Documents

ythalm

Description
for laboratory RAP
Categories
Published
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Laboratory and Field Investigation of MechanisticProperties of Foamed Asphalt Recycled Base CourseMaterials for High Volume Pavements in India Dhiraj Minotra 1 ; A. Veeraragavan 2 ; and Rajib B. Mallick, M.ASCE 3 Abstract:  The quality of cold in-place recycling (CIR) is strongly dependent on the capability of proper laboratory investigation and design,and quality control during construction. The objective of this study was to evaluate the quality of the first major foamed asphalt CIR recycledlayer in India, laid on a national highway in India. A nonnuclear density gauge (NNDG) and a portable seismic properties analyzer (PSPA)were utilized to determine the density and stiffness of the foamed asphalt recycled layer. In addition, cores and material samples from therecycled layer were also tested in the laboratory for density and stiffness using dynamic modulus tests and the seismic method. The results of tests were related to material properties and critical factors influencing the performance-related properties were identified. Comparisonsbetween laboratory and field sample data were made. Foaming temperature and gradation control of crusher dust were identified as param-eters requiring tight control to achieve the required quality. The detrimental effects of excess blending of cement were also revealed. TheNNDG and the PSPA were found to be capable of identifying poor quality areas both in midsections and at joints, and to be capable of monitoring the quality of recycled layer.  DOI: 10.1061/(ASCE)CF.1943-5509.0000673.  © 2014 American Society of Civil Engineers. Author keywords:  Foam asphalt; Recycling; Nondestructive testing; Portable seismic properties analyzer. Introduction Cold in-place recycling (CIR) is an environmentally friendly pave-ment rehabilitation method that produces a flexible crack-resistant layer from an existing distressed layer [Kandhal and Mallick 1997;Asphalt Recycling, and Reclaiming Association (ARRA) 2001;Wirtgen 2012]. Considering its advantages of savings in materialscost, energy, and time, it has gained wide acceptance in different parts of the world in the recent past (Reddy et al. 2013). Cold in-place recycling in India is just venturing beyond the technologydemonstrator stage. A vast road network in India running over 4.6 million km, with its huge maintenance needs, nationwide short-age of aggregate, and restricted budgets set the perfect platform for adoption of cold recycling on a large scale. The first set of foamedasphalt CIR projects has just begun in India, and there is a criticalneed for investigation of key properties and variability, and iden-tification of important factors for ensuring good performance of recycled pavements. Since the success of a CIR project is heavilydependent on the ability to adjust the construction process to therequirements of the in-place materials, continuous constructionquality monitoring is essential, and nondestructive testing (NDT)provides the perfect technology for use during CIR operations.Nondestructive testing enables the collection of a huge amount of data in a relatively short period of time and hence, providesgreater confidence to the inspectors regarding the quality of therecycled layers. Objective The objectives of this study were to evaluate the performance-related stiffness and density properties of foamed asphalt CIR layer using NDT tools and techniques, comparing data from tests con-ducted on field cores and laboratory samples, and identify the criti-cal mix properties that are responsible for the variability of thedensity and stiffness properties. Scope The scope of this study included the testing of the in-place foamedasphalt CIR layer with the nonnuclear density gauge (NNDG) andthe portable seismic properties analyzer (PSPA), testing offield mixand cores in the laboratory, and analysis of the data. The pavement selected for this study is part of the first large-scale foamed asphalt CIR project in India. This pavement had shown early signs of rut-ting distress and top down cracking. It was selected as a candidatefor the first major foamed asphalt CIR project on a national high-way in India. This paper presents a description of the project, data collected from different field, and laboratory tests and analysis of the data. Description of Project The present study is on a project on a typical national highway inIndia, and