Non-Nuclear Methods for Compaction Control of Unbound Materials (2014)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

90 chapter five STIFFNESS-BASED SPECIFICATIONS FOR COMPACTION CONTROL OF UNBOUND MATERIALS INTRODUCTION Although current common quality control specifications that use dry density and/or moisture content criteria are relatively straightforward and practical, they do not reflect the engineer- ing properties of unbound materials necessary to ensure a high-quality product. In addition, the design of pavements and embankments is based on stiffness and/or strength parameters. Thus, there is a missing link between the design process and construction quality control practices of unbound materials. To address this problem, several studies have been conducted by state DOTs to develop quality control procedures for construc- tion based on a criterion that closely correlates to the stiffness/ strength parameters used in the design process. This chapter presents a review of stiffness/strength-based specifications that have been developed in the United States and Europe. In addition, it summarizes the experience of state DOTs that have implemented these specifications in field projects. The infor- mation in this chapter is derived from published documents identified in the literature review as well as interviews with selected state DOTs that were conducted in this study. REVIEW OF CURRENT STIFFNESS/ STRENGTH-BASED COMPACTION CONTROL SPECIFICATIONS Currently, only a few state DOTs have developed compaction control specifications for unbound materials that are based on in situ stiffness/strength measurements. Table 19 provides the links for these specifications. Although some DOTs have developed such specifications primarily for subgrade soils or select types of base course materials, other states have devel- oped comprehensive specifications that cover a wide range of unbound materials. A summary of the stiffness- and strength- based compaction control specifications that have been devel- oped in the United States and Europe is presented in the following sections. Minnesota DOT Specifications The Minnesota DOT has been one of the leading state DOTs in developing and implementing stiffness-based specifications for compaction control of unbound materials. Currently, the Minnesota DOT has specifications for the using the DCP and the LWD in assessing the quality of compacted layers of unbound materials during construction. The following sec- tions discuss these specifications. DCP Specification The DCP has been used by the Minnesota DOT as a compac- tion quality control device for pavement edge drain trenches since 1993. In 1999, the agency implemented a new speci- fication using the DCP penetration index (DPI) for com- paction acceptance of three types of base course aggregate materials, specifically classes 5, 6, and 7. In this specifica- tion, the DPI value is recorded in millimeters (or inches) per blow and is the only parameter used to determine the compaction effort needed to reach the target CBR value for base aggregates. This specification was later modified to expand its applications to other granular materials by taking into account the effect of gradation and moisture content. The current Minnesota DOT DCP specification is used for compaction control of three types of unbound granular materials: base aggregates, granular subgrade materials, and edge drain trench filter aggregates. Although procedures for the granular materials are similar, a slightly modified proce- dure is specified for edge drain trench filter aggregate. For assessing compaction quality of base aggregates and granular subgrade materials, Minnesota DOT currently requires deter- mining the gradation and in situ moisture content of the com- pacted material as well as performing DCP tests on it. Readings for the first five DCP drops need to be recorded. The first two drops are treated as seating drops and are used to compute the SEAT value using Eq. 62. The penetration depth after the fifth drop is used to compute the DPI value using Eq. 63. The measured SEAT and DPI values are then compared with the maximum allowable SEAT and DPI values. These should be selected from Table 20 or by using Eq. 64, based on the tested material’s gradation properties [represented by the grading number (GN)] and recorded moisture content. The compacted material passes the test only if the measured SEAT and DPI values are found to be less than or equal to the maximum allowable values. However, Minnesota DOT also requires that the penetration of the five DCP drops be smaller than the tested layer thickness. –( )=SEAT Reading after seating 2 blows Initial Reading (62) [ ] =DPI Reading after 5 blows – Reading after 2 blows3 (63)

91 Maximum Allowable DPI mm blow 4.76 GN 1.68MC – 14.4 (64) ( ) = × + where MC = the moisture content at the time of testing, and GN = grading number obtained using the following equation: = + + + + + µ + µGN 25 mm 19 mm 9.5 mm 4.75 mm 2.00 mm 425 m 75 m 100 (65) The SEAT value requirement initially was included in the Minnesota DOT specification to ensure that the aggregate base layer had sufficient surface strength to allow construc- tion equipment, such as a paver, to operate on its surface without significant rutting. However, this requirement was maintained for other materials to make certain that thin, loose, or irregular surface material does not unduly affect measure- ments. Minnesota DOT’s specification uses the grading num- ber (GN) to represent the gradation properties of the tested granular base and subgrade materials. The GN is computed using Eq. 65, which requires performing a sieve analysis with seven of the most common sieves: 25, 19, 9.5, 4.75, and 2.00 mm and 425 and 75 µm. The GN increases with the increase of fine sand particles in the material. A material with an extremely fine gradation will yield a GN close to 7, whereas an extremely coarse material yields a GN close to zero. Materials with larger amounts of gravel and coarse sand State DOT Specification Links Minnesota DCP specification: http://www.dot.state.mn.us/materials/gbmodpi.html LWD specification: http://www.dot.state.mn.us/materials/gblwd.html Indiana DCP specification: http://www.state.in.us/indot/files/Fieldtesting.pdf LWD specification: http://www.in.gov/indot/div/mt/itm/pubs/508_testing.pdf Missouri http://www.modot.org/business/standards_and_specs/Sec0304.pdf Illinois http://www.dot.il.gov/bridges/pdf/S-33%20Class%20Reference%20Guide.pdf TABLE 19 LINKS FOR STIFFNESS- AND STRENGTH-BASED COMPACTION CONTROL SPECIFICATIONS GN In Situ Moisture (% by dry weight) Maximum Allowable Seating (mm) Maximum Allowable DPI (mm/blow) 3.1–3.5 <4.0 40 10 4.1–6.0 40 10 6.1–8.0 40 13 8.1–10.0 40 16 3.6–4.0 <4.0 40 10 4.1–6.0 40 12 6.1–8.0 45 16 8.1–10.0 55 19 4.1–4.5 <4.0 45 11 4.1–6.0 55 15 6.1–8.0 65 18 8.1–10.0 70 21 4.6–5.0 <4.0 65 14 4.1–6.0 75 17 6.1–8.0 80 20 8.1–10.0 90 24 5.1–5.5 <6.0 90 19 6.1–8.0 100 23 8.1–10.0 110 26 10.1–12.0 115 29 5.6–6.0 <6.0 110 22 6.1–8.0 120 25 8.1–10.0 125 28 10.1–12.0 135 32 Source: Minnesota DOT (2012). TABLE 20 MODIFIED DCP PENETRATION REQUIREMENTS

