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.

106 chapter six CONCLUSIONS INTRODUCTION For this report, all available knowledge and information from a variety of sources on various non-nuclear devices and meth- ods that have been used for compaction control of unbound materials were compiled and summarized. Included were non- nuclear devices that measure density, as well as those that eval- uate in situ stiffness- and strength-related properties of unbound materials. Information on the devices’ accuracy, repeatability, ease of use, test time, cost, Global Positioning System (GPS) compatibility, calibration, compatibility with various unbound materials, and depth of influence was documented and dis- cussed. In addition, the main advantages, disadvantages, and limitations of the devices were identified. All correlations between the measurements of the considered devices and den- sity, as well as the input parameters for designing transpor- tation and geotechnical structures, were provided. This report reviewed stiffness- and strength-based specifications that have been developed and implemented in the United States and in Europe for compaction control of unbound materials. The main findings and conclusions are provided in the sections that follow. In addition, gaps in knowledge and current practices, along with recommendations for future research to address these gaps, are highlighted. MAIN FINDINGS Density-Based Compaction Control Specification The majority of state departments of transportation (DOTs) use field density and moisture content measurements obtained by the nuclear density gauge (NDG) for compaction control of various types of unbound materials. Furthermore, almost all DOTs use impact compaction methods, such as AASHTO T99 and AASHTO T180, to determine the target density value to be achieved in the field. However, those methods are lim- ited for unbound materials that have 30% or less by mass of their particles with sizes greater than 19 mm (¾ in.). A review of states’ construction manuals indicated that there are dif- ferences in the relative compaction values required by each DOT for unbound materials in embankments, subgrade soils, and base course layers. Non-Nuclear Devices for Measuring Density and Moisture Content The majority of DOTs expressed interest in having non-nuclear density devices that could replace the NDG. As shown in Table 35, several of them have evaluated such devices. Sur- vey results showed that among the DOTs that evaluated non- nuclear density devices, overall satisfaction was so low that none of them recommended the use of such devices. As shown in Table 36, which presents a comparison between the non-nuclear density devices and the NDG based on the results of previous reported studies and the survey conducted in this synthesis, all currently available non-nuclear density devices are more difficult to operate and require longer testing time than does the NDG. In addition, the electrical density gauge (EDG) and moisture density indicator (MDI) were reported to have some limitations when used for testing high-plasticity clay and stiff soils. There are several non-nuclear devices that have been devel- oped to measure the moisture content; however, limited studies have been conducted to evaluate most of these devices. The speedy moisture tester and field microwave are the most com- mon non-nuclear devices used to measure in situ moisture con- tent of unbound materials. According to the survey conducted in this study, 13 state DOTs recommend their use but four do not. The main limitation of both devices is that they cannot be used for all types of unbound materials. According to previous stud- ies documented in this report, the speedy moisture tester cannot be used for highly plastic clayey soil or coarse-grained granular soil. Furthermore, the field microwave is suitable only for materials consisting of particles smaller than 4.75 mm (0.19 in.). In Situ Test Devices for Stiffness/Strength Measurements As shown in Table 37, the dynamic cone penetrometer (DCP), GeoGauge, and light weight deflectometer (LWD) are the devices most thoroughly evaluated by DOTs among all in situ test devices. The DCP and LWD have been implemented by some DOTs in the field for compaction control of unbound materials. Previous studies found the GeoGauge measure- ment to be very sensitive to the seating procedure and to the stiffness of the top 2 in. of the tested soil layer, which signifi- cantly affected its reliability. Table 38 presents a comparison between the different in situ test devices that have been used for compaction control of unbound materials, based on the information collected in this synthesis. All of the devices are quick and easy to use. Although the devices’ measurements were found to be influenced by the moisture content, none of these devices