it starts at chainage km   11  þ  000  and ends near  54  þ  365 , for a length of 43 km. The project is towiden the existingfour-lane highway (64.8 km) to six lanes (Fig. 1). The work involved a mix design for laying a base/binder course by cold 1 Former M.Tech Student, Dept. of Civil Engineering, Indian Instituteof Technology Madras, Chennai 600036, India. E-mail: dhirajminotra@gmail.com  2 Professor, Dept. of Civil Engineering, Indian Institute of TechnologyMadras, Chennai 600036, India. E-mail: av@iitm.ac.in 3 Professor, Dept. of Civil and Environmental Engineering, Worcester Polytechnic Institute (WPI), Worcester, MA 01609 (corresponding author).E-mail: rajib@wpi.eduNote. This manuscript was submitted on February 9, 2014; approved onAugust 8, 2014; published online on October 6, 2014. Discussion periodopen until March 6, 2015; separate discussions must be submitted for in-dividual papers. This paper is part of the  Journal of Performance of Con- structed Facilities , © ASCE, ISSN 0887-3828/04014176(7)/$25.00.  © ASCE 04014176-1 J. Perform. Constr. Facil.  J. Perform. Constr. Facil., 2015, 29(6): 04014176    D  o  w  n   l  o  a   d  e   d   f  r  o  m   a  s  c  e   l   i   b  r  a  r  y .  o  r  g   b  y   I  n   d   i  a  n   I  n  s   t   i   t  u   t  e  o   f   T  e  c   h  n  o   l  o  g  y   R  o  o  r   k  e  e  o  n   0   6   /   2   7   /   1   8 .   C  o  p  y  r   i  g   h   t   A   S   C   E .   F  o  r  p  e  r  s  o  n  a   l  u  s  e  o  n   l  y  ;  a   l   l  r   i  g   h   t  s  r  e  s  e  r  v  e   d .  in-place recycling using foamed bitumen to rehabilitate the existingfour lanes of the project road where the pavement showed distress. Site Inspection and Pavement Condition Assessment  The site was inspected and cores were taken from different chai-nages, where the pavement had shown extensive top down cracksand distress (Fig. 1). To verify the nature of the cracking in thepavement, a visual inspection of the cores was done. The cores re-vealed top down cracking and were found to be restricted only tothe hot mix asphalt (HMA) layer. The depth of the cracks variedfrom 60 to 80 mm from the top (Fig. 2). The existing pavement consisted of 40-mm bituminous concrete (BC) and 160-mm densebituminous macadam (DBM), laid in two layers of 80 mm thick-ness. Cold in-place recycling was selected to address the variousdistresses of rutting, shoving, and cracking observed in the existingpavement. Sample Collection and Properties of Reclaimed Asphalt Pavements  The reclaimed asphalt pavement (RAP) material was collectedusing milling machines from the longest stretch in the job to obtainrepresentative materials for the mix design. However, according toexisting technical guidelines (Asphalt Academy 2009), the samplecan be collected from any homogeneous stretch having uniform pavement composition, as bitumen stabilized material (BSM) isnot oversensitive material, small variability in untreated materialwillnot significantly affect the strength achieved through treatment.The characterization of RAP was carried out by conducting thefollowing tests: ã  Wet sieve analysis to determine the grading [ASTM D422(ASTM 2007)] ã  Atterberg limits to determine the plasticity index [ASTM D4318(ASTM 2010)] ã  Moisture/density relationship [AASHTO T-180 (AASHTO1995)]Blending of the milled RAP material with virgin aggregate wasrecommended in the design to meet the gradation limits of theTechnical Guideline 2 (TG 2) (Fig. 3.2, Asphalt Academy 2009),as shown in Fig. 3. Crusher dust with a nominal maximum aggre-gate size of 4.75 mm was blended with the RAP to achieve a mixthat was within the specified gradation limits. Accordingly, a blendof 80% RAP, 19% crusher dust, and 1% cement was recommendedfor the CIR. The filler was added according to the recommenda-tions of TG 2 to enhance the dispersion of the asphalt and reducemoisture susceptibility. The final design blend consisted of 4.7%fines passing 0.075 mm ( > 4%  laid down by TG 2). A viscositygraded 10 (VG 10) asphalt binder was used in the mix, with a foaming temperature of 180°C and 6% foaming water content. Theoptimum moisture content was determined to be 6.5%.The sequence of construction in the field consisted of spreadingthe materials to be blended with the RAP (crusher dust and cement)on the pavement surface, foaming and blending and finally, com-paction. After the initial trials showed variability in the quantity of the crusher dust (spread with a grader), a paver was used for