92 have small GN values. Such materials have larger strength and modulus values than do those with large amounts of fine sand. Therefore, strength and modulus values calculated from the DCP are expected to increase as the GN decreases. The DCP specification was modified to include GN as an input to ensure that a material with a larger GN remains acceptable even if it has lower modulus/strength values. This was done to remain consistent with Minnesota DOT’s preexisting con- struction standards. The Minnesota DOT provided a description of the stan- dard DCP device that should be used in this specification. However, there are a few configuration options available for the DCP, including change of hammer mass, type of tip, and recording method. The standard hammer mass is 8 kg, but a 4.6-kg alternative can be used. For pavement applica- tions, the 8-kg mass is used because of the highly compacted soil. The DCP tip can be a replaceable point or a disposable cone. The replaceable point stays on the DCP for an extended period—until damaged or worn beyond a defined tolerance— and is then replaced. The disposable cone remains in the soil after every test, making it easier to remove the DCP; however, a new disposable cone must be placed on the DCP before the next test. Manual or automated methods are available to gather penetration measurements. The reference ruler can be attached or unattached to the DCP. LWD Specifications Minnesota DOT has also developed specifications for using the LWD in compaction QC/QA of unbound materials. This specification provides two methods for developing accep- tance criteria for compacted unbound materials: the control strip method and the comparison test method. In the first method, a control strip meeting the requirements provided in Table 21 is constructed within the roadway core to determine the LWD target deflection value (LWD-TV). This is done by conducting LWD tests during the construction of the control strip at the rates shown in Table 22. The average deflection values of all LWD tests are plotted against the roller pass count to create a compaction curve. The LWD-TV for each lift is the minimum deflection of the compaction curve. Once the LWD-TV is determined, the LWD test is to be performed on the compacted layer at the rates specified in Table 23. The compacted layer is accepted only if the average deflection value of all performed LWD tests is less than 1.1 times the LWD-TV. For the comparison test method, six sets of tests compar- ing the LWD deflections with either the DPI (for granular and base) or the Minnesota DOT-specified density method (for nongranular) must be conducted. Each set should be spaced 304.8 mm (1 ft) longitudinally from the next, and spaced at Specification Material/Location Length Width Thickness Number of Lifts 2211, 2215 Base 100 m (300 ft) Layer Layer 1 2105, 2106 Roadbed embankment soil (excavation and borrow) Excavated embankment width Planned layer thickness, but not exceeding a maximum thickness of 4 ft (1.2 m) 1 every 304.8 mm (1 ft) Miscellaneous, trench, culvert or other tapered construction Embankment soil granular bridge approach treatments and other embankment adj. to structures (excavation and borrow) 100 m (300 ft) 1 every 608.6 mm (2 ft) Source: Minnesota DOT (2012). TABLE 21 REQUIRED CONTROL STRIP DIMENSIONS Specification Material Form Number Minimum Required Agency (Field Test Rate) 2211, 2215 Base G and B-601 4 tests/roller pass/lift 2105, 2106 Roadbed embankment soil (excavation and borrow) 2105, 2106 Miscellaneous, trench, culvert or other tapered construction Granular bridge approach treatments and other embankment adj. to structures (excavation and borrow) G and B-602 2 tests/roller pass/lift Source: Minnesota DOT (2012). TABLE 22 LWD TESTING RATES DURING CONTROL STRIP CONSTRUCTION

93 any width throughout the first 3,058.2 m3 (4,000 yd3). The LWD-TV can be determined by using the maximum deflec- tion measurement where DPI or density values are passing. This should continue until there are six passing comparison tests in locations that are close to failure. The LWD-TV should then be validated by conducting 10 additional LWD tests and performing three additional sets of comparison tests. The com- pacted layer is accepted only if the deflection value of the LWD test is less than the determined LWD-TV. Based on the results of several research projects, Minne- sota DOT developed LWD-TVs for different types of unbound materials. Table 24 provides a summary of LWD-TVs that were recommended for granular materials. These should be selected based on their grading number and moisture con- tent. For compacted, fine-grained soil, Minnesota DOT uses the plastic limit and field moisture content to determine the LWD-TVs, as shown in Table 25. It is worth noting that Min- nesota DOT requires that the moisture content of embank- ment materials should be maintained from 65% to 95% of the target moisture content. Because the properties of LWD devices might vary, Min- nesota DOT requires that the LWD used in testing have a plate diameter of 200 mm (7.9 in.) and a falling mass of 10 kg (22.1 lb), with a height of fall of about 500 mm (19.7 in.). Minnesota DOT specifies that the force applied by the LWD should be of 6.28 kN (1,539.9 lbf), which results in a stress of 0.2 MPa (29 psi) for the 200-mm (7.9-in.) diameter plate. Those requirements allow for obtaining consistent and reliable Specification Material Form Number Minimum Required Agency (Field Test Rate) 2211 Base GandB-604 1 LWD test/500 yd3 (CV) 1 LWD test/400 m3 (CV) 2215 Base 1 LWD test/3,000 yd 2 1 LWD test/2,000 m2 2105, 2106 Roadbed embankment soil (excavation and borrow) 1 LWD test/2,000 yd3 (CV) 1 LWD test/1,500 m3 Miscellaneous, trench, culvert or other tapered construction embankment soil (excavation and borrow) 1 per 2-ft thickness/250 ft 1 per 600-mm thickness/75 m Granular bridge approach treatments and other embankment adj. to structures (see Note 1) 1 per 2-ft thickness 1 per 600-mm thickness TABLE 23 LWD TESTING RATES (CONTROL STRIP METHOD) Grading Number (GN) Moisture Content (%) Target LWD Modulus Target LWD Deflection Zorn (mm) Keros/Dynatest (MPa) Zorn (MPa) 3.1–3.5 5–7 120 80 0.38 100 100 67 0.45 75 75 50 0.6 3.6–4.0 5–7 120 80 0.38 80 80 53 0.56 63 63 42 0.71 4.1–4.5 5–7 92 62 0.49 71 71 47 0.64 57 57 38 0.79 4.6–5.0 5–7 80 53 0.56 63 63 42 0.71 52 52 35 0.86 5.1–5.5 5–7 71 47 0.64 57 57 38 0.79 48 48 32 0.94 5.6–6.0 5–7 63 42 0.71 50 50 33 0.9 43 43 29 1.05 Source: Siekmeier et al. (2009). TABLE 24 LWD TARGET VALUES FOR GRANULAR MATERIAL