107 has the ability to measure it. The Briaud compaction device (BCD), DCP, LWD, and soil compaction supervisor (SCS) may not be suitable for very soft, fine-grained soils. Accord- ing to ASTM D6951, the DCP is also limited to use with materials that have a maximum particle size smaller than 50 mm (2 in.). The influence depth differs among the various in situ devices. Some devices, such as the BCD, have shallow depths that may not allow them to assess the properties of the entire lift; this creates a problem for their usage in compaction control procedures. On the other hand, careful consideration should be given when analyzing the results for relatively thin lifts because the zone of influence of some devices might exceed the lift thickness, thus providing a composite value of two layers, rather than only the tested layer. In addition, devices apply different load magnitudes during the test, so the measurement will be different, and various devices apply different load magnitudes during testing, which affects mea- surement. Thus, measurement must be corrected to account for design loads. Most devices were reported to have limita- tions with regard to the type of unbound material they can test such as that there is not one single in situ test device that can assess all types of unbound materials. Finally, all of the devices are comparable in price with the NDG, except the portable seismic property analyzer (PSPA), which is much more expensive than the others. As shown in Table 36, several correlations between the in situ test device measurements and those obtained by other standard in situ tests as well as design input parameters [e.g., Mr, California bearing ratio (CBR)] were reported in the lit- erature. However, those correlations are empirical and thus can be used only in conditions similar to those encountered during their development. In general, no strong correlation was found between in situ stiffness/strength measurements and in-place density because this relationship continuously changes with moisture content. All devices except the PSPA might have difficulties in establishing target field value in the laboratory owing to boundary effects on their measurement accuracy. Therefore, several DOTs attempted to establish those values based on pilot projects or by constructing control strips along a proj- ect. According to an interview with Indiana DOT staff, the use of control strips to develop target values of new in situ devices for compaction control facilitates the implementa- tion of these devices in field projects because it provides the opportunity to identify their limitations. It was also found to help familiarize contractors with device procedures and measurements. Stiffness- and Strength-Based Compaction Control Specifications The majority of DOTs are interested in implementing stiffness- and strength-based specifications for compaction control of unbound materials, but few DOTs have developed such spec- ifications. This was mainly attributed to the lack of trained personnel and funds, the need for new testing equipment, and Feature NDG MDI EDG SDG Test Method Nuclear Electrical Electrical Electrical ASTM standard D2922, D3017 D6780 D7698 None Measurement d, w d, w d, w d, w Moisture readings Yes Yes Yes Yes Calibration of device Requires calibration Laboratory testing in Proctor mold Field calibration using direct measurement of d, w Field calibration using direct measurement of d, w Portability Good Medium Medium Good Durability Good Good Good Good Operator skill Extensive, licensedtechnician Moderate Moderate Extensive Ease of use— training Medium—requires training Difficult Difficult Difficult Initial cost About $6,950 $6,000 $9,300 $10,000 Data storage Yes Yes Yes Yes Repeatability Good Good Mixed Results * Accuracy Good Mixed Results Mixed Results * GPS Yes No Yes Yes Main limitations - Contains radioactive materials that can be hazardous - Requires intense regulations - High costs to own and maintain - Complex and time consuming - Cannot test highly plastic clay - Complex and time consuming - NDG is required for calibration - Cannot test highly plastic clay - Extensive operator training *Not enough data were reported. TABLE 36 COMPARISON BETWEEN DIFFERENT DEVICES FOR IN-PLACE DENSITY MEASUREMENT

108 Features BCD CH DCP GeoGauge LWD PSPA SCS ASTM standard None D5874 D6951 ASTM 6758 E2583 None None Measurement Modulus CIV DPI Modulus Modulus Modulus Compaction indicator Moisture measurement No Yes No No No No No Calibration of device UC test and BCD test on rubber blocks Laboratory testing in Proctor mold None Calibration plate Required Laboratory stiffness testing Preset system Portability Good Medium Good Good Medium Medium Good Durability N/A Poor Good Good Good Good Box—good Sensors—fair Ease of use/training Easy—minimal Easy—minimal Easy—minimal Easy—minimal Moderate Moderate Easy—minimal Initial cost Not available $2,500 $1,000 $5,000–$5,500 $8,000–$15,000 $30,000 About $1,650 Data storage Yes Yes No Yes Yes Yes Yes Influence depth (inch) About 6 10–12 48 5–8 11 (1–1.5 D)a 30 (maximum) Repeatability Goodb Medium Good Fair Fair Goodb Goodb GPS No Yes No Yes No No No Main strengths - Simple, very quick - Can be used in the lab to determine the target modulus - Operated by one person - Simple, quick - Strong correlations with CBR - GPS - Simple, quick for shallow depth - Economical - Assess up to 4-ft- thick layers - Strong correlation with CBR and Mr - Used in many DOTs - Simple, quick and nonintrusive - Good portability and durability - Quick - Measure a wide range of modulus values - Not influenced by aggregate size - Can be compared with lab measurement - Measure properties of multiple layers separately - Not influenced by aggregate size - Economical - Requires minimum training Main limitations - Not evaluated by DOTs yet - Cannot be used for very stiff or soft soil - Shallow influence depth - Boundary effects during calibration - Different CIV for CH models - May require two persons - Maximum allowed particle size is 2 in. - Deeper testing can take as long as 15 min/location - Extremely sensitive to seating conditions - Inconsistencies in testing data - Unfavorable findings by several DOTs - High variability in weak soft soils - May require two persons - Can be time consuming and can require complex data processing - No ASTM procedure - May be affected by the surrounding geometry - Expensive - Not evaluated by DOTs yet - Fair durability of sensors - Does not provide any test results applicable to design or quality control purposes a D is the diameter of loading plate. b Based on limited data collected. N/A = No available information. TABLE 37 COMPARISON BETWEEN DIFFERENT IN SITU SPOT DEVICES MEASURING STIFFNESS/STRENGTH

109 Parameter Device Correlations Reference Soil Type Clegg Hammer CBR CBR = 0.07 (CIV)2 Clegg (1980) Wide range of soils CBR = 0.8610 (CIV)1.1360 (R2 = 0.757) Al-Amoudi et al. (2002) GM soil CBR = 1.3577 (CIV)1.0105 (R2 = 0.845) SM soil CBR = 1.3489 (CIV)1.0115 (R2 = 0.846) GM and SM combined soils CBR = 0.513 (CIV)1.417 (R 2 = 0.94) Aiban and Aurifullah (2006) Steel slag and limestone aggregate base materials CBR = 0.564 (CIV)1.144 Fairbrother et al. (2010) Subgrade soils Dynamic Cone Penetrometer (DCP) Mr Mr (psi) = 7,013.065 – 2,040.783 ln(DPI) Hassan (1996) Fine-grained soil ( ) 1.925 0.144 7.82 r 27.86M dr c LLDPI w γ−= + (R2 =0.71) George and Uddin (2000) Fine-grained soil ( ) -0.305 -0.935 0.674 r 90.68 log M dr cr u DPI w c γ= + (R2 = 0.72) Coarse-grained soils ( )r 1.096 1045.9M DPI = Mohammad et al. (2009) A-4, A-6, A-7-5, and A-7-6 r 1.46 1.27 1 1M 3.86 2020.2 619.4 DPI w = + + CBR CBR = 2,559.44 / (-7.35 + DPI1.84) + 1.04 for 6.31 < DPI < 66.67 mm/blow (R2 = 0.93) Abu-Farsakh et al. (2004) Subgrade and base course materials Log CBR = 2.465 – 1.12 log (DPI) or CBR = 292/(DPI)1.12 Webster et al. (1992) Granular and cohesive materials Log CBR = 2.62 – 1.27 log (DPI) Smith and Pratt (1983) Log CBR = 2.56 – 1.15 log DPI Livneh and Ishai (1987, 1991) Fine-grained soil Log CBR = 2.2 – 0.71 (log DPI)1.5 Granular soils Log CBR = 2.56 – 1.16 log DPI for DPI > 10 Harrison (1989) Clayey soil Log CBR = 2.70 – 1.12 log DPI for of DPI < 10 (mm/blow) Granular soil E Log (E) = 3.05 – 1.07 log (DPI) De Beer (1990) Subgrade soil Log (E) = 3.25 - 0.89 log (DPI) Log (E) = 3.652 - 1.17 log (DPI) Pen (1990) Subgrade soil E = 2,224 (DPI)-0.99 Chai and Roslie (1998) Subgrade soil PLT Log (EPLT) = (-0.88405) Log (DPI) + 2.90625 Konard and Lachance (2000) Unbound aggregates and natural granular soils E PLT (i) = 1.6 9770 (DPI) - 36.9 -0.75 (3.27 < DPI < 66.67) (R2 = 0.67) Abu-Farsakh et al. (2004) Subgrade and base course materials E PLT (R2) = 1.4 4374.5 2.16 (DPI) 14.9 −− (3.27 < DPI < 66.67) (R2 = 0.72) FWD MFWD = 78.05 × (DPI)-0.67 Chen et al. (2007) Base soils MFWD = 338 (DPI)-0.39 for 10 mm/blow < DPI < 60 mm/blow and subgrade soils ln (EFWD) = 2.04 + 5.1873 ln(DPI) (3.27 < DPI < 66.67) (R2 = 0.91) Abu-Farsakh et al. (2004) Subgrade and base course materials TABLE 38 CORRELATION REPORTED IN LITERATURE FOR IN SITU DEVICES (continued on next page)

110 contractors’ unfamiliarity with stiffness- and strength-based specifications. Only the Indiana and Minnesota DOTs have widely implemented stiffness- and strength-based specifica- tions, and both use the DCP and LWD in those specifications. Both DOTs also reported that they had positive experiences with using the DCP as a tool for compaction control of unbound materials. Other states, such as Missouri, have used the DCP in compaction control but only for a specific type of granular base material. The Indiana and Minnesota DOTs use the DCP and LWD. Continuous Compaction Control and Intelligent Compaction Some studies documented in this synthesis have reported good correlations between intelligent compaction measurement values GeoGauge Mr 1.54 r G= M 46.48+0.01E (R2 = 0.59) 0.78 0.8 r G-M 13.94 + 0.0397 +E 601.08= w 1 (R2 = 0.72) Mohammad et al. (2009) Subgrade and base course materials CBR CBR = 0.00392 (EG) 2 -5.75 (R2 = 0.84) for 40.8 MPa < EG < 184.11 MPa Abu-Farsakh et al. (2002) Subgrade and base course materials PLT E PLT (i) = -75.58 + 1.52 (EG) (R2 = 0.87) for 40.8 MPa < EG < 194.4 MPa Abu-Farsakh et al. (2002) Subgrade and base course materials E PLT (R2) = -65.37 + 1.50 (EG) (R 2 = 0.90) for 40.8 MPa < EG < 194.4 MPa FWD MFWD = 37.65 HSG – 261.96 Chen et al. (2000) Base course materials MFWD = -20.07 + 1.17 (EG) (R2 = 0.81) for 40.8 MPa < EG < 194.4 MPa Abu-Farsakh et al. (2002) Subgrade and base course materials Light Weight Deflectometer (LWD) Mr 0.18 LWD=M 27.75× Er (R2 = 0.54) Mohammad et al. (2009) Subgrade and base course materials ( ) 111.23 242.32=M +12.64 0.2LWDE w+r (R2= 0.7) = –3.907 + 5.435 EFWD EFWD MR95 MR95 PI P200 D(f/95) – 0.370 M(f/o) (R2 = 0.70) George (2006) Subgrade soil = –2.30 + 3.860 D(f/95) – 0.316 M(f/o) – 0.635 (R2 = 0.83) CBR CBR = -14.0 + 0.66 (ELWD) for 12.5 MPa < ELWD < 174.5 MPa (R2 = 0.83) Abu-Farsakh et al. (2004) Subgrade and base course materials FWD MFWD = 0.97 (ELFWD) for 12.5 MPa < ELFWD < 865 MPa (R2 = 0.94) Abu-Farsakh et al. (2004) Subgrade and base course materials EFWD = 1.09ELWD, 2,240 psi < EPFWD < 30,740 psi (R2 = 0.64) George (2006) Subgrade soil MFWD = 1.031 ELWD(Prima 100) Fleming et al. (2000) Granular layer over silty clay MFWD = 1.05 to 2.22 EGDP Fleming et al. (2000) Granular layer over silty clay MFWD = 0.76 to 1. 32 ETFT Log ( LWD 30 k k ) = 0.0031 log (kLWD ) + 1.12 Kamiura et al. (2000) Subgrade soil PLT EPLT (i) = 22 + 0.7 (ELWD) for 12.5 MPa < ELWD < 865 MPa (R2 = 0.92) Abu-Farsakh et al. (2004) Subgrade and base course materials EPLT (R2) = 20.9 + 0.69 (ELWD) for 12.5 MPa < ELFWD < 865 MPa (R2 = 0.94) CBR = California bearing ratio; CIV = Clegg impact value; DPI = DCP penetration index; E = elastic modulus; EG = GeoGauge elastic stiffness modulus; EGDP = stiffness modulus from German dynamic plate; ELWD = stiffness modulus of LWD; EPLT = modulus from plate load test; ETFT = deformation modulus from Transport Research Laboratory Foundation Tester; HSG = GeoGauge stiffness reading (MN/m); MFWD = FWD back-calculated modulus; MR = resilient modulus. Parameter Device Correlations Reference Soil Type TABLE 38 (continued)

111 (ICMVs) and spot in situ tests (particularly DCP and LWD) when project scale averages were used rather than point-to- point comparisons. However, all correlations were project spe- cific and not universal because they were affected by different factors, including the heterogeneity in conditions of underlying layers, moisture content variation, and differences in influence depth between intelligent compaction (IC) rollers and other in situ test measurements. Continuous compaction control (CCC) or IC measurements areas during quality control/quality assur- ance (QC/QA) were found to be affected by the roller vibration amplitudes. As documented in this report, most research and implemen- tation projects that were conducted by the FHWA and state DOTs focusing on the use of CCC and IC reported consider- able success with and numerous benefits of these technologies. However, currently only three state DOTs (Indiana, Minne- sota, and Texas) have IC specifications. These specifications include the selection of a target ICMV based on acceptable stiffness or density spot-testing measurements obtained on control strips compacted using IC rollers. Acceptance is based on achieving the target ICMV for at least 80% to 90% of the compacted area. SUGGESTIONS FOR FUTURE RESEARCH The following suggestions are meant to address the gaps in knowledge and practices that were identified in this synthesis: • Future research is needed to develop more suitable labo- ratory compaction tests for unbound granular materials to better replicate field conditions in the laboratory. • There is a need to fully understand the effects of using stiffness- and strength-based compaction control speci- fications on a pavement structure’s longevity. This can be done by comparing the performance of similar pave- ment structures where conventional and stiffness- and strength-based compaction control specifications have been used. Future studies might investigate the rela- tionship between the in situ stiffness measurements of unbound pavement materials and subgrade soils and ulti- mate pavement performance. • The swell potential of fine-grained soils may not be opti- mized if stiffness/strength properties are used for their compaction control. Thus, future work could identify a criterion to address this issue in stiffness- and strength- based compaction control specifications. • Limited research has been conducted on the cost- effectiveness of using non-nuclear devices for com- paction control of unbound materials. Thus, future research might include life-cycle cost studies to eval- uate the economic benefits of using such devices. • A database for target values of in situ stiffness/strength measurements needs to be established for different soil types and moisture contents to facilitate the use of these devices in compaction control specifications. This data- base should be verified for local materials in each state before it is used in quality control. It is recommended that DOTs start with the DCP or LWD because the use of these devices has been successfully implemented in some states. • A database of the relationships between in situ test devices used in compaction control and the resilient mod- ulus design input value for different types of unbound materials should be developed. This will ensure that con- struction and design processes are fully integrated. • Because different models and types of the same in situ devices provide different measurements, there is a need to promote standardized protocols for in situ test devices used in compaction control specifications. Pooled fund studies, such as the one recently developed for LWD (Standardizing Lightweight Deflectometer Measurements for QA and Modulus Determination in Unbound Bases and Subgrades), will help in developing such protocols. In addition, state DOTs are to be encouraged to develop certification and training programs, such as those devel- oped in Indiana and Minnesota. • Future research might investigate the development of sta- tistical specifications for compaction control of unbound materials, which can account for the spatial variability of earthwork projects. The advent of the use of new devices that can rapidly assess the in situ stiffness/strength of unbound material can help to facilitate the implementa- tion of such specifications. • There is still a lack of experience and knowledge of how CCC and IC technologies can be used to improve con- ventional earthwork operations. Thus, more pilot and implementation projects of these technologies can help to optimize their usage and facilitate their implementa- tion in DOT construction practices and specifications. It is essential to monitor the cost and long-term perfor- mance of future projects in which CCC and IC tech- nologies are used, to fully understand their benefits and further support their effectiveness.