thepurpose of achieving better process control. A cement spreader wasused throughout the project to spread the dry cement. The perfor-mance of the equipment showed a dosing variability of up to 2x thestipulated quantity. This was attributed to the high fineness content of the filler [ordinary portland cement (OPC) grade 53].The lack of the in-built heating arrangement in the bulk asphalt tanker used in the recycling train led to a drop in foaming temper-atures down to 150°C in some cases, compared with the designrequirement of 180°C. A Wirtgen WR2400 (Wirtgen, Windhagen,Germany) recycler was used in the project.To achieve the final base course layer thickness of 200 mm (required to match the newly constructed adjacent lane), 20% of the recycled layer material was required to be removed by a grader.The lane-wise grading of this high quantity of material ideallyrequired two graders in tandem, which would have ensured that thefinal profilewas achieved before the stiffening of the recycled layer.The process was however slowed down by the presence of only onegrader with a blade width of 2.5 m. This led to compaction andprofile correction being carried out for up to 5 – 6 h after recycling,which most likely resulted in stiffening of the recycled layer andimpeded adequate compaction.Stretches of the recycled pavement (before the application of theoverlay) was opened to traffic at differing time periods after thefinal compaction, without the use of any proof rolling. No standardtest method was available to ascertain whether adequate strengthhad been achieved before letting traffic ply on the freshly con-structed stretch. Finally, the application of the HMA overlay, whichis recommended to be completed by the end of 2 – 4 weeks after recycling (TG 2, Asphalt Academy 2009) was delayed for up to12 weeks in certain cases. This exposed the recycled layer to trafficwithout the protection of a wearing course for 8 weeks longer thanrecommended. The field laboratory of the construction agencymade a dry recycler pass after spreading of the crusher dust andcement and carried out gradation of the blend to verify the correct proportioning of the constituents; that is, the RAP, crusher dust, andcement. This method was used instead of measuring individualconstituents using tray tests, which require the continuous presenceof the quality control (QC) team at the construction site. However,the validation of this process in the laboratory revealed that thein situ moisture has a dominant effect on the measured percentageof fines passing through a 0.075 mm sieve. The presence of moisture leads to the cement forming agglomerates with the finesin the RAP and crusher dust. A reduction of up to 40% in the fineswas observed due to this phenomenon in a controlled test in the Fig. 1.  Location of the project (length: 64.8 km) (map data © 2014Google)  © ASCE 04014176-2 J. Perform. Constr. Facil.  J. Perform. Constr. Facil., 2015, 29(6): 04014176    D  o  w  n   l  o  a   d  e   d   f  r  o  m   a  s  c  e   l   i   b  r  a  r  y .  o  r  g   b  y   I  n   d   i  a  n   I  n  s   t   i   t  u   t  e  o   f   T  e  c   h  n  o   l  o  g  y   R  o  o  r   k  e  e  o  n   0   6   /   2   7   /   1   8 .   C  o  p  y  r   i  g   h   t   A   S   C   E .   F  o  r  p  e  r  s  o  n  a   l  u  s  e  o  n   l  y  ;  a   l   l  r   i  g   h   t  s  r  e  s  e  r  v  e   d .  laboratory when 2.5% of water of the RAP by weight was added tosimulate the effect of the in situ moisture. NNDG Data The use of the PaveTracker Plus NNDG (Troxler ElectronicLaboratories, Research Triangle Park, North Carolina) enabledthe collection and analysis of a high volume of data, which isnot feasible by sand replacement tests and core extraction. The de-vice was calibrated using modified Marshall Laboratory samples inthe laboratory. The PaveTracker, exhibited high repeatability andwas sensitive to measurements taken along different orientations.The higher moisture content and air voids of the foamed asphalt before final compaction strongly influences the dielectric constant of the material, which directly impacts the NNDG measurements.After the completion of the initial compaction by the soil compac-tor, the NNDG was used to take density measurements across a given cross section. This process resulted in density measurementsof the material that were higher than the expected range. It isanticipated that this is due to the high moisture content and air voids. However, the sensitivity of the NNDG was adequate to iden-tify the locations of inadequate compaction at the longitudinal con-struction joints at 2.4 and 6 m offset from the median. Ideally, thearea should be reworked at the joints and the density measurements Fig. 2.  