94 LWD data. They also ensure that the LWD influence depth extends to the bottom of a common lift thickness. Minnesota DOT also provides guidelines for conducting LWD tests. First, tests should be conducted immediately after compaction. Second, before the LWD test is performed, the surface of the tested layer should be leveled by removing any loose or rutted surface material to a depth of 50 to 100 mm (1.97 to 3.94 in.). Third, three LWD seating drops should be performed before the data are collected at any test location. This ensures that any permanent deformation of the surface material will not affect the LWD measurements. Once the LWD has been seated, three additional LWD drops should be performed at the same test location. The average of the maximum deflections or modulus values resulting from these drops is used as the mean value at the test location. As the modulus or deflection values change slightly during the three measurement drops, Minnesota DOT requires that this change does not exceed 10% of original modulus or deflection value. Otherwise, the material has to be additionally compacted before more LWD tests are conducted. Minnesota DOT also recommends that the LWD maximum deflections recorded during the three measurement drops range between 0.3 and 3.0 mm (0.012 and 0.12 in.). In addition, LWD tests are pro- hibited when the water table is less than 600 mm (23.6 in.) or when the embankment thickness is less than 300 mm (when no site preparation is needed) and 460 mm (when site preparation is needed). Finally, LWD devices are not to be used when the temperature falls below 5°C (41°F) to ensure that the device’s components, particularly the rubber buffers, work as intended. There is no practical upper limit on the temperature. Minnesota Experience In a follow-up interview, representatives of the Minnesota DOT stated that the DCP is used as the default device for Plastic Limit (%) Estimated Optimum Moisture (%) Field Moisture as a Percent of Optimum Moisture (%) DCP Target DPI at Field Moisture (mm/drop) Zorn Deflection Target at Field Moisture Minimum (mm) Zorn Deflection Target at Field Moisture Maximum (mm) Nonplastic 10–14 70–74 12 0.5 1.1 75–79 14 0.6 1.2 80–84 16 0.7 1.3 85–89 18 0.8 1.4 90–94 22 1 1.6 15–19 10–14 70–74 12 0.5 1.1 75–79 14 0.6 1.2 80–84 16 0.7 1.3 85–89 18 0.8 1.4 90–94 22 1 1.6 20–24 15–19 70–74 18 0.8 1.4 75–79 21 0.9 1.6 80–84 24 1 1.7 85–89 28 1.2 1.9 90–94 32 1.4 2.1 25–29 20–24 70–74 24 1 1.7 75–79 28 1.2 1.9 80–84 32 1.4 2.1 85–89 36 1.6 2.3 90–94 42 1.8 2.6 30–34 25–29 70–74 30 1.3 2 75–79 34 1.5 2.2 80–84 38 1.7 2.4 85–89 44 1.9 2.7 90–94 50 2.2 3 Source: Siekmeier et al. (2009). TABLE 25 TARGET LWD DEFLECTION VALUES FOR FINE-GRAINED SOILS

95 compaction control of granular materials and is currently used in about 50% of projects involving such materials. The LWD has been used to a much lesser extent by some districts mainly for granular materials but also for projects involving nongranular materials. The selection of compaction control device depends on the district’s experience with the different devices as well as the particular application. The DCP is not preferred in projects with large aggregate particles because of the high DCP refusal rate. At the same time, the LWD has posed problems in some districts because it is heavy, less portable, and needs annual calibration. The Minnesota DOT staff stated that the department had good experience with DCP and is planning to extend its use to other applications. How- ever, the DOT had mixed success with the LWD. Indiana DOT Specifications The Indiana DOT has recently developed and implemented comprehensive specifications for compaction control of various unbound materials based on the results of DCP and LWD tests. In those specifications, the Indiana DOT uses the DCP for clay, silty, or sandy soils, and granular soils with aggregate sizes smaller than 19 mm (¾ in.), and structural backfill sizes 1 in., ½ in., and Nos. 4 and 30. The LWD test is implemented to assess granular soils with aggregate sizes greater than 19 mm (¾ in.), coarse aggregate sizes Nos. 43, 53, and 73, and structural backfill sizes 2 and 1.5 in. Finally, the Indiana DOT allows using either the DCP or LWD for chemically modified soils. DCP Specifications The DCP acceptance criterion is based on the type of soil being tested. For clayey, silty, and sandy soils, the DCP accep- tance criteria is determined using Figure 92 based on the max- imum dry density and optimum moisture content of the soil of interest. These values are estimated in the field using the Indiana DOT’s modified version of the one-point Proctor test procedure (AASHTO T272). Indiana DOT specifies that the strength is measured after completion of the compaction for each 152.4 mm (6 in.) of the clayey soils, and for each 304.8 (12 in.) of silty and sandy soils. Therefore, the acceptance criterion for the DCP, determined from Figure 92, represents the minimum required number of DCP blows for a penetra- tion of 152.4 mm (6 in.) for clay soils and a penetration of 304.8 (12 in.) for silty and sandy soils. For granular materials—defined by the Indiana DOT as noncohesive soils with 35% or less passing sieve No. 200— the DCP test should be performed after the completion of compaction for each 457.2 mm (18 in.) lift of material. The DCP should first be penetrated into the material a depth of 152.4 mm (6 in.). The number of DCP blows is measured for the penetration between 152.4 and 457.2 mm (6 and 18 in.) into the granular material. This should be compared with the minimum required DCP blow count determined using Eq. 66 or Table 26, based on the optimum moisture content of the tested granular material. It is worth noting that the optimum moisture content is used as an index of the amount of fine particles contained in the tested granular material. FIGURE 92 DCP criteria for different types of soils (Indiana DOT 2013).

96 ∼ wcoptNDCP req 0 12 in. 59exp 0.12 (66))()( = − where wcopt = the optimum moisture content, and (NDCP)req|0~12 in. = the minimum required blow count for 0- to 12-in. penetration, which implies an RC of 95% with high probability. The minimum required blow count should be rounded up to the nearest integer. The Indiana DOT specifies that the moisture content of compacted clayey, silty, sandy, and granular soils should be maintained within -3% and +1% of the optimum moisture content. In addition, one moisture content test is required for each day that density or strength measurements are taken. The sample for moisture content is required to be representa- tive of the entire depth of the compaction lift being tested. LWD Specification For controlling the compaction of aggregate materials and chemically modified soils, the Indiana DOT specifies per- forming LWD tests at 609.6 mm (2 ft) from each edge of the construction area and at the midpoint of the site width. To accept the compacted layer, the average value of the maxi- mum deflection obtained in the three LWD tests is to be equal to or less than the maximum allowable deflection determined from a test section. In addition, the Indiana DOT requires that the moisture content of the aggregate be within -6% of the opti- mum moisture content. One moisture content test of the com- pacted aggregates is required per day. In addition, the Indiana DOT specifies that compaction control testing be done for each 800 tons of compacted aggregate and for each 1,400 yd3 of chemically modified soil. The agency provides two options for determining the max- imum allowable LWD deflection using a test section. In both options, the test section should have an area approximately 30 m (100 ft) long by the width of the layer to be constructed. In addition, the test section is to be constructed with the avail- able equipment of the contractor to determine the roller type, pattern, and number of passes for the maximum allowable deflection. In the first option, only LWD tests are conducted on the test section. In this case, the roller is operated in the vibratory mode, and 10 random LWD tests should be con- ducted at the approximate locations, as shown in Figure 93, after the completion of four and five roller passes on the test section. If the difference between the average values of the maximum deflections of the LWD tests conducted after the fourth and fifth pass is equal to or less than 0.01 mm, the com- paction will be considered to have peaked and the average of the 10 LWD values after the fifth pass will be used as the max- imum allowable LWD deflection. However, if the difference between the average deflection values of LWD tests is greater than 0.01 mm, an additional roller pass in the vibratory mode is applied and 10 LWD tests performed at the same locations. This procedure is continued until the difference of the aver- age maximum deflection values of the 10 LWD tests between consecutive roller passes is equal to or less than 0.01 mm. The maximum allowable deflection will be the lowest average maximum deflection value of the last conducted LWD tests. The Indiana DOT also provides another option for determin- ing the maximum allowable LWD deflection value. With this option, LWD and nuclear density gauge tests are concur- rently conducted on the test section. The maximum allowable deflection will be the average deflection value of the 10 LWD test values conducted once the compacted test section has met Indiana DOT density requirements. In a follow-up interview, the Indiana DOT staff stated that using control strips in their current specification for LWD had made the contractor more comfortable when using this new device. The Indiana DOT requires that the LWD used in compac- tion control tests have a metal loading plate with a diameter Optimum Moisture Content (wcopt) NDCP 0~12 in. 10 18 11 16 12 14 13 13 14 11 Source: Indiana DOT (2013). TABLE 26 (NDCP)req0~12 in. FOR TYPICAL OPTIMUM MOISTURE CONTENT VALUES FIGURE 93 Test location for LWD tests (Indiana DOT 2013).

97 of 300 mm (11.8 in.), an accelerometer attached to the cen- ter of the loading plate for measuring the maximum vertical deflection, and falling mass of 10 kg (22 lb). In addition, the agency specifies that the maximum force applied by the LWD should be 7.07 kN (1,589.4 lbf). The Indiana DOT’s specifications include guidelines for conducting the LWD test. According to those guidelines, the test area must be leveled so that the entire undersurface of the LWD load plate is in contact with the material being tested. Loose and protruding material is to be removed and, if required, any unevenness be filled with fine sand. The LWD load plate also should be rotated approximately 45° back and forth to seat it. The LWD test includes conducting six drops. The first three are considered seating drops, and the mea- surements of the last three are averaged and reported as the LWD deflection value. Additional compaction of the tested material is required if the change in deflection for any two consecutive LWD drops is 10% or greater. The Indiana DOT also specifies that the LWD plate not move laterally with suc- cessive drops. Indiana Experience Indiana DOT staff stated in a follow-up interview that the DCP and LWD specifications are currently being used for compaction control of unbound materials in at least 80% of their construction projects. Indiana DOT has found that stiff- ness- and strength-based compaction control specifications allow for the linking of construction and design processes. This has helped to provide more accurate design input values. For example, for lime-treated soils the modulus design input value was increased by 25% based on the results of a DCP test conducted during and after construction of lime-treated soil layers. The thickness of lime-treated layers was also reduced from 16 to 14 in., which resulted in huge savings. So far, Indiana DOT staff stated, the DOT has had success with the DCP and LWD for most projects and will continue using them. Indiana DOT will be selling 90% of its nuclear density devices this year. According to the Indiana DOT website, the projected annual savings for using the DCP, rather than nuclear testing equipment, as a tool for soil compaction qual- ity control is $480,000 annually. Indiana DOT noted that the main limitation in the imple- mentation LWD specifications was the difficulty of using the LWD in confined areas and small projects for which con- trol strips cannot be constructed. In addition, the DCP posed problems for use with sand. However, the agency is working on solving these issues. Missouri DOT DCP Specification The Missouri DOT has recently implemented a compaction control specification for Type 7 aggregate base material under both roadways and shoulders that is based on DCP test results. In this specification, Missouri DOT requires that Type 7 aggregate base material be compacted to achieve an average DPI value through the base lift thickness less than or equal to 0.4 in. (10 mm) per blow. DCP tests should be conducted within 24 h of placement and final compaction, and the device used have an 8-kg (17.6-lb) hammer and meet the require- ments of ASTM D6951. DPI values are determined using Eq. 63, which was proposed by the Minnesota DOT. According to Missouri’s specification, the moisture content of the Type 7 base material is not to be less than 5% during compaction. In a follow-up interview, the Missouri DOT staff stated that the main reason for developing this specification was that the Proctor test cannot be conducted on the Type 7 base because it contains large amounts of aggregates with sizes greater than 19 mm (¾ in.). In addition, Missouri DOT staff indicated that the 0.4 in./blow DCP limiting criterion corresponds to a CBR value of 10. Illinois DOT DCP Specifications The Illinois DOT uses the DCP to assess the stability of sub- grade soil before and during construction activities. The DCP test is conducted to ensure that the subgrade provides adequate support for the placement and compaction of pavement lay- ers and will not develop excessive rutting and shoving during and after construction. The Illinois DOT acceptance criterion for subgrade is based on the immediate bearing value (IBV), which is computed from the DPI using Eq. 67. Illinois DOT requires a minimum subgrade IBV of 6% to 8% for construc- tion activities. For values less than 6%, subgrade treatment before construction is required. The IBV values from DCP testing are also used as subgrade inputs in most Illinois DOT flexible pavement design procedures for local roads. 10 (67)0.84 1.26XLOG PRIBV = [ ]( )− Iowa DOT Quality Management-Earthwork Specifications Owing to concerns about the quality compaction of roadway embankments that caused premature high roughness pave- ments in the state of Iowa, a study was initiated by the Iowa DOT to address such problems and improve construction practices for roadway embankments. The study consisted of a four-phase research project conducted from 1997 to 2007. Based on the findings of the study’s first phase, it was con- cluded that current Iowa DOT specifications failed to consis- tently produce quality embankments. The compaction quality was monitored using standard Proctor testing, which was found to be inadequate because it tends to overestimate the degree of compaction. Subsequent phases developed and evaluated a new specification for quality management- earthwork (QM-E).

98 The QM-E specification includes five key evaluation crite- ria for compaction control of soil embankment layers: mois- ture content, density, lift thickness, stability, and uniformity. The tests to assess the moisture content, density, and lift thickness are required for every 500 m3 of fill. The stability and uniformity criteria of the compacted materials are evalu- ated using DCP tests, which are conducted to a depth of 1 m (39.4 in.) for every 1,000 m3 (1,308 yd3) of compacted fill. To ensure stability, Iowa DOT specifies that the four-point mov- ing average of the mean DPI of the compacted soil does not exceed the maximum value provided in Table 27 for the type and grade of borrow material. In addition, for uniformity of compaction, the mean change in the DPI should be less than the maximum values shown in Table 28. As part of the QM-E provisional specification, control strips are to be constructed to establish proper rolling patterns and lift thickness. Four random locations within each test area are tested for thickness, moisture content, density, mean DPI, and mean change in DPI of the compacted lift. Density, mean DPI, and mean change in DPI are recorded as a function of roller passes. The number of roller passes and the lift thick- ness are adjusted until the minimum test criteria are met. Europe’s LWD Specification In 2008, the European Union developed a standard for using the LWD in evaluating compacted layers of unbound materi- als (CEN ICS 93.020: Measuring Method for Dynamic Com- pactness & Bearing Capacity with Small-Plate Light Falling Weight Deflectometer). This standard specifies performing LWD tests in accordance with the technical requirements and specifications shown in Table 29. The test process involves using the LWD with a loading plate diameter of 163 mm, a falling weight of 10 kg (22 lb), and a drop height of 720 mm (28.3 in.), generating a load of 7 kN (1,589.4 lbf) for testing unbound materials. The LWD is used to obtain two main parameters: the dynamic modulus and the dynamic compact- ness rate. Although the dynamic modulus is used to assess the bearing capacity of the tested unbound material, the dynamic compactness rate is used to evaluate the quality of compac- tion. The testing process to obtain those two parameters involves performing six sequences consisting of three LWD drops (for a total of 18 drops) on the loose, noncompacted material at the site. From the second measuring sequence, the average deformation of the three LWD drops is used to deter- mine the initial dynamic modulus. From the sixth measur- ing sequence, the final modulus is obtained. In both cases, the dynamic modulus is computed based on the elastic Boussinesq method using the following equation: E c v p r s d dyn a = −( ) ×1 682 1 ( ) where Ed = LWD dynamic modulus; c = Boussinesq plate multiplier (considering p/2 rigid plate); s1a = average vertical travel of the center of the plate, s s s s a = + + 3 ;1 11 12 13 v = Poisson ratio; r = radius of the loading disc (mm); pdyn = theoretical pressure applied to, computed using as Fdyn/A; A = loading plate area (mm2); Fdyn = applied load, F m g h Kdyn i i i i= 2 ; Soil Performance Classification Maximum Mean DCP Index (mm/blow) Cohesive Select 75 Suitable 85 Unsuitable 95 Intergrade Suitable 45 Cohesionless Select 35 Source: White et al. (2007a). TABLE 27 REQUIREMENTS FOR MEAN DCP INDEX INDICATING STABILITY Soil Performance Classification Maximum Mean Change in DCP Index (mm/blow) Cohesive Select 35 Suitable 40 Unsuitable 40 Intergrade Suitable 45 Cohesionless Select 35 Source: White et al. (2007a). TABLE 28 REQUIREMENTS FOR MEAN CHANGE IN DCP INDEX INDICATING UNIFORMITY

99 where TrE = site relative compaction at a given water content; and Trw = the moisture correction coefficient to adjust for dif- ferences between the measured moisture content and optimum moisture content. This is determined based on the results of a modified Proctor test conducted on samples of the material used in the field using the following equation: (71) max Trw di d = ρ ρ where rd max = maximum dry density value obtained in the modi- fied Proctor test; and rdi = dry density value on compaction curve of the modi- fied Proctor tests corresponding to the in situ mois- ture content. United Kingdom Specifications The United Kingdom specifications (Highway Agency 2009) define four classes of foundation material (base course materi- als) based on the long-term, in-service foundation surface modulus value (a composite value with contributions from all underlying layers). Table 30 presents the four foundation classes and their corresponding surface moduli. Those mod- ulus values are then used in design. A minimum subgrade modulus of 30 MPa (2.5 CBR) also is specified. If a subgrade modulus is less than 30 MPa, it should be stabilized or treated before being included in the permanent pavement works. For construction quality control, a target mean and minimum sur- m = mass of falling weight; g = acceleration of gravity (9.81 m/s2); h = drop height (m); and K = spring constant (N/m). According to the CEN ICS 93.020 standard, the work imparted on the material during the six LWD sequences (the 18 LWD drops) is equivalent to that applied in the modified Proctor test (see Table 29). Thus, a compaction curve can be generated based on the average deflection value obtained in each of the six LWD sequences. The rela- tive compactness rate (TrE) at the field moisture content is determined from this compaction curve using the follow- ing equation: % 100 (69)0T DrE m= − Φ ∗ where F0 = a linear coefficient of the calculated from the Proc- tor test results (in general, it is taken to be 0.365 ± 0.025); and Dm = deformation index, it is calculated from the sum of the elements of the data line formed from the differ- ence of the subsequent deflections up to the drop. This relative compactness rate (determined based on the soil’s moisture content at the site) must be adjusted to the opti- mal moisture content using Eq. 70. This adjustment is needed to make the relative compactness rate in the field equivalent to the relative compaction ratio determined based on the maximum dry density obtained in a modified Proctor test. The adjusted value is referred to as the dynamic compactness rate, Trd. (70)iT T Trd rE rw= • Mass of the falling weight (including handle) 10.5 ± 0.5 kg • Total mass of guide rod (including the spring consisting of spring elements, transportation protection of the falling weight, triggering structure and tilting protection) Maximum 5 ± 0.5 kg • Dynamic loading 0.35 MPa • Loading time 18 ± 2 ms Design requirements of the loading plate: • Diameter of the loading plate 163 ± 2 mm • Thickness of the loading plate Minimum 20 mm • Total mass of the loading plate complete masse (including measuring cell for the sensor and handles) 15 ± 1.0 kg Fixed technical data of acceleration gauge applied for deformation measurement: • Measurement range of in-built acceleration gauge 0–50 g In case of applying other strain gauge and the acceleration gauge: Measurement time 18 ± 2 ms minimum signals/18 ms • Processed measurement signal Minimum 0.01 mm • Reading accuracy of deformation Maximum ± 1.5 s per day • Quartz clock accuracy • Reading accuracy of deformation Minimum 0.01 mm Source: CEN ICS 93.020. TABLE 29 INFORMATIVE REQUIREMENTS OF THE LWD TEST

100 correction factor based on Table 32. The correction factor needs to be added to the LWD moduli measured in the field to achieve an equivalent LWD modulus at the optimum mois- ture content. In this procedure, the authors required using a LWD with the 850-mm (33.5-in.) drop height, 20-kg (44-lb) drop weight, and 300-mm (12-in.) diameter loading plate. In addition, they recommended conducting six LWD drops at each test loca- tion, with the first drop discarded and the measurements of the remaining five drops averaged to determine the modulus value at that test location. Three locations should be tested within a 3-m (10-ft) diameter area and the average used to determine a representative value for that particular station. Louisiana Study Based on a field study in Louisiana, Abu-Farsakh et al. (2005) recommended using a limiting DCP penetration ratio of 5.5 mm/blow for compacted base course layers. This was the average DCP penetration ratio measured for all base course test sections that had satisfactory FWD and PLT stiffness moduli and acceptable compaction levels. The DPI value of 5.5 mm/blow corresponds to a CBR value of 87%, according to Eq. 72. This value was found to meet the minimum CBR value of 80% specified by the U.S. Army Corps of Engineers for stone bases in airfield flexible pavements. face modulus value is specified for each type of pavement layer, as shown in Table 30. The moving mean of five consecu- tive in situ foundation surface modulus measurements should be equal to or greater than the target mean foundation surface modulus in the table. In addition, all individual in situ founda- tion surface moduli should exceed the target minimum foun- dation surface before construction of the overlying pavement layers. According to UK specifications, the in situ foundation surface modulus should be measured using the standard FWD. The LWD can be used if a correlation between LWD and FWD measurements is developed by conducting both tests at 25 points within the demonstration site. SYNTHESIS OF PAST AND ONGOING STUDIES IN OTHER STATES New England Transportation Consortium Study In a study funded by the New England Transportation Con- sortium, Steinart et al. (2005) proposed a procedure that used the LWD to assess the compaction of granular base course layers. In this procedure, a target modulus at the optimum moisture content is selected for a base course material from Table 31, based on the required relative compaction value. LWD tests should be conducted on the compacted base course materials to ensure that a target modulus value is achieved. In addition, moisture content is obtained to determine the Surface Modulus (MPa) Class 1 Class 2 Class 3 Class 4 Long-term in-service surface modulus 50 100 200 400 Mean foundation surface modulus Unbound mixture types 40 80 * * Fast-setting mixture types 50 100 300 600 Slow-setting mixture types 40 80 150 300 Minimum foundation surface modulus Unbound mixture types 25 50 * * Fast-setting mixture types 25 50 150 300 Slow-setting mixture types 25 50 75 150 *Unbound materials are unlikely to achieve the requirements for Class 3 and 4. Source: Highway Agency (2009). TABLE 30 TOP OF FOUNDATION SURFACE MODULUS REQUIREMENTS Relative Compaction Based on AASHTO T180 (%) Equivalent LWD Composite Modulus (MPa) at Optimum Water Content 90 92 95 115 98 130 100 139 Source: Steinart et al. (2005). TABLE 31 TENTATIVE EQUIVALENCES BETWEEN PERCENT COMPACTION AND COMPOSITE MODULUS AT OPTIMUM WATER CONTENT FOR BASE AND SUBBASE COURSE AGGREGATE

101 consists of selecting the most suitable material to ensure a durable layer. In the third step, the selected material for each layer should be tested in the laboratory at the field compaction and moisture conditions to obtain a representa- tive design modulus. The fourth step involves establishing a target modulus value that will be used in the field based on laboratory tests performed in conjunction with determin- ing the representative design modulus. Alternatively, this value is set based on field test strips. In the fifth step, field moduli are measured during construction with an appropriate device to ensure that the target modulus value is achieved. The final step consists of developing statistical control charts to ensure that the modulus and its variability along the proj- ect are in control. NCHRP 10-84 Study Currently, the ongoing NCHRP project 10-84 (2011) is antic- ipated to develop a modulus-based construction specifica- tion for compaction of unbound materials. Based on the results of Phase I of this project, the researchers indicated that the modulus-based compaction control specification that will be developed in this study ideally may follow the flowchart shown in Figure 94. As this figure demonstrates, several interrelated steps should be included in such speci- fication. In the first step, a mechanistic-empirical design procedure such as the Mechanistic–Empirical Pavement Design Guide (MEPDG) should be selected and construc- tion specifications tied to the procedure. The second step FIGURE 94 An ideal flowchart for modulus-based specifications (NCHRP 10-84, 2011). Water Content Relative to Optimum Correction Factor to Be Added to Composite Modulus (MPa) Measured at Field Moisture Content Dry of OMC -4% -31 -3% -23 -2% -15 -1% -8 At OMC 0 Wet of OMC +1% 8 +2% 15 +3% 23 +4% 31 Source: Steinart et al. (2005). TABLE 32 FACTOR TO CORRECT COMPOSITE MODULUS MEASURED AT FIELD WATER CONTENT TO EQUIVALENT VALUE AT OPTIMUM WATER CONTENT

102 Minnesota IC Specification A pilot specification was implemented by the Minnesota DOT in 2006 during the construction of TH 64 in Akeley, Minnesota, and was later updated in 2010 and 2012. In this specification, all segments of projects in which IC rollers are used should be compacted so that at least 90% of the IC measurements are equal to or exceed 90% of the target ICMV before the next lift is placed. If localized areas have IC measurements less than 80% of the target ICMV, those areas should be recompacted. In addition, the target ICMV should be reevaluated if a sig- nificant portion of the project is more than 30% of it. The Minnesota DOT requires that moisture content of compacted material be between 65% and 100% of its optimum moisture content value. The Minnesota DOT also specifies determining the target ICMV using control strips at least 100 m (300 ft) by 10 m (32 ft) at their base, with a thickness equal to that of the layer to be constructed. One control strip is required for each differ- ent type/source of grading material used on the construction site. As shown in Table 35, smooth drum or padfoot vibratory IC rollers weighing at least 25,000 lb can be used. Texas IC Specification In 2008, the Texas DOT developed a special specification for quality compaction of subgrade soils and base course layers using IC rollers. A list of Texas DOT-approved IC rollers was released in 2009 (see ftp://ftp.dot.state.tx.us/pub/txdot-info/ cmd/mpl/ic_rollers.pdf). In this specification, a control strip of at least 500 ft in length and a width equal to that of the material course is compacted using the same IC roller proce- dures intended for the remainder of the project. The moisture content during compaction must be no less than 1% below the optimum moisture content. Control strip results are used to determine the IC compaction parameters and the level of com- paction necessary to achieve the maximum target dry density. It is worth noting that as part of this specification, seismic modulus should be determined at the same locations on respective mea- surement passes as those used for density measurements. The acceptance of compacted sections is based on density and mois- ture content measurements, which should be obtained within 24 h of compaction completion. A section is deemed acceptable if a maximum of one of the five density measurements taken falls no more than 3 pcf below the target density value. SUMMARY This chapter reviews practices and experiences regarding the implementation of stiffness/strength-based specifications for compaction control of unbound materials in Europe and the United States. Although several states have evaluated the use of in situ device(s) to assess stiffness/strength for compaction control of unbound materials, only the DOTs of Indiana and Minnesota have developed and implemented comprehensive REVIEW OF CURRENT INTELLIGENT COMPACTION SPECIFICATIONS During the past two decades, several European countries have developed specifications for intelligent compaction. Recently, a few state DOTs, including those of Indiana, Minnesota, and Texas, have developed specifications to facilitate the imple- mentation of IC into earthwork construction practices. In addi- tion, the FHWA in July 2011 released generic specifications for compaction of unbound materials using IC technologies. These generic specifications are to be modified by individual agencies to meet specific requirements (see www.intelligent compaction.com). More recently, two approaches for com- paction control of unbound materials that included using IC technologies were proposed as part of TRB’s second Strate- gic Highway Research Program (SHRP 2) Renewal Project R07 (Scott et al. 2013). Tables 33 and 34 provide a summary of key elements of the current IC specifications in the United States and other countries, respectively. Two main approaches were followed in these specifications. In the first approach the ICMV measurements obtained in the final pass of the IC roller are used to map the weak areas in the compacted layers. Acceptance of these layers depends on satisfying a minimum density or stiffness target value in these weak areas based on the results of spot in situ tests, such as the NDG, PLT, or LWD. Another approach that has been pursued in the specifications developed by state DOTs and the FHWA involves evaluat- ing the ICMV measurements between successive passes until the target ICMV is achieved in a minimum percentage of the compacted area (typically 80% to 90%). The target ICMV is determined based on calibration tests on control strips selected at the construction site. A review of the most recent IC specifi- cations in the Indiana, Minnesota, and Texas DOTs is provided in the following sections. Indiana IC Specification The Indiana DOT recently developed a specification for IC construction. This specification can be used only when the construction area evaluated is equal to or greater than 5,000 ft2 (500 m2). In this specification, at least 90% of the construc- tion area is to be mapped with an approved IC roller, and a minimum of 70% of the mapped construction area should have or exceed the target ICMV. Deficient areas that do not meet the ICMV target and are larger than 1,500 ft2 should be reworked and retested. The reworked areas will be accepted if the ICMVs meet the minimum target ICMV. The Indiana DOT specifies using the IC roller on selected test sections to establish the target ICMV that corresponds to DCP test results. Test sections are to be approximately 100 ft long and 20 ft wide. Moisture tests at two random locations and DCP tests at four random locations in each test section should be performed. The moisture content should be within -3% to +2% of the optimum moisture content for silt-clay soils and controlled within -6% of the optimum moisture content for granular materials.

103 Agency Equipment Field Size Location Specs Documentation Compaction Specs Speed Frequency FHWA (2012) Vibratory self- propelled single- drum roller 225 ft (75 m) long and 24 ft (8 m) wide Results from the moisture, strength, and maximum dry density, and optimum moisture content tests; IC roller data (manufacturer, model, type, positioning); analysis of IC roller data for coverage area and uniformity; limit of construction area At least 90% of the individual construction area shall meet the optimal number of roller passes and 70% of the target ICMV determined from the test sections. Rework and reevaluate if areas do not meet the IC criteria. Constant speed and frequency throughout a section Indiana (2012) Self-propelled vibratory roller with drum accelerometers and smooth or padfooted drums Approximately 100 ft long and 20 ft wide One calibration/ control strip per type of unbound material Results from the moisture, strength, and maximum dry density, and optimum moisture content tests; IC roller data (manufacturer, model, type, positioning); analysis of IC roller data for coverage area and uniformity; limit of construction area At least 90% of the construction area shall be mapped with the IC roller. The percentage of the mapped area that equals or exceeds the target ICMV shall be at least 70%. The reworked deficient areas will be accepted if the ICMVs meet a minimum of 100% of the target ICMV. Moisture and density: one test for each 1,400 yd3 of each lift. Minnesota DOT (2007) Smooth drum or padfoot vibratory roller (25,000 lb) 300 ft by 32 ft (minimum at base). Maximum 4 ft thick One calibration/ control strip per type or source of grading material Compaction, stiffness, moisture, QC activities, and corrective actions (weekly report) 90% of the stiffness measurements must be at 90% of the compaction target value. Same during calibration and production compaction Texas (2008) Self-propelled IC rollers equipped with a measurement and documentation system At least 500 ft in length, and width must be equal to the full width of the material course Uniform layer, free of loose or segregated materials Roller speed, frequency, amplitude, roller measurement values (RMV); dry density, moisture content and seismic modulus of soil Accept if one of the five most recent density values is below the target density and the failing test is no more than 3 pcf below the target density. Rework, recompact, and refinish material that fails to meet the criteria. After Chang et al. (2010). TABLE 33 SUMMARY OF IC SPECIFICATIONS IN THE UNITED STATES

104 TABLE 34 SUMMARY OF INTERNATIONAL IC SPECIFICATIONS Agency Equipment Field Size Location Specs Documentation Compaction Specs Speed Frequency ISSMGE Roller chosen by experience 100 m by the width of the site Homogenous, even surface. Track overlap 10% drum width. Rolling pattern, sequence of compaction and measuring passes; amplitude, speed, dynamic measuring values, frequency, jump operation, and corresponding locations Correlation coefficient 0.7. Minimum value 95% of Ev1, and mean should be 105% (or 100% during jump mode). Dynamic measuring values should be lower than the specified minimum for 10% of the track. Measured minimum should be 80% of the specified minimum. Standard deviation (of the mean) must be 20% in one pass. Constant 2–6 km/h (± 0.2 km/h) Constant 2–6 km/h (± 0.2 km/h) Constant (± 2 Hz) Earthworks (Austria) Vibrating roller compactors with rubber wheels and smooth drums suggested 100 m long by the width of the site No inhomogeneitie s close to surface (materials or water content). Track overlap 10% drum width. Compaction run plan, sequence of compaction and measurement runs, velocity, amplitude, frequency, speed, dynamic measuring values, jump operation, and corresponding locations Correlation coefficient 0.7. Minimum value 95% of Ev1, and median should be 105% (or 100% during jump mode). Dynamic measuring values should be lower than the specified minimum for 10% of the track. Measured minimum should be 80% of the set minimum. Measured maximum in a run cannot exceed the set maximum (150% of the determined minimum). Standard deviation (of the median) must be 20% in one pass. Constant (± 2 Hz) Research Society for Road and Traffic (Germany) Self- propelled rollers with rubber tire drive are preferred; towed vibratory rollers with towing vehicle are suitable Each calibration area must cover at least three partial fields ~20 m long Level and free of puddles. Similar soil type, water content, layer thickness, and bearing capacity of support layers. Track overlap 10% machine width. Dynamic measuring value; frequency; speed; jump operation; amplitude; distance; time of measurement; roller type; soil type; water content; layer thickness; date, time, file name, or registration number; weather conditions; position of test tracks and rolling direction; absolute height or application position; local conditions and embankments in marginal areas; machine parameters; and perceived deviations The correlation coefficient resulting from a regression analysis must be 0.7. Individual area units (the width of the roller drum) must have a dynamic measuring value within 10% of adjacent area to be suitable for calibration. Constant Vägverket (Sweden) Vibratory or oscillating single-drum roller; minimum linear load 15–30 kN; roller- mounted compaction meter optional Thickness of largest layer 0.2– 0.6 m Layer shall be homogenous and nonfrozen. Protective layers < 0.5 m may be compacted with subbase. Bearing capacity or degree of compaction requirements may be met. Mean of compaction values for two inspection points 89% for sub-base under road base and for protective layers over 0.5 m thick; mean should be 90% for road bases. Required mean for two bearing capacity ratios varies depending on layer type. Constant 2.5–4.0 km/h After Chang et al. (2010).

105 Minnesota, and Texas), also has developed IC specifications. These specifications include selecting target ICMVs based on acceptable stiffness or density spot-testing measurements obtained on control strips compacted using IC rollers. Accep- tance is based on achieving the target ICMV for a minimum percentage of compacted area (range, 80% to 90%). Another type of IC specification that has been adopted by some Euro- pean countries, including Sweden, consists of using ICMV measurements to identify weak areas at a project site. These areas should be assessed using in situ point tests (e.g., den- sity, plate load, and/or LWD) for acceptance. stiffness- and strength-based compaction control specifica- tions for various types of unbound materials. Both DOTs use the DCP and LWD in their specifications. Other states, such as Missouri, have used the DCP in compaction control but only for a specific type of unbound material. Staff of the Indi- ana and Minnesota DOTs reported that they have had good experiences with the DCP in compaction control of unbound materials. Although the Minnesota DOT had a mixed experi- ence with the LWD, the staff of the Indiana DOT noted its success with the device in compaction control of unbound materials. The FHWA, along with three state DOTs (Indiana, State Non-Nuclear Density Devices Devices for In Situ Stiffness/Strength Measurements Alaska Alabama Arkansas Arizona California Colorado CH, DCP, LWD Connecticut Delaware Delaware DCP, GeoGauge, LWD Florida EDG, MDI CH, DCP, GeoGauge, LWD, PSPA Georgia DCP Hawaii EDG DCP, GeoGauge, LWD Iowa DCP, LWD Idaho EDG, SDG GeoGauge Illinois MDI DCP, GeoGauge Indiana MDI CH, DCP, LWD Kansas GeoGauge Kentucky Louisiana EDG DCP, GeoGauge, LWD Massachusetts Maryland DCP, GeoGauge, LWD Maine DCP, LWD Michigan Minnesota EDG, MDI, SDG CH, DCP, GeoGauge, LWD Missouri EDG DCP Mississippi DCP, GeoGauge, LWD Montana DCP, GeoGauge North Carolina DCP, GeoGauge, LWD North Dakota DCP, LWD Nebraska EDG, MDI, SDG DCP, GeoGauge, LWD New Hampshire EDG GeoGauge New Jersey EDG, MDI New Mexico MDI CH, DCP, GeoGauge Nevada New York EDG, SDG DCP, GeoGauge, LWD Ohio Oklahoma DCP, LWD Oregon Pennsylvania GeoGauge Rhode Island South Carolina GeoGauge South Dakota GeoGauge Tennessee Texas MDI, SDG CH, DCP, GeoGauge, LWD, PSPA Utah DCP Virginia DCP, GeoGauge Vermont EDG, MDI Washington Wisconsin DCP, GeoGauge West Virginia GeoGauge Wyoming GeoGauge TABLE 35 IN SITU TEST DEVICES EVALUATED BY STATE DOTs