assigned, the user should save the file before performing an early-age CRCP analysis. The analysis screen for an early-age CRCP strategy is shown in figure 132. After the analysis is run, a window appears showing the analysis status. The analysis output, in tabular format, includes the average crack spacing and its standard deviation, the average crack width, the maximum steel stress, and the bond development length. Interpretation of the analysis is performed by comparing the cracking results with the thresholds previously set in "Desirable CRCP Cracking Characteristics" in the geometry input screen. These thresholds are also presented in the analysis screen for comparison.

The plot can either show the incremental or cumulative frequency versus crack spacing. Plots for three different points in time are provided: 1) at 3 days after construction; 2) for the critical analysis period when construction traffic is applied; and 3) at 1 year of age ("Time Period to Plot" pulldown in figure 133). The crack spacing frequency also is presented in tabular format. The numerical value of crack spacing frequency can also be obtained by hovering the mouse over the point of interest in the plot. As shown in figure 134, a help window is provided to caution the user on the validation of the crack width model incorporated in HIPERPAV II. This window also defines the analysis period used for CRCP behavior analyses.

Also, because more than one strategy can be analyzed with HIPERPAV II, comparisons can be made between the analysis results more easily. It is possible to toggle between the analysis screens by selecting one strategy and then the other one (for this, the user must have analyzed at least two strategies previously for comparison and toggled between them).

An analysis screen also is provided for evaporation rate. The evaporation rate analysis is the same as for the early-age JPCP (see section 7.2.6.2).

Figure 132. Screen Shot. Analysis screen for early-age CRCP.  Here, analysis is highlighted, which allows the user to enter information such as Display Units: U.S. Customary and Metric; and Crack Spacing Plot: Incremental and Cumulative.  Also shown is a help icon button, a table illustrating the frequency in percent values for various crack spacing lengths, and a plot of the frequency in percent values versus the crack spacing in feet.

Figure 132. Analysis screen for early-age CRCP.

Figure 133. Screen Shot. Drop-down menu for time period to plot.  A continuation of figure 132.  Here, the drop-down menu for time period to plot is shown, with options the user can enter for the time period to plot, such as 72 Hours, Critical Analysis Period, and 1 year.

Figure 133. Drop-down menu for time period to plot.

Figure 134. Screen Shot. Help icon under CRCP analysis screen. A continuation of figure 132.  Here, the help information is displayed when the user chooses the help icon.

Figure 134. Help icon under CRCP analysis screen.

7.4.8 Interpretation of CRCP Analysis

One of the objectives of the CRCP analysis is to achieve a crack spacing distribution that falls within the predefined thresholds, with only a small percentage of cracks falling above and/or below the maximum and minimum crack spacing limits, respectively, as shown in figure 132. In this analysis, both crack width and crack spacing are within acceptable limits. An example of a CRCP analysis with poor cracking characteristics is shown in figure 135. In this second analysis, neither crack spacing nor crack width fall within acceptable limits. In addition, a large percentage of crack spacings fall below the predefined minimum crack spacing threshold.

Figure 135. Screen Shot. CRCP analysis with poor cracking characteristics.  Here, analysis is highlighted, which allows the user to enter information such as Display Units: U.S. Customary and Metric; and Crack Spacing Plot: Incremental and Cumulative. Also shown is a help icon button, a table illustrating the frequency in percent values for various crack spacing lengths, and a plot of the frequency in percent values versus the crack spacing in feet.

Figure 135. CRCP analysis with poor cracking characteristics.

It is also recommended to compare the maximum steel stress predicted with the steel yield strength to evaluate the potential of wide cracks or steel rupture. The bond development length is predicted to compare with the predicted crack spacing. The user should achieve a minimum average crack spacing of twice the bond development length to ensure good CRCP behavior. For additional information on CRCP behavior, refer to section.

7.5 HIPERPAV II Reports

For any HIPERPAV II analysis, reports can be generated by clicking on the print icon on the upper icon toolbar or by selecting "Print" from the file menu. After selecting "Print," a HIPERPAV II report screen is displayed as shown in figure 136. In this screen, options are available to select printing options under the setup icon, send the report to the printer as-is under the print icon, preview the report under the preview icon, and export the results to a Microsoft Excel spreadsheet under the export icon.

Figure 136. Screen Shot. Hiperpav II report screen.  Here, a report screen is shown with values for Project Information: Project Name, Project ID, Section Name, Begin Station, End Station, and Comments; and Project Environment Data: Latitude, Longitude, Elevation, Thornwaite Index.

Figure 136. HIPERPAV II report screen.

7.6 COMET Module

The Concrete Optimization, Management, Engineering, and Testing (COMET) module is a simplified derivation of the Concrete Optimization Software Tool (COST) that was developed by FHWA and the National Institute of Standards and Technology (NIST) for optimizing concrete mixes. COMET can be accessed within HIPERPAV II by selecting "COMET" in the tools top menu bar shown in figure 137. COMET optimizes concrete mixes based on early-age strength, 28-day strength, and cost.

Figure 137. Screen Shot. Tools drop-down menu in Hiperpav II.  Here, tools is selected from the Hiperpav toolbar. The drop-down menu is shown and COMET is highlighted. Other options in the tools drop-down menu are Event Log, File Checker, Reference Database, and Options.

Figure 137. Tools drop-down menu in HIPERPAV II.

Once selected, the COMET initial window appears as shown in figure 138. Inputs and outputs can be selected on the tree view on the left side of the screen. The inputs for the screen selected are displayed on the right. The first item on the tree view, "Event Log," reports potential problems that may arise while running the software. Also, the tree view shows two levels of inputs and outputs. Level 1 inputs generate the experiment design of trial batches required for optimization. Level 2 inputs produce a set of optimum mixes based on the trial batch test results, and desirability functions defined for each response (early-age strength, 28-day strength, and cost).

As shown in figure 138, mix constituents are limited to cement, pozzolan, water, coarse aggregate, and fine aggregate. For each constituent, the specific gravity and cost is required.

Figure 138. Screen Shot. Mix constituents input screen in COMET.  Here, mix constituents is highlighted, and the user can define information for the Specific Gravity and Constituent Cost, respectively, of Cement, Pozzolan, Coarse Aggregate, and Fine Aggregate.

Figure 138. Mix constituents input screen in COMET.

The user provides optimization ranges for coarse aggregate fraction, cementitious content, pozzolan substitution, and w/cm ratio (see figure 139).

Figure 139. Screen Shot. Factor limits input screen in COMET.  Here, factor limits is highlighted, and users can define the low and high limits for Coarse Percent Total Aggregate Weight, Cementitious Content, Pozzolan Percent Total Cementitious, and Water/Cementitious.

Figure 139. Factor limits input screen in COMET.

With the inputs in level 1, trial batches are developed for the experimental program by running the "Create Trial Batches" command button in the level 1 outputs screen, as shown in figure 140. As a result, 29 trial batches are generated, which can be displayed in volumetric (percent of total volume) and gravimetric (weight per unit of volume) form. The gravimetric form is presented in figure 141.

Figure 140. Screen Shot. Create trial batches command button in level 1 outputs.  Here, level 1 outputs is highlighted and a create trial batches button is shown. Also shown is a text box with information summarizing the amount and description of trial batches that will be created.

Figure 140. Create trial batches command button in level 1 outputs.

Figure 141. Screen Shot. Trial batches in kilograms per meter cubed, gravimetric form.  Here, trial batches—gravimetric is highlighted, with a table illustrating the following gravimetric data for each trial batch: Cement, Pozzolan, Water, Coarse Aggregate, Fine Aggregate, and Total.

Figure 141. Trial batches in kg/m3, gravimetric form.

In addition, as shown in figure 142, the cost for every trial batch is computed and default models in COMET are used to predict early-age strength and 28-day strength as a function of mix constituents. Predicted responses are only approximate and may be used in a planning stage in which the mix constituents may not be known or are not available for testing.

Figure 142. Screen Shot. Predicted responses for each trial batch.  Here, trial batches—response predictions is highlighted, with a table illustrating the Cost, Early-Age Strength, and 28-Day Strength for each trial batch.

Figure 142. Predicted responses for each trial batch.

After the level 1 inputs and outputs are complete, the user can proceed with level 2 analysis. To continue with this analysis, the user must enter the laboratory testing results for each response for the trial batches generated in level 1 (see figure 143). Alternatively, the user may fill in these values using default models by clicking on the "Reset to Predicted Values" command button, if a preliminary analysis is desirable.

Figure 143. Screen Shot. Lab results screen.  Here, response—lab results is highlighted, with a text box informing the user that the predicted responses from the previous section may be used, or values from lab tests may be inserted by the user. Also shown is a reset to predicted values button and a table in which the user may edit the Cost, Early-Age Strength, and 28-Day Strength values for each trial batch.

Figure 143. Lab results screen.

The user can assign desirability functions for each of the optimization responses under the cost, early-age strength, and 28-day strength desirability screens (a sample desirability function is shown in figure 144 for 28-day strength).

Figure 144.  Screen Shot. Desirability function for 28-day strength.  Here, desirability—28-day strength is highlighted and a plot of this table is shown, illustrating the desirability versus the 28-day strength in four points. Also, a table is shown where the user may define the value and desirability values for each point, and an update chart button is located below the table, in which the user may update the chart when new values are added.

Figure 144. Desirability function for 28-day strength.

After the level 2 inputs are entered, optimization of mixes is performed by clicking the "Optimize Mixes" command button in the level 2 outputs screen (see figure 145). With this command, regression models based on the results from the experimental program or from the predicted responses are developed as a function of the mix proportions.

Figure 145. Screen Shot. Level 2 outputs—command to optimize mixes.  Here, Level 2 outputs is highlighted and an optimize mixes button is shown with a text box. The text informs the user that optimized mixes may be created, and only the top 100 mixes that match the desirability curve will be shown.

Figure 145. Level 2 outputs—command to optimize mixes.

These models predict responses for a comprehensive set of mixtures within the factor limits specified in the factor limits screen. Finally, optimum mixes are identified in terms of the individual desirability for every response and the maximum overall desirability for all responses, and are presented in volumetric and gravimetric form. Optimum mixes are shown in volumetric form in figure 146.

Figure 146. Screen Shot. Optimum mixes sorted by desirability in volumetric form.  Here, volumetric is highlighted, and a table with the following information is shown for each optimized mix: Desirability, Cement, Pozzolan, Water, Coarse Aggregate, and Fine Aggregate.

Figure 146. Optimum mixes sorted by desirability in volumetric form.

In addition, the optimum mixes are also displayed in terms of the optimization factors, predicted responses, and individual response desirabilities as shown in figures 147, 148, and 149, respectively.

Figure 147. Screen Shot. Optimum mixtures in terms of optimization factors.  Here, factors is highlighted. A table illustrating the following calculated values for each optimized mix is shown: Desirability, Coarse Percent Total Aggregate Weight, Cementitious Content, Pozzolan Percent Total Cementitious, and Water/Cementitious.

Figure 147. Optimum mixtures in terms of optimization factors.

Figure 148. Screen Shot. Response predictions for optimum mixtures.  Here, response predictions is highlighted, with a table illustrating the following predicted response values for each mix: Desirability, Cost, Early-Age Strength, and 28-Day Strength.

Figure 148. Response predictions for optimum mixtures.

Figure 149. Screen Shot. Individual response desirabilities for optimum mixtures.  Here, desirability is highlighted, and a table illustrating the following predicted desirability values for each mix is shown: Desirability, Cost, Early-Age Strength, and 28-Day Strength.

Figure 149. Individual response desirabilities for optimum mixtures.

A report that can be used for printing the inputs and outputs for the entire analysis is generated, as shown in figure 150. To print, click on the print icon on the upper icon toolbar.

Figure 150. Screen Shot. Report view for printing purposes.  Here, report is highlighted. Three tables are shown with values for Level 1 Inputs: Mix Constituents: Specific Gravity and Constituent Cost, respectively, for Cement, Pozzolan, Coarse Aggregate, and Fine Aggregate; Level 1 Inputs: Factor Limits: Low and High Limits, respectively, for Coarse Percent Total Aggregate Weight, Cementitious Content, Pozzolan Percent Total Cementitious, and Water/Cementitious; Level 1 Outputs: Trial Batches—Volumetric: Values for Run Number, Mix Number, Cement, Pozzolan, Water, Coarse Aggregate, and Fine Aggregate.

Figure 150. Report view for printing purposes.

 

Previous | Table of Contents | Next


The Federal Highway Administration (FHWA) is a part of the U.S. Department of Transportation and is headquartered in Washington, D.C., with field offices across the United States. is a major agency of the U.S. Department of Transportation (DOT).
The Federal Highway Administration (FHWA) is a part of the U.S. Department of Transportation and is headquartered in Washington, D.C., with field offices across the United States. is a major agency of the U.S. Department of Transportation (DOT). Provide leadership and technology for the delivery of long life pavements that meet our customers needs and are safe, cost effective, and can be effectively maintained. Federal Highway Administration's (FHWA) R&T Web site portal, which provides access to or information about the Agency’s R&T program, projects, partnerships, publications, and results.
FHWA
United States Department of Transportation - Federal Highway Administration