Cracking in existing roadway, core from existing roadway, and PSPA testing on recycled base course (images by Dhiraj Minotra)  © ASCE 04014176-3 J. Perform. Constr. Facil.  J. Perform. Constr. Facil., 2015, 29(6): 04014176    D  o  w  n   l  o  a   d  e   d   f  r  o  m   a  s  c  e   l   i   b  r  a  r  y .  o  r  g   b  y   I  n   d   i  a  n   I  n  s   t   i   t  u   t  e  o   f   T  e  c   h  n  o   l  o  g  y   R  o  o  r   k  e  e  o  n   0   6   /   2   7   /   1   8 .   C  o  p  y  r   i  g   h   t   A   S   C   E .   F  o  r  p  e  r  s  o  n  a   l  u  s  e  o  n   l  y  ;  a   l   l  r   i  g   h   t  s  r  e  s  e  r  v  e   d .  taken again. Recycling was conducted through four recycler passesas follows: Pass 1: 0 – 2.3 m; Pass 2: 2.3 – 4.6 m; Pass 3: 4.6 – 6.9 m;Pass 4: 6.2 – 8.5 m (recycled aligned to the edge of pass 4).The variability in compaction along each recycled pass isexpected to be lower than the variability across the cross sectiondue to the compaction sequence typically followed in constructionsites. In order to evaluate the sensitivity of the NNDG to detect a possibility of raveling, a comparison of a raveled stretch (ravelingobserved in approximately 20% of the area) was carried out with anadjacent intact stretch of 50 m. Measurements were taken in themiddle of the recycler pass at 1.2 and 3.6 m offset from medianto avoid the influence of longitudinal construction joints (Fig. 4).The higher coefficient of variation (COV) of the measured values inthe raveled section (Table 1) gives an indication of the poor qualityof compaction. The poor performance of the raveled section is,however, indiscernible from the individual or average values of density readings. Ideally, the process parameters must be recordedduring execution. This information, in conjunction with NNDGmeasurements can be used to decide if a reworking is requiredin a section detected with higher COV before the result of poor quality manifests on the surface as raveling or some other distresssubsequently. Analysis of Overall Compaction Sequence  Density measurements taken across different cross sections, 50 m apart, are plotted in Fig. 5. Calculation of COVof these measure-ments at each offset exhibited a moderate variability (5 – 10%),which indicates that a similar compaction sequence was followedin all sections. Relatively lower densities are observed at longitu-dinal constructionjoints at an offset of 2.4 and 6m, which was most likely due to a reduced temperature of the foamed asphalt. Measurements on Either Side of a Transverse Construction Joint  Transverse construction joints at the junction of the end of a con-structed section and the commencement of a new section requireparticular attention to avoid low densities. Measurements weretaken by the NNDG on either side of different transverse jointsat six offsets, each at an interval of 1.2 m (Fig. 6). These measure-ments show cross sections with low density on the side of thetransverse joint toward which work commenced. This may beattributable to the tendency to avoid reversing the soil compactor  Fig.3. Gradation limits as per TG2 (data from  Asphalt Academy 2009)and Design blend. Note: RAP Chennai TADA represents the gradationof the milled material Fig. 4.  NNDG measurements at different sections Table 1.  Average and Variability of NNDG Data Location Offset Average density (kg = m  3 ) COV (%)km 31,800 (raveled) 1.2 m 1,918 5.852.4 m 2,037 5.97km 31,900 (intact) 1.2 m 1,944 2.962.4 m 2,033 3.22 Fig. 5.  Density at different offsets from median for different compac-tion passes Fig.6. Density at start and end of sections for different offsets from themedian  © ASCE 04014176-4 J. Perform. Constr. Facil.  J. Perform. Constr. Facil., 2015, 29(6): 04014176    D  o  w  n   l  o  a   d  e   d   f  r  o  m   a  s  c  e   l   i   b  r  a  r  y .  o  r  g   b  y   I  n   d   i  a  n   I  n  s   t   i   t  u   t  e  o   f   T  e  c   h  n  o   l  o  g  y   R  o  o  r   k  e  e  o  n   0   6   /   2   7   /   1   8 .   C  o  p  y  r   i  g   h   t   A   S   C   E .   F  o  r  p  e  r  s  o  n  a   l  u  s  e  o  n   l  y  ;  a   l   l  r   i  g   h   t  s  r  e  s  e  r  v  e   d .
Search
Similar documents
Tags
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks