Illllllllllllllllllllllllllf . THESIS “ z fill This is to certify that the thesis entitled OPTIMIZING THE EFFICIENCY OF TRANSVERSE JOINTS ANDCRACKS IN ROLLER-COMPACTED CONCRETE (RCC) PAVEMENTS presented by Jacob E. Hiller has been accepted towards fulfillment of the requirements for MS. degree inCivil Engineering 4 M/y WW professor Date ‘QI/gl/OO 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution J..- M LlBRARY Michigan State _ University L3,- _ PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE @3ng 7 2.0 4 v: oz; as moo amass-p.14 0PTl\llZl\G l ROU OPTMZING THE EFFICIENCY OF TRANSVERSE JOINTS AND CRACKS IN ROLLER-COMPACTED CONCRETE (RCC) PAVEMENTS By Jacob Eskel Hiller A THESIS Submitted to - Michigan State University in partial firlfillment of the requirements for the degree of MASTER OF SCIENCE Department of Civil and Environmental Engineering 2000 TEEEGHEP ROLLER-‘ Reliermmpacté’d it mist-am of aspha mac garment R( m heaiily loaded 1 Bf and associated labc vam'oml concrete pa hhmaummi mm “id’JLs‘ whiz? This thesis viii aI 1011' - Westmcm'e deli; ifi‘ Dr 5 A ‘ ‘ 3mm rehabii. m.“ H ABSTRACT OPTINIIZING THE EFFICIENCY OF TRANSVERSE JOINTS AND CRACKS IN ROLLER-COMPACTED CONCRETE (RCC) PAVEMENTS Jacob Eskel Hiller Roller-compacted concrete (RCC) has been utilized since the 1930’s. It combines the practicality of asphalt paving procedures with the durability of a portland cement concrete pavement. RCC pavements provide an initial cost-effective solution to low volume, heavily loaded pavements. Due to lack of forms, reinforcing steel, temperature steel, and associated labor costs, RCC pavements have resulted in savings of 10-58% over conventional concrete pavement construction. The applications of RCC pavements have been limited due to the varying surface roughness (and corresponding fiction) as well as large crack widths, which develop from the long crack spacings. This thesis will aim to evaluate existing RCC pavements in service through the use of non-destructive deflection data to estimate the stresses due to a traffic load. Alternatives (such as increased RCC design thickness, engineered joint spacing at closer distances than the natural crack spacing, or dowel bar retrofitting of natural cracks) will be examined in order to produce an efficient design based on cost. Options for in-service RCC pavement rehabilitation will be produced and future RCC pavement design recommendations will be made. laoald to thank arms a sell as mtl‘t c L357?) luould also like ETM’Cll work and 3 mt: pat merits and laould also like arise regeding the l Aspecia‘ tl‘anlts freeman and Mia? 3331“ng data of ? laould like to g 323m): prodded su A final that}; S r - g, writ 'n 9 ‘ M“ ths Stage 0:" ACKNOWLEDGEMENTS I would to thank Dr. Neeraj Buch for his support and direction on numerous prOjects as well as with career objectives throughout my graduate career at Michigan State University. I would also like to thank the Portland Cement Association (PCA) for sponsoring this research work and providing financial support. Specifically, I would like to thank Jan Prusinski of PCA for providing technical support on the subject of roller-compacted concrete pavements and Steve Kosmatka for his help on this project. I would also like to give thanks to Tom Yu of ERES Consultants for his technical expertise regarding the ISLABZOOO rigid pavement analysis program. A special thanks goes out to Linda Pierce of the Washington Department of Transportation and Michael Eacker of the Michigan Department of Transportation for providing FWD data of FCC retrofitted dowel sites. I would like to give a special thanks to my parents, John and Sharon Hiller, who have always provided support and encouragement in my academic career as well as in life. A final thanks goes to my wife, Renee, for her patience and understanding throughout this stage of our life together. iii Llil 0f TAB LES XS". 0F IlGl'RES C‘APTZR l [\TRODL'CIIO,‘ Problem 3 Objecttxes Scope ot‘i Contents I i lllERtTlRE R Construct: TABLE OF CONTENTS Page LIST OF TABLES ................................................................................................ vii LIST OF FIGURES ............................................................................................... x CHAPTER I WTRODUCTION ..................................................................................... 1 Problem Statement ......................................................................... 1 Objectives of Research .................................................................... 1 Scope of Research .......................................................................... 2 Contents of Thesis .......................................................................... 3 II LITERATURE REVIEW ........................................................................... 5 Construction Practices .................................................................... 5 History ................................................................................ 5 Underlying Layer Construction ............................................ 6 Test Sections ...................................................................... 7 Production and Transportation ............................................ 8 Paving Operations ............................................................... 10 Compaction ........................................................................ 12 Curing ................................................................................. 13 Contraction Joints and Load Transfer Devices ..................... 13 RCC Material Properties ................................................................. 14 Overview ............................................................................ 14 Materials ............................................................................. 16 Mechanical Properties ......................................................... 20 Structural Design of RCC Pavements .............................................. 25 Overview ............................................................................ 25 PCA Method ....................................................................... 26 US. Army COE Method ..................................................... 27 Comparison and Contrast .................................................... 28 Field Performance of RCC Pavements ............................................. 28 Future of RCC Pavements .............................................................. 29 III RCC PAVEMENT SITES ......................................................................... 31 Description of Test Sites ................................................................. 31 Austin, TX .......................................................................... 31 Fort Campbell, KY .............................................................. 34 Fort Drum, NY ................................................................... 34 Fort Hood, TX .................................................................... 3 5 iv S I DmaCo Data An I I I Load Ir Load Ir \ I ll RCC PMBIE 0xeme‘ I I Guide tt‘ 5 1 l l I I Compy NESTIGAII Design : l 1 D0“ el [ l l Page Spring Hill, TN ................................................................... 37 Edmonton, AB, Canada ...................................................... 37 Data Collection ............................................................................... 38 Data Analysis .................................................................................. 39 ERES Method ..................................................................... 40 ECOPP Method .................................................................. 44 Backcalculation Procedure Comparison ............................... 47 Load Transfer Efliciency Background ............................................. 52 Load Transfer Efiiciency Trends ..................................................... 56 Waterways Experiment Station Study Trends ...................... 56 Edmonton, AB Study Trends .............................................. 63 IV RCC PAVEMENT DESIGN ..................................................................... 68 Overview of Design Methodology .................................................. 68 Effect of Load Positioning on Tensile Stress ........................ 68 Effect of Slab Dimensions on Tensile Stress ........................ 69 Guide to RCC Pavement Structural Design ..................................... 70 Stress Ratio and Allowable Stress ....................................... 72 Temperature Gradients ........................................................ 76 Lateral Load Location ......................................................... 79 Determination of Pavement Thickness ................................. 83 Limiting Subgrade Stress ..................................................... 87 Mixed Traffic Designs ......................................................... 89 Comparison of Results to Other RCC Pavement Design Methods 92 V INVESTIGATION OF RCC PAVEMENT ALTERNATIVES .................. 97 Design Alternatives ......................................................................... 97 Engineered Joint Spacing .................................................... 97 Increased RCC Slab Thickness ............................................ 100 Dowel Bar Retrofitting as a Rehabilitation Alternative .................... 102 Overview ............................................................................ 103 Importance of Dowel Bar Retrofitting ................................. 105 Current Methodology of Selection ...................................... 108 Analytical Modeling ............................................................ 109 Field Validation of Analytical Results ............................... - 114 Benefits of Selective Retrofitting ......................................... 120 VI CONCLUSIONS, RECOMMENDATIONS, AND FUTURE RESEARCH NEEDS ................................................................................. 122 Conclusions and Recommendations ............................................. 122 Backcalculation Procedures ................................................. 122 Load Transfer Efliciency Correlations ................................. 123 RCC Pavement Design ........................................................ 125 RCC Pavement Rehabilitation Alternatives .......................... 126 Future t??E\DI\' A TABl ,XIPINDIX B DO“ IP‘F’IENDIFIC CAI. 3.9.930“ D RN PPEXDL‘IE GU: FEREXCES Page Future Research Needs ................................................................... 127 APPENDIX A: TABULATED CRITICAL TENSILE STRESSES ...................... 131 APPENDD( B: DOWEL BAR RETROFIT GUIDE ............................................. 152 APPENDIX C: CATALOG OF RCC DESIGN THICKNESSES .......................... 206 APPENDIX D: RAW LOAD TRANSFER DATA ............................................... 222 APPENDIX E: GUIDE FOR LIMITING SUBGRADE STRESS ......................... 232 REFERENCES ...................................................................................................... 246 ABLE RCC Paretnen’. RCC Pat ement Deflation Data Regression C Oct Regression C oet“ Comparison oIE Fatigue Relation Fatigue Relatior Reiablit} LC‘» e' “ember for l “IOTILSheet for 1 Data to be L‘se. Worksheet for Data to be I'se RCC niches CHIN SIIESSQ (meal SUESSQ (Erica. SUesse Ciitical 536ng C ; t. . Ina; SUesse LIST OF TABLES TABLE Page 1 RCC Pavement Site Information (After Pittman [1]) ................................... 32 2 RCC Pavement Mix Proportion Information (After Pittman [1]) ................. 33 3 Deflection Data from Different RCC Test Sites .......................................... 40 4 Regression Coefficients for 5: (After [23]) ................................................. 42 5 Regression Coefiicients for al, a2, a3, and a; (After [26]) ............................ 46 6 Comparison of Backcalculated Parameters from Different RCC Test Sites. 47 7 Fatigue Relationship (After PCA [14]) ....................................................... 73 8 Fatigue Relationship (After Tayabji and Halpenny [34]) .............................. 74 9 Reliability Levels for Different Wheel Load Locations ................................ 82 10 Worksheet for Example on Determination of RCC Thickness ..................... 85 11 Worksheet for Determination of RCC Thickness ........................................ 86 12 Data to be Used in Mixed Traflic RCC Design Example ............................. 90 13 Worksheet for Mixed Traffic RCC Design Example ................................... 89 14 Data to be Used in RCC Design Comparisons ............................................ 91 15 RCC Thickness Results of RCC Design Comparison in Inches ................... 92 A-1 Critical Stresses (psi) for a 12 Kip Single Axle Load .................................. 132 A-2 Critical Stresses (psi) for an 18 Kip Single Axle Load ................................ 136 A-3 Critical Stresses (psi) for a 24 Kip Tandem Axle Load ............................... 140 A—4 Critical Stresses (psi) for a 30 Kip Tandem Axle Load ............................... 144 A-S Critical Stresses (psi) for a 36 Kip Tandem Axle Load ............................... 148 lkBLI {-1 r‘“, 1.. RCC Desrgn I HOG p51 in RCC Design I: Hit psi il'l -. RCC Deszgn TI Hill: psr lli RCC Design Ti l=l00 psitn l‘ZSO psi In {‘5 RCC Design T HOG psi In ['7 RCC Deggn I L: l 0‘3 W in . Ii Rcc Design T HBO PSi In [.51 ’ RCC DeSiEn T NY; .‘ ~ "V Psi m .. RCC Design] TABLE C-1 C-2 C-3 C-4 C-5 C-9 C-lO C-ll RCC Design Thicknesses for a 12 Kip Single Axle Load and k=100 psi/In .............................................................................................. 207 RCC Design Thicknesses for a 12 Kip Single Axle Load and k=250 psi/in .............................................................................................. 208 RCC Design Thicknesses for a 12 Kip Single Axle Load and k=400 psi/in .............................................................................................. 209 RCC Design Thicknesses for an 18 Kip Single Axle Load and k=100 psi/in .............................................................................................. 210 RCC Design Thicknesses for an 18 Kip Single Axle Load and F250 psi/In .............................................................................................. 211 RCC Design Thicknesses for an 18 Kip Single Axle Load and k=400 psi/1n .............................................................................................. 212 RCC Design Thicknesses for a 24 Kip Tandem Axle Load and k=100 psi/In .............................................................................................. 213 RCC Design Thicknesses for a 24 Kip Tandem Axle Load and k=250 psi/In .............................................................................................. 214 RCC Design Thicknesses for a 24 Kip Tandem Axle Load and k=400 psi/In .............................................................................................. 215 RCC Design Thicknesses for a 30 Kip Tandem Axle Load and k=100 psi/In .............................................................................................. 216 RCC Design Thicknesses for a 30 Kip Tandem Axle Load and k=250 psi/In .............................................................................................. 217 viii Page llBLI C-lZ S.) 0) -‘.‘ RCC Design I are Design "~ k=100 psi In I= 103 psi tn RCC Design I -1 r . - l--."J psr In '3: RCC Desrgn] k=400 on In LIE Tests on ”E IeStS 0n TABLE Page C-12 RCC Design Thicknesses for a 30 Kip Tandem Axle Load and k=400 psi/In .............................................................................................. 218 C-13 RCC Design Thicknesses for a 36 Kip Tandem Axle Load and k=100 psi/tn .............................................................................................. 219 014 RCC Design Thicknesses for a 36 Kip Tandem Axle Load and k=250 psi/In .............................................................................................. 220 C-15 RCC Design Thicknesses for a 36 Kip Tandem Axle Load and k=400 psi/In .............................................................................................. 221 D-l LTE Tests on Transverse Cracks (After Pittman [6]) ................................. 223 D-2 LTE Tests on Transverse Joints (after Wu and Todres [3 9]) ...................... 228 ix IlGIRI 1 Rite-"iii“ended rm.Sitaft PL: 3 HOW CIR“ . Pater with DOL- 5 RCC Being Pia; é Dial-Drum R053 Close-l'p \‘iev. . In lieu ot‘RCl Close-In View Edge View of R longitudinal Co ~ Results of Bach Resins of Back. Results of Back; Deletion of 0°: Depiction of 101 Relationship Bet Rflattomhip Bet Relax] ' . 0 4‘, lib-up Bet Rilationshjp Bet LIST OF FIGURES FIGURE Page 1 Recommended Test Section Setup (After Pittman [2]) ................................ 8 2 Twin-Shaft Pugrnill Mixing Plant ................................................................ 9 3 Hopper of RCC/Asphalt Paver ................................................................... 10 4 Paver with Double Tarnping Bar Screed ..................................................... 1 1 5 RCC Being Placed and Tested for Density Before Rolling .......................... 1 1 6 Dual-Drum Roller Compacting Freshly Laid RCC ...................................... 12 7 Close-Up View of RCC Surface (After One Day) ....................................... 15 8 Far Vrew of RCC Surface (After One Day) ................................................ 16 9 Close-Up View of Fresh RCC .................................................................... 17 10 Edge View of RCC Pavement (After One Day) .......................................... 21 1] Longitudinal Cold Joint (After One Day) .................................................... 24 12 Results of Backcalculated Elastic Modulus of the Concrete Comparison ..... 48 13 Results of Backcalculated Modulus of Subgrade Reaction Comparison ...... 50 14 Results of Backcalculated Radius of Relative Stiffness Comparison ............ 52 15 Depiction of 0% Deflection Load Transfer Efficiency (After Buch [29]) ..... 55 16 Depiction of 100% Deflection Load Transfer Efficiency (After Buch [29]). 55 17 Relationship Between Crack Width and Crack Spacing ............................... 57 18 Relationship Between Crack Width and LTE5 ............................................. 59 19 Relationship Between Crack Spacing and LTEa .......................................... 59 20 Relationship Between Modulus of Subgrade Reaction and LTEs ................ 60 IlGLRlE Il Relationship l Reiationship II Relationship I) Relationship Ii Reiations'nip 8 Relationship Bt Slab lhmensior F10“ C hart for Companson of DWWUd Slat [tiara Slab o It?“ Probahi [”3 OfRelta‘r; E5“ GIReiiab 55w of Rehab “mating Plate llh’lucing 0f JOIr tom (‘7 h. We... 1 RPM“) a FIGURE Page 21 Relationship Between Radius of Relative Stiffness and LTEa. ..................... 61 22 Relationship Between Average Joint Spacing and LTE5 .............................. 63 23 Relationship Between Joint Spacing and LTE5 at Edmonton, AB Site ......... 64 24 Relationship Between 8 and LTEa at Edmonton, AB Site. .......................... 65 25 Relationship Between k and LTE;> at Edmonton, AB Site ........................... 66 26 Relationship Between Eh3 and LTEs at Edmonton, AB Site ........................ 66 27 Slab Dimensions and FEA Mesh for Critical Load Position at Edge ............ 69 28 Flow Chart for Methodology of “New” RCC Pavement Design Process ..... 71 28 Comparison of PCA and Tayabji/Halpenny Fatigue Relationships ............... 75 3O Downward Slab Curling due to Positive Gradient. ...................................... 78 31 Upward Slab Curling due to Negative Gradient. ......................................... 78 32 Typical Probabilistic Lateral Load Distribution on 12’ Wide Lane ............... 80 33 Level of Reliability for Load Positioning on 12’ Wide Lane. ....................... 81 34 Effect of Reliability on “New” Method Thickness Without Temperature ..... 95 35 Effect of Reliability on “New” Method Thickness With Temperature. ......... 96 36 Vibrating Plate with Welding Fin for Use in Inducing Joints. ...................... 99 37 Inducing of Joints Using the Vibrating Fin in an RCC Base ......................... 99 38 Effect of LTE; on Tensile Stress near the Crack or Joint for k=100 psi/in (27.1 kPa/mm) and No Temperature Gradient ............................................ 10] 39 Effect of LTE; on Tensile Stress near the Crack or Joint for k=250 psi/in (67.9 kPa/mm) and No Temperature Gradient. ........................................... 101 40 Effect of LTE; on Tensile Stress near the Crack or Joint for k=400 psifrn (108.6 kPa/mm) and No Temperature Gradient ................................. 102 IlGIRI ll ,2" 5-3 if Distinction Bl Comparison c Donel Bat R Sill) D‘imefls‘. Example of l Example ot‘l Example of' Iieid Vent]; Iieicl Verjfi; Reid Vent?“ VIEW of Sp; Fieid Vent} Cnfiw Tm Located 24' FIGURE Page 41 Distinction Between Preventive and Corrective Maintenance. ..................... 107 42 Comparison of Performance Benefits of Rehabilitation Over Time. ............. 108 43 Dowel Bar Retrofit Locations for ISLAB2000 Modeling ............................ 1 10 44 Slab Dimensions and FEA Mesh for Joint/Crack Load Position. ................. 111 45 Example of LTEs Increase for Given Levels of AGG. ................................. 112 46 Example of Reduction in Maximum Tensile Stresses By Using DBR .......... 113 47 Example of Final LTEa for Given Levels of Original LTEa .......................... 114 48 Field Verification of LTEs for DBR Site on I-69 in Michigan ...................... 116 49 Field Verification of LTEs, for DBR Site on I-75 in Michigan ...................... 1 16 50 Field Verification of LTE5 for DBR Site on M-14 in Michigan ................... 117 51 View of Spalling in Grout Covering of M-14 DBR Site .............................. 118 52 Field Verification of LTEs for DBR Site on I-90 in Washington .................. 119 A-1 Critical Tensile Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle Located 24” (61 mm) from the Pavement Edge with AT=0°F (0°C). .......... 133 A-2 Critical Tensile Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle Located at the Pavement Edge with AT=0°F (0°C) ..................................... 133 A-3 Critical Tensile Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle Located 2 ” (61 mm) from the Pavement Edge with AT=15°F (83°C) ....... 134 A4 Critical Tensile Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle Located at the Pavement Edge with AT=15°F (83°C) ................................ 134 A-5 Critical Tensile Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle Located 24” (61 mm) from the Pavement Edge with AT=—15°F (-8.3°C) 135 xii IlCi‘tRI it Critical Tensile located at the l-' Critical Tensile located 24" it ll Cntical Tensile I located at the l) 51-“. Critical Tensile S locatcd 24" (01 ll Critical Tensile 5 located at the P. .l-ll Critical Tensile S heated 24“ (61 III (nocal Tensile S l-Ocaed at the P .l-Il Critical Tensile S located :4“ to: ‘34 Critical Tensile S Wedmhep FIGURE Page A-6 Critical Tensile Stresses for a 12 kip (53.4 km Dual-Tired Single Axle Located at the Pavement Edge with AT=-15°F (-8.3°C) ............................. 135 A-7 Critical Tensile Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle Located 24” (61 mm) from the Pavement Edge with AT=O°F (0°C) ........... 137 A—8 Critical Tensile Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle Located at the Pavement Edge with AT=0°F (0°C) ..................................... 137 A-9 Critical Tensile Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle Located 24” (61 mm) from the Pavement Edge with AT=15°F (83°C) ....... 138 A-10 Critical Tensile Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle Located at the Pavement Edge with AT=15°F (83°C) ................................ 138 A-ll Critical Tensile Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle Located 24” (61 mm) from the Pavement Edge with AT=-15°F (-8.3°C) 139 A-12 Critical Tensile Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle Located at the Pavement Edge with AT=~15°F (-8.3°C) ............................. 139 A-13 Critical Tensile Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=0°F (0°C) ........... 141 A-14 Critical Tensile Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=0°F (0°C) ..................................... 141 A-15 Critical Tensile Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=15°F (83°C) ....... 142 A-l6 Critical Tensile Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=15°F (83°C) ................................ 142 IlClRI ti? Critical Tensiiv. Located 24" t' til Critical Tensile i located at the ill Critical Tensile located 24" it .‘s-Il Cnncal Tensile located at die I .l-Bl Critical Tensile! located 24“ to 3-: CntICal Tensile 1 located at the 1 *3 Critical Tensile Located :4" 1'6 FIGURE Page A-17 Critical Tensile Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=~15°F (-8.3°C) 143 A-l8 Critical Tensile Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=—15°F (-8.3°C) ............................. 143 A—19 Critical Tensile Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=0°F (0°C) ........... 145 A-20 Critical Tensile Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=0°F (0°C) ..................................... 145 A21 Critical Tensile Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=15°F (83°C) ....... 146 A-22 Critical Tensile Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=15°F (83°C) ................................ 146 A-23 Critical Tensile Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=-15°F (-8.3°C).... 147 A24 Critical Tensile Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=~15°F (-8.3°C) ............................. 147 A-25 Critical Tensile Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=O°F (0°C) ........... 149 A-26 Critical Tensile Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=0°F (0°C) ..................................... 149 A-27 Critical Tensile Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=15°F (83°C) ....... 150 xiv HGIRI loZS Critical Tens.- .i-lq CTlllCAl Tensi'. t .l-ll‘ C natal Tensile 33 3-5 located at the located 34" l’ Located at the mi mth Res; (152 nan). k=l: ”Ev” “Ill! Resp llSZ mu [:1 ”£6 Villh Res; ”53%). k=1 FIGURE Page A-28 Critical Tensile Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=15°F (83°C) ................................ 150 A-29 Critical Tensile Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=-15°F (-8.3°C).... 151 A-30 Critical Tensile Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=—15°F (-8.3°C) ............................. 151 B-l LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=100 psi/In (27.1 kPa/mm), and AT=O°F (0°C) ........................ 152 8-2 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=100 psi/In (27.1 kPa/mm), and AT=+15°F (+8.3°C) ............ 152 8-3 LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=100 psi/In (27.1 kPa/mm), and AT=—15°F (-8.3°C) .............. 154 8-4 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=250 psi/tn (67.9 kPa/mm), and AT=O°F (0°C) ................... 154 B-S LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=250 psi/in (67.9 kPa/mrn), and AT=+15°F (+8.3°C) ............. 155 8-6 LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=250 psi/In (67.9 kPa/mm), and AT=—15°F (-8.3°C) .............. 155 B-7 LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=400 psi/In (106.8 kPa/mm), and AT=0°F (0°C) ................... 156 IlGlRI. M ITEe “Till RS) (152mm. l“ 33 [Th will Rt“ (152mm). k are lilit~ with R“ (203 mm). k=l in HE,- m‘th Res (”203 rnrni. k=l 322 ll}; with Res (203 min). k=l 3‘13 HE, with Res (203 mm). H 314 LTE, nith Res l303 mm. H 325 ”E; Willi Res; FIGURE Page 8-8 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=400 psi/1n (106.8 kPa/mm), and AT=+15°F (+8.3°C) ............ 156 8-9 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=400 psi/In (106.8 kPa/mm), and AT=-15°F (-8.3°C) ............. 157 B-10 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=100 psi/1n (27.1 kPa/mm), and AT=0°F (0°C) ..................... 157 3-11 LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=100 psi/tn (27.1 kPa/mm), and AT=+15°F (+8.3°C) ............. 158 B-12 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=100 psi/tn (27.1 kPa/mm), and AT=-15°F (-8.3°C) ............... 158 8-13 LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=250 psi/In (67.9 kPa/mm), and AT=O°F (0°C) ..................... 159 B-l4 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=250 psi/In (67.9 kPa/mm), and AT=+15°F (+8.3°C) .............. 159 8-15 LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=250 psi/tn (67.9 kPa/mm), and AT=-15°F (-8.3°C) ............... 160 B-16 LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=400 psi/In (106.8 kPa/mm), and AT=0°F (0°C) .................... 160 B-1 7 LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=400 psi/In (106.8 kPa/mm), and AT=+15°F (+8.3°C) ............. 161 IlGlRI . I 3.18 LTE Mill Rt 033 ml. lit 3' .. 9 UR mth Re. l35~l m). R‘- 520 lTL nith Re: (:54 31ml. [(2 ‘ ill LTE-é with Res M l LTE mth Re- FIGURE B-18 8-19 3-20 B-21 8-22 8-23 13-24 3-25 B-26 B-27 LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=400 psi/In (106.8 kPa/mm), and AT=-15°F (-8.3°C) ............ 161 LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=100 psi/m (27.1 kPa/mm), and AT=O°F (0°C) .................. 162 LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=100 psi/In (27.1 kPa/mm), and AT=+15°F (+8.3°C) ............ 162 LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=100 psi/In (27.1 kPa/mm), and AT=-15°F (-8.3°C) ............. 163 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=250 psi/in (67.9 kPa/mm), and AT=O°F (0°C) .................. 163 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=250 psi/In (67.9 kPa/mm), and AT=+15°F (+8.3°C) ............ 164 LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=250 psi/In (67.9 kPa/mm), and AT=-15°F (-8.3°C) ............. 164 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=400 psi/In (106.8 kPa/mm), and AT=0°F (0°C) .................. 165 LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=400 psi/in (106.8 kPa/mm), and AT=+15°F (+8.3°C) .......... 165 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=400 psi/In (106.8 kPa/mm), and AT=—15°F (-8.3°C) ............ 166 xvii Page IllilRI ill 3-39 330 ll .__‘ an 333 it»: 835 LII.- on n.‘ (305 mm), it was Mill Re llC'S mm}. k; “E mth Re~ lin m)‘ [-1 LTE with Res ‘3‘35 mm). b: LTE-a saith Res (305 mm). RI: LIE with Res (305 mm). it: LTE,- Rith R65 (3'35 mm )‘ ha LIE; Rith R65 FIGURE Page 8-28 LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=100 psi/m (27.1 kPa/mm), and AT=O°F (0°C) .................. 166 8-29 LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=100 psifrn (27.1 kPa/mm), and AT=+15°F (+8.3°C) ............ 167 8-30 LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=100 psi/1n (27.1 kPa/mm), and AT=—15°F (-8.3°C) ............. 167 B-31 LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=250 psi/in (67.9 kPa/mm), and AT=0°F (0°C) ................... 168 B-32 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=250 psi/In (67.9 kPa/mm), and AT=+15°F (+8.3°C) ............ 168 B-33 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=250 psi/in (67.9 kPa/mm), and AT=—15°F (-8.3°C) ............. 169 B-34 LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=400 psi/tn (106.8 kPa/mm), and AT=O°F (0°C) .................. 169 B-35 LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=400 psi/m (106.8 kPa/mm), and AT=+15°F (+8.3°C) .......... 170 B-36 LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=400 psi/in (106.8 kPa/mm), and AT=-15°F (-8.3°C) ............ 170 8-37 LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=14” (356 mm), k=100 psi/In (27.1 kPa/mm), and AT=0°F (0°C) ................... 171 IlGlRI. I38 tram” (356mm).k 339 HR. nitth (Sionnll M lTEznith Re (356mm Re; 541 iii} mil“. Res (356 mm). It: 54. Ill, nith Res: (356 turn). l=2 34‘ k; UL; mil Res (356 mm). H M LTE-1 “Rh Res (356 mm]. litsi H) UT,- “lih Res ‘mjllCal Deli 547 . Wines] Dete ”E; [of h=6“ FIGURE 3-38 3-39 8-40 B-41 B-42 B-43 B-44 B-45 B-46 B-47 LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=14” (356 mm), k=100 psi/m (27.1 kPa/mm), and AT=+15°F (+8.3°C) ............ 171 LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=14” (356 mm), k=100 psi/In (27.1 kPa/mm), and AT=~15°F (-8.3°C) ............. 172 LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=14” (3 56 mm), F250 psi/tn (67.9 kPa/mm), and AT=O°F (0°C) ................... 172 LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=14” (356 mm), k=250 psi/In (67 .9 kPa/mm), and AT=+15°F (+8.3°C) ............ 173 LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=14” (356 mm), k=250 psi/1n (67 .9 kPa/rnm), and AT=-15°F (-8.3°C) ............. 173 LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=14” (3 56 mm), k=400 psi/1n (106.8 kPa/mm), and AT=0°F (0°C) .................. 174 LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=14” (356 mm), k=400 psi/in (106.8 kPa/mm), and AT=+15°F (+8.3°C) .......... 174 LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=l4” (356 mm), k=400 psi/In (106.8 kPa/mm), and AT=-15°F (-8.3°C) ............ 175 Analytical Determination of LTEa After Dowel Bar Retrofit from Initial LTEs for h=6” (152 mm) and AT=0°F (0°C) ..................................... 175 Analytical Determination of LTEs After Dowel Bar Retrofit from Initial LTE; for h=6” (152 mm) and AT=15°F (83°C) ................................. 176 Page IlG'tRE B48 Bis git! RSI set 353 iii-4 356 Analytical 0 UL for h' 1 atoms Dc lTE forh S : Anahtrcal De life for hi? Anahtica' De ”£5 for h=1 Analytical D LTE: for h: Warm C LTE; for h: Wines I LTE: f0T h: FIGURE Page 8-48 Analytical Determination of LTE5 After Dowel Bar Retrofit from Initial LTEa for h=6” (152 mm) and AT=-15°F (-8.3°C) ............................... 176 B-49 Analytical Determination of LTEs After Dowel Bar Retrofit from Initial LTE5 for h=8” (203 mm) and AT=0°F (0°C) ..................................... 177 B-50 Analytical Determination of LTE5 After Dowel Bar Retrofit from Initial LTEa for h=8” (203 mm) and AT=15°F (83°C) ................................. 177 B-Sl Analytical Determination of LTEa After Dowel Bar Retrofit from Initial LTEs for h=8” (203 mm) and AT=—15°F (-8.3°C) ............................... 178 8-52 Analytical Determination of LTEs After Dowel Bar Retrofit fi'om Initial LTE5 for h=10” (254 mm) and AT=0°F (0°C) .................................... 178 B-53 Analytical Determination of LTE5 After Dowel Bar Retrofit from Initial LTE5 for h=10” (254 mm) and AT=15°F (83°C) ............................... 179 B-54 Analytical Determination of LTEa After Dowel Bar Retrofit from Initial LTEs for h=10” (254 mm) and AT=~15°F (-8.3°C) .............................. 179 8-55 Analytical Determination of LTE5 After Dowel Bar Retrofit from Initial LTEs for h=12” (305 mm) and AT=O°F (0°C) .................................... 180 B-56 Analytical Determination of LTEs After Dowel Bar Retrofit from Initial LTE5 for h=12” (305 mm) and AT=15°F (83°C) ............................... 180 B-57 Analytical Determination of LTE5 After Dowel Bar Retrofit fi'om Initial LTEa for h=12” (305 mm) and AT=-15°F (-8.3°C) .............................. 181 IlC'lRI is initial 0‘ [TB for h=T 550 Mistral Dr . I ”E! {OT h: ‘ ' B-ol Analytical De'. lTE for h?- I ~' B-ri Principal Tensi i=6“ (152 tn'r. i=1: Pincipal Tensi it“ ll 52 nm: 5.5.3 Principal Tens: eerie -. IILT. til Ptgncipal Tens: eerie .. mm 5.5 ' Pfineipal Tens u. mt" ’41 -fi mn- «35 . [moan . 5 hi? (1s; FIGURE Page B-58 Analytical Determination of LTE5 After Dowel Bar Retrofit from Initial LTE5 for h=14” (3 56 mm) and AT=O°F (0°C) .................................... 181 B-59 Analytical Determination of LTEs After Dowel Bar Retrofit from Initial LTEa for h=14” (356 mm) and AT=15°F (83°C) ............................... 182 B-6O Analytical Determination of LTE5 After Dowel Bar Retrofit from Initial LTEs for h=14” (356 mm) and AT=-15°F (-8.3°C) .............................. 182 8-61 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=100 psi/in (27.1 kPa/mm), and AT=0°F (0°C) ............ 183 8-62 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=100 psi/1n (27.1 kPa/mm), and AT=+15°F (+8.3°C). . . .. 183 B-63 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=100 psi/in (27.1 kPa/mm), and AT=-15°F (-8.3°C) ...... 184 B-64 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=250 psi/in (67.9 kPa/mm), and AT=0°F (0°C) ............ 184 B-65 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=250 psi/in (67.9 kPa/mm), and AT=+15°F (+8.3°C). . . .. 185 B-66 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=250 psi/In (67.9 kPa/mm), and AT=-15°F (-8.3°C) ...... 185 8-67 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=400 psi/in (106.8 kPa/mm), and AT=0°F (0°C) ........... 186 B-68 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=400 psi/1n (106.8 kPa/mm), and AT=+15°F (+8.3°C). .. 186 llC'lRI all Principal Ter‘ H1152 m It Printipal Tens H" 1393 Int" l'l Paternal Tens- tel” (203 min I”. Pnneipal Tens h=8"(203 rm I? Empal Tens “5 (:03 In; I’d Principal Ten H“ (:03 m 3.5 Pfimpd lei FIGURE B-69 B-70 B-71 B-72 B-73 B-74 B-75 B-76 B-77 B-78 B-79 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=400 psi/in (106.8 kPa/mm), and AT=-15°F (-8.3°C). . . .. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=100 psi/in (27.1 kPa/mm), and AT=0°F (0°C) ........... Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=100 psi/tn (27.1 kPa/mm), and AT=+15°F (+8.3°C). . . .. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=100 psi/in (27.1 kPa/mm), and AT=—15°F (-8.3°C) ..... Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=250 psi/in (67.9 kPa/mm), and AT=0°F (0°C) ........... Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=250 psi/in (67.9 kPa/mm), and AT=+15°F (+8.3°C). .. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=250 psifrn (67.9 kPa/mm), and AT=-15°F (-8.3°C) ..... Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=400 psi/in (106.8 kPa/mm), and AT=0°F (0°C) ......... Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=400 psi/In (106.8 kPa/mm), and AT=+15°F (+8.3°C). .. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=400 psi/1n (106.8 kPa/mm), and AT=-15°F (-8.3°C). . .. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=100 psi/1n (27.1 kPa/mm), and AT=0°F (0°C) ......... xxii Page 187 .187 188 .188 . 189 189 . 190 .. 190 191 .191 ..192 illilRI R80 Principal Te: lFlO" 125-l T“ lit Principal Ten h=10"125~l Ir ll: Principal Ten: h=10"1254 m l3} Principal Tens 11:10“ 1254 It. 3’34 Plifldpal Ten: 11:10’1254 n Ml Principal Ten MO“ 1254 n M Pump“ Ten P10“ (254 1 All Pump“ TeI h‘10"1‘254 1 tag . . Princrpa] Te: FIGURE Page 8-80 B-81 3-82 8-83 B-84 B-85 B-86 8-87 8-88 8-89 B-9O Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=100 psi/in (27.1 kPa/mm), and AT=+15°F (+8.3°C). .. 192 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=100 psi/In (27.1 kPa/mm), and AT=-15°F (-8.3°C). . . .. 193 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=250 psi/in (67.9 kPa/mm), and AT=0°F (0°C) ........... 193 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=250 psi/In (67.9 kPa/mm), and AT=+15°F (+8.3°C). .. 194 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=250 psi/in (67.9 kPa/mm), and AT=-15°F (-8.3°C). . . .. 194 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=400 psi/tn (106.8 kPa/mm), and AT=0°F (0°C) ......... 195 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=400 psi/In (106.8 kPa/mm), and AT=+15°F (+8.3°C)... 195 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=400 psi/1n (106.8 kPa/mm), and AT=—15°F (-8.3°C). .. 196 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=100 psi/in (27.1 kPa/mm), and AT=0°F (0°C) ........... 196 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=100 psi/tn (27.1 kPa/mm), and AT=+15°F (+8.3°C). .. 197 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=100 psi/in (27.1 kPa/mm), and AT=-15°F (-8.3°C). . . .. 197 xxiii IlGlRI Ell Pnncipal Te: l’FlZ"(305 fl ll: Principal Ter hill“ 1.335 7 iii Pnncipal Ten _ 1"r H ll. 13mm iii Pnncipal Tens h=l3" 1305 It: iii Principal Tens lFlf (305 In “A Principal Tern l’FlTBGS m 3;? Principal Tens W (356 m HS Prim-Pal Tens FIGURE 3-91 8-92 B-93 3—94 8-95 B-96 B-97 B-98 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=250 psi/m (67.9 kPa/mm), and AT=0°F (0°C) ........... 198 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=250 psi/in (67.9 kPa/mm), and AT=+15°F (+8.3°C). .. 19s Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=250 psi/1n (67.9 kPa/mm), and AT=—15°F (-8.3°C). . . .. 199 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=400 psi/tn (106.8 kPa/mm), and AT=0°F (0°C) ......... 199 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=400 psifrn (106.8 kPa/mm), and AT=+15°F (+8.3°C). 200 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=400 psi/1n (106.8 kPa/mm), and AT=~15°F (-8.3°C). .. 200 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (356 mm), k=100 psi/1n (27.1 kPa/mm), and AT=0°F (0°C) ........... 20] Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (356 mm), k=100 psi/In (27.1 kPa/mm), and AT=+15°F (+8.3°C). .. 20] Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (356 mm), k=100 psi/In (27.1 kPa/mm), and AT=~15°F (-8.3°C). . . .. 202 B-lOO Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (3 56 mm), k=250 psi/in (67.9 kPa/mm), and AT=0°F (0°C) ........... 202 B-101 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (356 mm), k=250 psi/In (67.9 kPa/mm), and AT=+15°F (+8.3°C). .. 203 xxiv Page nG'tllf an: Internal I" will“ lint» Pinata Tel :14“ (356 if its Principal W p14" 1M 11 E1 Sucgade Si 1.151“W R I.” Subfu'ade S l. H mm! l 5.3 SubETadeS (:34 mm’ in Subgrades 1315. W“ {.5 Suhgade‘ (ass amt Es. . Suhgrad (152 rm :3 Subgrade (23.) It!” FIGURE Page 8-102 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (356 mm), k=250 psi/in (67.9 kPa/mm), and AT=-15°F (-8.3°C). . . .. 203 B-103 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (356 mm), k=400 psi/in (106.8 kPa/mm), and AT=0°F (0°C) ......... 204 B-104 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (356 mm), k=400 psifrn (106.8 kPa/mm), and AT=+15°F (+8.3°C). 204 B-105 Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (356 mm), k=400 psi/1n (106.8 kPa/mm), and AT=-15°F (-8.3°C). .. 205 El Subgrade Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle on a 6” (152 mm) RCC Pavement Before and After Retrofit Dowels ............... 233 E-2 Subgrade Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle on an 8” (203 mm) RCC Pavement Before and After Retrofit Dowels ............... 233 E3 Subgrade Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle on a 10” (254 mm) RCC Pavement Before and After Retrofit Dowels ............... 234 E-4 Subgrade Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle on a 12” (305 mm) RCC Pavement Before and After Retrofit Dowels. .. . . . . .. 234 E-5 Subgrade Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle on a 14” (3 56 mm) RCC Pavement Before and After Retrofit Dowels ............... 235 E-6 Subgrade Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle on a 6” (152 mm) RCC Pavement Before and After Retrofit Dowels ............... 235 E-7 Subgrade Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle on an 8” (203 mm) RCC Pavement Before and After Retrofit Dowels ............... 236 HERE E3 E-9 {-10 E-ll E-IS Subgrade (254 mm - Subgrade I (305 mm» Subgrade S (356 mm Subgrade Subgradg (:93 mm Subgadt— FIGURE E-8 E-9 E-lO E-ll E-12 E-13 E-l4 E-15 E-16 E-17 E-18 Subgrade Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle on a 10” (254 mm) RCC Pavement Before and After Retrofit Dowels ............... 236 Subgrade Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle on a 12” (3 05 mm) RCC Pavement Before and After Retrofit Dowels ............... 237 Subgrade Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle on a 14” (3 56 mm) RCC Pavement Before and After Retrofit Dowels ............... 237 Subgrade Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle on a 6” (152 mm) RCC Pavement Before and After Retrofit Dowels ............... 238 Subgrade Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle on an 8” (203 mm) RCC Pavement Before and After Retrofit Dowels ............... 238 Subgrade Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle on a 10” (254 mm) RCC Pavement Before and After Retrofit Dowels ............... 239 Subgrade Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle on a 12” (305 mm) RCC Pavement Before and After Retrofit Dowels ............... 239 Subgrade Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle on a 14” (3 56 mm) RCC Pavement Before and After Retrofit Dowels ............... 240 Subgrade Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle on a 6” (152 mm) RCC Pavement Before and After Retrofit Dowels ............... 240 Subgrade Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle on an 8” (203 mm) RCC Pavement Before and After Retrofit Dowels ............... 241 Subgrade Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle on a 10” (254 mm) RCC Pavement Before and After Retrofit Dowels ............... 241 Page HGLRE E49 E-Zl E-II E-ZS Subgrade S (305 mm P Subgrade S: (356 mm» R Subgrade S: llSZmnHRt Subgrade SI: (:03 1mm R1 “ 5115873536 St: (254 mm , R1 S”bgfade Sm (305 mm) It SUbiEade Str' 1356mm, M FIGURE E-19 Subgrade Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle on a 12” E-ZO E-21 E-ZZ E-23 E-24 E—25 (305 mm) RCC Pavement Before and After Retrofit Dowels ............... 242 Subgrade Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle on a 14” (3 56 mm) RCC Pavement Before and After Retrofit Dowels ............... 242 Subgrade Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle on a 6” (152 mm) RCC Pavement Before and After Retrofit Dowels ............... 243 Subgrade Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle on an 8” (203 mm) RCC Pavement Before and After Retrofit Dowels ............... 243 Subgrade Stresses for a 36 kip (160.1 kN) Dual—Tired Tandem Axle on a 10” (254 mm) RCC Pavement Before and After Retrofit Dowels ............... 244 Subgrade Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle on a 12” (305 mm) RCC Pavement Before and After Retrofit Dowels ............... 244 Subgrade Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle on a 14” (3 56 mm) RCC Pavement Before and After Retrofit Dowels ............... 245 xxvii Page PROBLEM STA' During the m uere not cut lemme instabli: {$23910 24 ml! extremely flick sect; These large < ides ofthe crack 5 resses W11 be incu: ifsi'gred for due to 113 be developed I ”imam is the ne - CHAPTER I - Introduction PROBLEM STATEMENT During the early years of roller-compacted concrete (RCC) pavements, engineered joints were not cut in order to save on initial construction and sealant maintenance costs. Volumetric instability along with curling resulted in crack spacing ranging from 30 to 80 feet (9 to 24 m) for conventional RCC thicknesses and up to 450 feet (137 m) in extremely thick sections. These large crack spacings resulted in wide crack openings. These large crack widths are not conducive to high load transfer between the two sides of the crack. Since the two sides of the crack may act as different slabs, higher stresses will be incurred. This will bring the pavement to failure at an earlier time than designed for due to fatigue of the concrete. To combat this problem, design strategies must be developed to ensure a long-temr, cost-effective pavement. An additional issue of importance is the need to assess when and how to repair pavements with low load transfer capacities. OBJECTIVES OF RESEARCH This project will aim to study load stresses for in-service RCC pavements through non-destructive deflection data and finite element analysis. Alternatives (such as increased RCC design thickness, engineered joint spacing at closer distances than the natural crack spacing, or dowel bar retrofitting of natural cracks) will be examined in order to produce at eff-cent desrir beliscussed and : ll- the ICC UT should prove RCC Ila: can been leg. :ensfieration of R1 loci COllC’ClOI. an; Sf OPE 0f RESP..- The reeomm failevtng research r made a better und Task 1 ~ Re design proee “id:- RCC pr Task 2 ~ G; Smhesize tl elm mod; deflemon lc rih'tge of Va; Ink 3 ~ Fr PTOVzde rye; 00‘» - “WSW an efficient design based on cost. Options for in-service RCC pavement rehabilitation will be discussed and future RCC pavement design recommendations will be made. Ifthe recommendations from this project are utilized, the findings from this study should prove RCC pavements are a reasonable long-term alternative for low volume areas that carry heavy loads. A secondary consideration of this project is to also promote the consideration of RCC pavement construction for other firnctions, such as subdivision, local, collector, and arterial roads. SCOPE OF RESEARCH The recommendations from this project will be made from the findings of the following research plan. The research plan has been divided into six “tasks” in order to provide a better understanding of the different phases of this project Task 1 — Review and synthesize national and international literature as well as design procedures for RCC in order to characterize known and potential problems with RCC pavements. Task 2 — Gather FWD and HWD data from existing RCC pavement sites and synthesize this data to backcalculate pavement structural parameters (such as elastic modulus, modulus of subgrade reaction, radius of relative stiffness, deflection load transfer efficiency, etc.). This will be done in order to describe the range of values that are typical for RCC pavements. Task 3 — From the existing sites found in Task 2, an attempt will be made to provide trends with statistical certainties of the following parameters in comparison to deflection load transfer efficiency for each test section: Task 4 - D fadgue 1613* alternarhes do“ el bfll 1’6 benefit ol‘loa induced $1195 Task 5 - De Existing RU [ONTINTS OF T This thesis c Emits a discus E12 m; and a air Reach. A more d rt)“ ' «her ll Includes man of the ma‘ the, Crack width or opening Crack or Joint spacing Elastic modulus of the RCC Modulus of subgrade reaction Radius of relative stiffness Task 4 — Develop a design scheme for new RCC pavement construction based on fatigue relationships found in Task 1 and inventory data from Task 2 including alternatives such as increased thickness design, joint cutting and maintenance, and dowel bar retrofitting of naturally cracked pavements. It will incorporate the benefit of load transfer at joints or cracks as well as the effect of temperature- induced stresses, which is not included in most design methods. Task 5 - Develop rehabilitation strategies towards extending the service life of existing RCC pavements found in Task 2. CONTENTS OF THESIS This thesis contains background on information roller-compacted concrete pavements, a discussion of the analyses performed and the results obtained fi'om data in this study, and a summary of the conclusions and recommendations derived from this research. A more detailed breakdown of the contents of individual chapters follows. Chapter 11 includes: background on the construction practices of RCC pavements, a summary of the material properties of RCC; a review of the most common structural design methods of RCC pavements; a review of the field performance of in—service RCC pavements; and, a description of the future application where RCC pavements are to be utilized. Chapter 111 provides a description of the test sites and data collected for this study. It also discusses the trends found between load transfer efficiency and many variables from these test sites. A review of a new RCC pavement structural design method is raven in rethod to existing memes of RC ( dearer also cites a | r..ebilrta:ron m‘tem. :fiw‘ 9h) 3 ll! - uh“ L. C URUAI' 56) 0 edited to the load 1 n Li‘k \ a - Mam. konelUdE) och method is given in Chapter IV. Also included in Chapter IV is a comparison of this new method to existing design methods. Chapter V discusses design and rehabilitation alternatives of RCC pavements to insure the design life of an RCC pavement. This chapter also gives a detailed methodology as to the timing of dowel bar retrofitting as a rehabilitation alternative. A summary of the conclusions and recommendations arising from the analyses performed in this study as well as a listing of future research needs related to the load transfer efficiency in RCC pavements are contained in Chapter VI, which concludes this thesis. CONSTRl‘C Tl lllston .chng years ’he idea r 1"" Re CD, 3e»- - CHAPTER II - Literature Review CONSTRUCTION PRACTICES History Although the concept of soil-cement and lean concrete has been in use for over 50 years, the idea of using higher-quality aggregates to form a high-strength concrete is relatively new. Roller-compacted concrete combines the practicality of asphalt paving procedures with the durability of portland cement concrete pavement. The ability of roller-compacted concrete to carry a large of stress placed on it by an external load as a surface layer is what separates it from normal soil-cement. In this respect, RCC pavements are very similar to conventional concrete pavements. The earliest reported use of roller-compacted concrete was as an Australian pavement in 1932 as well as in a Yakima, Washington airport pavement in 1941. Another early example of RCC was the construction of two fieight yards in Hokkaido, Japan in 1956 [1]. Roller-compacted concrete was used for mass concrete dams in both Pakistan and Oregon in the early 19705. The no-slump mixture was placed in lifts and compacted, resulting in a savings of about one-third over the costs of using conventional concrete materials and techniques. These projects demonstrated the economic advantages of this construction technique which could easily be adapted to pavement construction [2]. RCC pavements in North America were constructed on a larger scale soon after the mass concrete dams in the mid-19705. The US. Army Corps of Engineers Waterways EW’ 1135 :Wnl h; rm on RC C an. RCC rexements {I ln 1970. R sormg mid storage sadmahle and ecc | humerus has them. he forear) industr) [radian watem a} that conditio rt» ln luh of ‘1‘ We RCC an arm 3515 and other an T35 ahffnath‘e reg 30mm Me Pmemen‘ toS‘Cttchon as ho meap} “Err T“ paVelher Waterways Experiment Station (WES) built two test sections in Vicksburg. Mississippi. This experiment had two purposes. The first examined the effect of different vibratory rollers on RCC and the second studied the consequences of using marginal materials in RCC pavements [2]. In 1976, RCC pavements were constructed as a low—cost alternative for use as log sorting and storage yards in British Columbia, Canada [1]. The birth of RCC pavements as a durable and economical alternative to both conventional concrete and asphalt pavements has been attributed to this project. Due to a governmental regulation forcing the forestry industry to sort logs on land thereby reducing wood debris contamination in Canadian waterways, over 28 acres of RCC sorting yards has been built in the harsh climatic conditions of British Columbia as of 1983 [2]. In July of 1984, the US. Army Corps of Engineers constructed the first large- scale RCC pavement in the United States. A 20,000 yd2 (16,700 m2) parking area for tanks and other armored vehicles was built at Fort Hood in Texas over an 11-day period. This alternative resulted in a savings of 15 percent over a conventional reinforced concrete pavement [2]. This project was the beginning of an increase in RCC pavement construction as both military and civilian projects in the United States. Underlying Layer Construction From a physical standpoint, a RCC pavement is very similar to a conventional concrete pavement. Knowing this, a RCC pavement subgrade and base are designed and constructed to fimction in the same way as a normal PCC pavement would [2]. ln frost susce hrnctions in consider snort so that the c taming and rollers ] The Camdri Elng ofthe suhgra limit} in accordmc should he rtpaned z 9m: should re ofh'ater from the p lest Sections For larger 1 ‘0‘ the contractor . {flare [he 90mm m one month in 38' -. ~ Qflrw .bmtmn Shin 0 ~ c In frost susceptible areas, the base course should be designed with drainage functions in consideration. As the base is constructed, it should also provide enough support so that the over-lying RCC layer can be fully consolidated when compacted with tamping and rollers [2, 3]. The Canadian Portland Cement Association suggests that after excavation or filling of the subgrade, it should be compacted to a minimum of 95% Modified Proctor density in accordance with ASTM D1557. After this is done, the disturbed subgrade should be repaired and debris removed fi'om the surface. Finally, the surface of the subgrade should be moistened without ponding or creating mud to minimize absorption of water from the RCC mixture [4]. Test Sections For larger RCC pavement constructions sites, test sections can be utilized in order for the contractor to address mixing, transporting, placing, compaction, and curing issues before the construction of the actual pavement. These test sites should be constructed at least one month in advance in order to strength test samples can be taken directly from the RCC. It is suggested that both cores and beams be sawed from the existing RCC after 28 days in order to determine the correlation between the flexural strength and splitting tensile strength of that particular RCC mixture [2]. Samples from the actual construction site are often tedious to collect and do not provide the immediate feedback needed to make adjustments on-site [5]. Rolling patterns should be tested in order to test for the procedure which will result in optimum density with the least amount of passes. After the site is to achieve suco The mo. cold in tnts shot order to attain [I]. From this A recommend :- rx ‘1 After the site is assessed from these perspectives, techniques and designs can be modified to achieve successful construction. The most critical component of a RCC pavement is the joint system Fresh and cold joints should be constructed in both the transverse and longitudinal directions in order to attain smooth surface transition between the slabs as well as consistent densities [2]. From this test section, typical transverse cracking patterns can be observed as well. A recommended layout of a RCC test section can be viewed in Figure 1. Fresh joint DAY ONE Three 12 to 14 foot DAY TWO Widc T lanes Fresh jornt Cold jornt * 150 feet I foot=0.305 m Figure 1. Recommended Test Section Setup (After Pittman [2]). Production and Transportation Since RCC has relatively little water in the mixture, it is typically mixed with a twin-shaft pugrnill mixer. This apparatus is commonly used in asphalt concrete mixing and can be viewed in Figure 2. Both continuous mixing and weigh-batching plants can be utilized to mix RCC batches. For larger jobs, the continuous mixing plant is preferred due to its larger capacity, simpler transportation of the plant, and set-up time. The advantage of the weigh-batching plant is the superior control over the proportions of each inditidual hatch i would need to de individual batch in comparison to a continuous mixing operation [2]. The contractor would need to decide which plant better suits the needs of a particular project. Figure 2. Twin-Shaft Pugmill Mixing Plant. The mixing plant should be located as close to the construction site as possible to minimize the hauling time and reduce uncontrollable delays. Dump trucks are typically used to haul the RCC mix from the plant to the paving site. If environmental concerns are prevalent, the dump trucks should be equipped with protective covers to reduce effects of precipitation or extreme cold or heat. When the trucks arrive on site, the concrete is dumped directly into the paver’s hopper as seen in Figure 3 [2]. If RCC material is transported using ready-mix trucks, the drum should be in good condition and as clean as possible in order to aid in the flow of the stiff mixture. Using these trucks will slow the construction process regardless of the condition of the drums however and are not recommended in most cases [6]. Figure 3. Hopper of RCC/Asphalt Paver. Paving Operations RCC pavements are placed with asphalt pavers in most instances. The same technology that allows asphalt pavers to control grade and depth of asphalt pavements (using traveling skis or electronic string lines) monitors the same properties of RCC pavements. Most RCC is placed using a paver employing a vibrating screed and tamping bar as seen in Figure 4 [2]. This allows the paver to achieve compaction levels up to 90% Modified Proctor after it is initially placed and before it is rolled as tested in Figure 5. Some adjustments are usually needed to handle RCC on asphalt pavers. Feeding gates between the hopper and screed need to be enlarged in order to accommodate the volume of RCC that is placed at a given time. Spreading screws also need to be adjusted in fiont of the screed to promote uniformity throughout the entire width of paving [2]. RCC should be placed as soon as possrble after the initial addition of water. To form a fresh joint and assure good bond between two slabs of concrete, two adjacent hemmw 1 twmnmg m1 .IH Tarn"? rttitre 4. in ——-—-.--~.-—— I lanes need to be placed within 60 minutes of each other. The use of two pavers working together can greatly minimize the chances of cold joints forming in RCC pavements [2]. , 4 "' ’;—' ~,..-:'-~: y- " A! ' I; .‘ A. ’37 . -f . '7 . a. - . - w. _ . f \_ FL”':§..1—_.« . . '3'” I" :2. r4 , ' ,. , ~ ~ .99“ ..£I ,. , .30? ' a» . ' ., . "1"] '.._ 3 "f“ £.;,p {finhl‘ " a in?» 4"} ' Figure 5. RCC Being Placed and Tested for Density Before Rolling. ll Compaction After pl rollers as seen i .‘39 t\'t dual-d1 gases ot‘a 20-1 closure of surfs roller marks \s‘l that in Spring htgtudiral col med the roll erder to rnimmi “£11116, Dual h '5 recc Lift ‘~ before {he Compaction After placement using an asphalt paver, RCC pavements are compacted using rollers as seen in Figure 6. This is best achieved by making several passes of a lO-ton (89 kN) dual-drum vibratory roller in most cases. Often, this is followed by two or more passes of a 20-ton (178 kN) rubber tired roller for increased smoothness. This will aid in closure of surface voids and cracks. A dual-drum static roller can be used to remove any roller marks which may be left by the first two rollers if necessary [2]. At a Saturn auto plant in Spring Hill, Tennessee, a two roller method was implemented to minimize longitudinal cold joints between lanes with much success [7]. If a single roller technique is used, the rollers should be used in the direction parallel to the shortest dimension in order to minimize the length of possible cold joints [8]. firm \'\ Figure 6. Dual-Drum Roller Compacting Freshly Laid RCC. It is recommended that no more than 10 minutes elapse after the placement of the RCC before the rolling commences. The rolling should be finished within 45 minutes of 12 addition of v order to atta Cunng RCC tang ot‘th sortietmes t hows using Aft: mother addition of water at the RCC plant if possible [2]. Enough passes should be made in order to attain the necessary density required (typically 98% Modified Proctor). Curing RCC mixtures typically have a very low water content. To prevent drying and scaling of the fi'eshly laid surface, a combination of moist curing and membrane curing is sometimes utilized. Freshly compacted RCC should be kept moist for a minimum of 24 hours using any of a variety of methods [2]. After the initial 24 hour curing period, RCC pavements should be covering using any of the following methods: water curing spray, wet sand, wet burlap, or a membrane- forming curing compon [2,6]. If a water spray method is employed, carefirl attention must be made in order to prevent the washing on fines on the surface. All traffic (with the exception of curing trucks) should be kept off the new RCC pavement until 14 days for best results [2]. Contraction Joints and Load Transfer Devices Since RCC pavements are an economical alternative to both deep-strength asphalt pavements and conventional concrete pavements, contraction joints are often not sawed. Most projects have allowed cracks to naturally form and in many cases, no distress has been noted at these cracks. These shrinkage cracks typically form at spacings greater than that of conventional concrete pavements (between 40 to 60 feet, or 12.1 to 18.3 m) [8]. DuringRCC pavement construction at Fort Hood (Texas) and Fort Lewis (Washington), an attempt to saw joints in RCC pavements produced a ragged edge due to the saw blade knocking off pieces of cement paste and aggregate [2]. 13 then cracks or him entering the base ‘ Hotteter then cracks . option in many cases do Special attention contraction joints is atte as hgh temperatures. lc paced RCC. This sens contraction joints [6]. In the initial C0 Pars or keyed joints m concrei When cracks are allowed to form naturally, sealant can be used to prevent water from entering the base when fieeze-thaw is a concern or when drainage is a problem [9]. However when cracks are allowed to form naturally, sealant has been discarded as an option in many cases due to aesthetic and economic reasons. Special attention needs to be paid with respect to timing if the sawing of contraction joints is attempted. With such a low water content to start with, factors such as high temperatures, low humidity, and wind can greatly affect the hydration of freshly placed RCC. This sensitivity can seriously affect the correct timing of sawing of the contraction joints [6]. In the initial construction of RCC pavements, load transfer devices such as dowel bars or keyed joints have been very limited in use. The stiff consistency and placing operations of RCC pavements make dowel placement difficult and arduous in comparison to conventional concrete dowel placement. At this time, dowels have been either driven into the RCC pavement before final set or carefully placed and worked around with fresh RCC covering the dowels bars before rolling [2]. Dowel bar retrofitting is a viable option for placement after load transfer has diminished past a sought after threshold as is done in conventional concrete pavements. RCC MATERIAL PROPERTIES Overview While the constituents of roller-compacted concrete are the same as those used in conventional concrete, the difference between the two is the proportions of these 14 ingredients. Norn resembles that of. ASTM Di (crest Mixtures l's'ng \lodifted l M CASES. the n HMVU. both c mm motstur Ward 0f mod ml} Shaped b} will While F Wham am Wuhan! m! '50”!!th MI w (M I: ingredients. Normal RCC is very dry and stiff mix with no slump properties. Its surface resembles that of a gray asphalt pavement as seen in Figures 7 and 8. ASTM D558 (Standard Test Methods for Moisture-Density Relations of Soil- Cement Mixtures) and ASTM D1557 (Laboratory Compaction Characteristics of Soil Using Modified Effort) treat RCC like a soil-cement instead of conventional concrete. In most cases, the modified Proctor method is used in assessing compaction efforts on RCC. However, both of these methods treat RCC like a soil with an established relationship between moisture and density for a given compactive effort and measurable using standard or modified Proctor compaction testing [10]. When fieshly mixed, RCC can be easily shaped by hand. Conventional concrete relies solely on chemical reactions to gain strength while RCC requires both from hydration as well as mechanical force to achieve a significant amount of strength. as! gay rm H a: ‘ (l V ‘0 ' . - ”use! . Jar-Q em .- -' 05,4, 2";- -22.. 43.1%,“? a): _ . ‘ 43 ‘ m ., " ‘ I v ". F ‘7 ‘1 2 d' ‘ (- ’0‘ “5,1411, 473“” - H-- r i C , . . , -_ .li “ha-h. * ,4 r' Figure 7. Close-Up View of RCC Surface (After One Day). Figure 8. Far View of RCC Surface (After One Day). RCC is generally not air-entrained, has lower water and paste contents, and higher fine aggregate volumes in comparison to conventional concretes as can be seen in Figure 9. RCC must be dry enough to hold the weight of a heavy roller, yet wet enough to assure distribution of the water available throughout the mixing and placing process [1 1]. Materiak Basic RCC is comprised of water, cement, and aggregates. Admixtures, fly ash, and fibers can also be found in some RCC mixes. Aggregates Aggregates normally comprise 75 to 85 percent of the volume of RCC mixtures. Because of this, the volume highly influences fi'esh RCC properties such as workability, segregation potential, and ability to consolidate properly. The volume and quality of the 16 and fatigue trend BOlll fine meshed graxe combination it it seem to hold me hound strength ntersntmost c Q“ aggregates have an effect on the strength, deformation characteristics, thermal properties. and fatigue trend [11]. Both fine and coarse aggregates are used in RCC mixtures. Both crushed and uncrushed gravel as well as crushed stone have been used in RCC projects in combination with both natural and manufactured sand [11]. Crushed stone and gravel seem to hold mortar better than natural aggregates in RCC and usually provides a higher flexural strength. However, crushed aggregate requires more compactive effort with rollers in most cases in comparison with natural stones [1 1]. “'1'.” .J‘r n . 4“ ." . -— s" e‘ .I met, "1}“; . of Sr ffggn '0’ . f’afl ‘4‘ ' _ ‘ Figure 9. Close-Up View of Fresh RCC. While greater economy is usually associated with a larger nominal maximum size aggregate due to the lesser cement paste content, segregation can become a problem It is recommended that the nominal maximum size aggregate not exceed 3/4 inch (19 mm) [1]]. Since the 5 [33:7 aggregates Fine aggr he total aggrega (emeutitiow‘ if. Man} dit These include pt WHO-huts su. hide I or ll por RCC [13h mitt. l" ., t Winkle [0 [ will be fO‘dc in user COHIen‘, In RC< tie-w . QR} anew l‘ggm Ergo [11]. Since the surface of RCC pavement can wear down as much as ‘A inch (6 mm). larger aggregates become exposed and provide a less pleasing look aesthetically. Fine aggregate aids in the compaction process of RCC. As much as 14 percent of the total aggregate material can pass through the #200 sieve [12]. C ementitious Materials Many different types of cementitious materials have been used in RCC mixtures. These include portland and blended hydraulic cements, ground blast firmace slag, as well as pozzolans such as fly ash. The most typical cementitious materials used have been Type I or II portland cement (ranging from 3.5 to 6 bags/yd3, or 4.6 to 7.8 bags/m3, of RCC [12]) with 15 to 20 percent mixtures of Class C (binder) or F (filler) fly ash by volume. The fly ash provides additional fine material not found in many standard fine aggregate gradations to aid in compaction [11]. It is recommended that the entire cementitious content be between 12 to 14 percent of the aggregate content by weight [10]. As in conventional concrete, the design strength is the driving factor to determine the amount of cementitious materials in an RCC mixture. Since aggregates are still susceptible to chemical reactions like alkali reactivity, conventional concrete practice should be followed when setting limits on the content of cementitious materials [1 1]. Water Content In RCC, the amount of water is proportioned to achieve no slump and maximum density after compaction. In most cases, the water-cement ratio is between 0.3 and 0.5 [10] and in some cases as high as 0.6 [14]. 18 immature! It is difficult ti 3mm \Vatero‘a}'$ in air by using mm mggested for cont emt Most RC C pm some damage to sweatshirt of the Rt ti}. it is difficult for comment-e eitort als tsfieezethaw dural Water-reduc 0i? in m sections ’9ng agents to l iggregate passing t' A dmixtures It is diflicult to entrain air in a RCC mixture. Research at the US. Army Corps of Engineers Waterways Experiment Station has showed that RCC mixes can be entrained with air by using normal air-entraining admixtures at dosages 5-10 times greater than suggested for conventional concrete [11]. Most RCC pavements that are located in freeze-thaw susceptible zones have minimized damage by using low water-cementitious ratios, thereby reducing the permeability of the RCC pavement. It has been shown that once RCC has hydrating fully, it is difficult for outside moisture to enter the RCC layer of the pavement. The compactive efion also aids in reducing the permeability of the pavement, which increases its freeze-thaw durability without air-entrainment [11]. Water-reducing admixtures have had limited use in RCC pavements and usually only in test sections and research studies. It appears that the ability for the water- reducing agents to lower water contents is dependent on the type and volume of aggregate passing the #200 sieve [11]. Set retarding admixtures have been used and are sometimes crucial in allowing for greater time for compaction as well as improving the bond between lanes that would otherwise form a cold joint [11]. Fibers Research has been done by Nanni [13] showing the benefits of using steel fibers in RCC pavements. Most RCC pavements are allowed to crack naturally. These cracks form at larger spacings than in conventional concrete. However, the larger cracks 19 stating usuall} aggregate inter mmafim This increase it ttcreasing fatfg HIOUl'ki 25° 0 C8 menu and t that. RC( l3 5‘5 as much Whittle reduct RCC into one MeClinical p 7551 d; his ‘0 that 0‘: Commilon‘r “55 in the R- beamg lasers fishes (305 to moflhe; 'k teem“ spacing usually lead to large crack openings, thereby reducing load transfer from aggregate interlock. When large diameter steel fibers are included in the RCC mix, they can act as a “micro-dowel”, increasing the load transfer by keeping the cracks tighter. This increase in load transfer will benefit the RCC pavement by lowering stresses and increasing fatigue life. His research has shown that pavement thickness reductions around 25% can be utilized with proper addition of steel fibers. This will save money in materials and labor. Savings can be realized from the reduction in RCC that is produced. Typically, RCC is laid in lifts no more than 10 inches (254 mm) although it has been laid in lifts as much as 12 inches (305 mm) in some projects. Savings are made fi'om the possrble reduction in labor, which could reduce a project that once required two lifts of RCC into one lift. Mechanical Properties Test data fiom RCC pavement projects shows that the behavior of RCC is very similar to that of conventional concrete. Although its construction my differ fi'om that of conventional concrete, it performs as a rigid plate by absorbing the vast majority of stress in the RCC layer. This allows for the subgrade and base to perform as non-load bearing layers for the most part. It is important to note that the properties within 12 to 18 inches (305 to 457 mm) of the edge of the pavements vary slightly from those near the center of the pavement as seen in Figure 10. This is due to inadequate compaction in these areas [11]. 20 ‘." ~9n=1-.:m-gam.mrnm=m — lit-lure 1t iiitfl .; ' " " PM I ‘ l \‘1 are“ D- l,%a‘; l9 {8 M} 1 3“ "has r 1.“: I . . 14% ...- " ’ n “Meir r~- ' s M 4" Figure 10. Edge View of RCC Pavement (After One Day). Compressive Strength The range of typical 28 day compressive strengths has been found to range fi‘om 3500 psi to over 5000 psi (24.1 to 34.5 MPa). Other data from cored project indicates an even wider range of compressive strengths after several years [14]. This range of tested value is not different than conventional concrete. FlexuraI Strength The range of typical flexural strengths of RCC is fi'om 500 psi to over 700 psi (3.5 to 4.8 MPa). The Portland Cement Association recommends the following relationship in equation (1) to approximate the flexural strength of RCC: 21 [=Clrrin where: limited data has i lhe Value of9 “'5 tesgned using 9'~ dermal Sirens"sh " hitting Tensile TeSts has if!) to over 600 tests need to be 1 scs'le strength I tssimersII 1]. Madam 07‘ Eta. Altho ugfi $5M“ has She t ‘; this Ihe Portl E:C(rt)l1 r. = C (mm (1) where: fr = flexural strength of RCC, psi. C = constant (equal to 9). fc = compressive strength at 28 days, psi. Limited data has shown that C varied from 9.4 to 10.8 depending on the RCC mixture. The value of 9 was chosen as a conservative value. Airport pavements are typically designed using 90 day flexural strengths. This is usually about 10% greater than the flexural strength at 28 days [14]. Splitting Tensile Strength Tests have shown that the splitting tensile strength of RCC cores has varied form 400 to over 600 psi (2.8 to over 4.1 Mpa) after 28 days of curing. The flexural strength tests need to be performed on sawed beams which can be cumbersome. The splitting tensile strength test provide more reliable results and are more easily obtained from core specimens [11]. Modulus of Elasticity Although little data has been accumulated from RCC cores, the modulus of elasticity has shown similar to slightly higher values than those of conventional concrete [ 1 1]. The Portland Cement Association recommends the relationship in equation (2): E = C (1%)“2 (2) 22 “here: limited data has 5 m: [l4]. Film 7 251mg JUSI hm: c reinionslip bE'M ESiHfilatbaxm 1W SWRglh Since RC Tm bond betwa or as a Partially 1. 3W1 Str Well“. LS mud “Goof W “length Colin'uqed R C 711: Show that . “an as W where: E = elastic modulus of RCC. psi. C = constant (equal to 57,000). f‘c = compressive strength at 28 days, psi. Limited data has shown that C varied from 59.000 to 67,000 depending on the RCC mixture [14]. Fatigue Testing Just like conventional concrete, RCC is subject to fatigue behavior. The relationship between RCC fatigue and conventional concrete fatigue has been shown to be similar based on limited testing [11]. Bond Strength Since RCC is often put down in multiple lifis, bond strength is a key property. The bond between difierent lifts will determine if the slab will act as one monolithic layer or as a partially bonded layer, which would signify a lower load carrying capacity [11]. Bond strength is also important for the minimization of cold joints. The bond strength is much lower for cold joints (Figure 11) than those of fresh joints. “Good” bond strength would be regarded as at least 50 percent of the direct tensile strength of the RCC material. Based on limited data, it appears that properly constructed RCC pavements exhibit the necessary bond strength. However, some data has shown that longitudinal construction joints may not have nearly as much bond strength as interior locations exhibit [1 1]. 23 Freeze-77m Du! Many RC mined “ell. neon) using his test for RC( the field tests th F reeze-Thaw Durability Many RCC pavements have been built in freeze-thaw susceptible areas and have performed well. However, when RCC samples have been tested for freeze-thaw durability using ASTM C 666, the samples have performed poorly. The applicability of this test for RCC is unclear at this time. A better indication of the durability of RCC is the field tests themselves [1 1]. Figure 11. Longitudinal Cold Joint (After One Day). The performance of many of the RCC pavement projects has been good with respect to freeze-that durability. Very little evidence has been found to indicate that RC pavements in cold weather regions are not durable in this aspect. Even where deicers have been used, there has been no evidence of scaling problems [15]. Studies by Ragan [16] as well as Pigeon and Malhotra [17] have shown that even though some RCC samples exhibited high bubble spacing factors that are normally 24 initiated with mites contains RCC nix “like aggregates and mos. RC C pax Studies that dwabilin entrance. RCC could result in sight reductio. SlRl' CTL'R Overview Since 1 Mammy. 1mm fiexm Wilt this in m 13.3?an 0U associated with low freeze-thaw durability, the low water-cement ratios of some RCC mixes contains very little freezable water in the paste. This creates a low permeability RCC mix which is difficult to critically saturate. This along with non-frost-susceptible aggregates and proper curing conditions helps overcome the deficient air void system of most RCC pavements. Studies by Dolen [l 8] have shown that air-entrained RCC samples show freeze- thaw durability factors on the order of 60 to 400 percent greater than those of non air- entrained RCC samples. These samples also showed an increase in workability which could result in a reduction in the water-cement ratio. Just as in conventional concrete, a slight reduction in compressive strength was associated with air-entraining however. STRUCTURAL DESIGN OF RCC PAVEMENTS Overview Since RCC pavements respond similarly to conventional concrete pavements mechanically, the design procedures for both are alike. The procedure is based on limiting flexural stresses and the corresponding fatigue damage caused by wheel loads. With this in mind, all concrete pavements have different stress and deflection responses depending on wheel load placement. Generally, wheel loads near the edge of a pavement create higher stresses than those placed at the interior of a pavement away from discontinuities [1 1]. Two predominant design procedures exist for RCC pavements. One design method was developed by the Portland Cement Association (PCA) while the other method was developed by the U.S. Army Corps of Engineers (U .S. Army COE). 25 PCA Method The P( mistrial RC ( operations [I l '13 Wheel loat raster at eit‘: Mistrial faei he edge oft} seamen. ll}: PCA pm 9313‘ times 1h For tl PCA Method The PCA’s RCC pavement design method is intended for use in the design of industrial RCC pavements [14]. However, it can also be used for similar paving operations [11]. This procedure uses the Westergaard interior stress calculation due to the wheel loads as the critical stress, thereby ignoring any variability or degree of load transfer at either natural cracks or engineered joints [1]. This assumes that wheel loads at industrial facilities do not operate like highway pavements where the wheel load is near the edge of the pavement. To account for edge loading conditions, the PCA method recommends that a 20 percent increase in the thickness of the edge be constructed [14]. The PCA procedure does not give any recommendations on contraction joint spacing. but only notes that natural cracks will form [1]. For the analysis of fatigue life, the PCA method utilizes the concept of the stress ratio. The stress ratio is defined as the stress of the design load divided by the flexural strength of the RCC used in the pavement. The PCA procedure then relates the design stress ratio to the expected load repetitions until failure of the RCC pavement using either equation (3) or (4): SR = 118.31 — 10.73(logw(NA)) for NA 5 600,000 load repetitions (3) or SR = 40 for NA 5 600,000 load repetitions (4) where: SR = stress ratio (ratio of interior stress due to the design wheel load to the 28 day flexural strength of the RCC). NA = number of load applications until failure. 26 this equation “'3 miles using W an incorporated mttt‘lllllflllfll ta LS Army COE The US. aconerxatite we contentioml cor 0&359‘0 matimt mister capabiiit 31??wa near c i‘tss equations. 3010 60 feet «‘91 W18 areas. ar m m. In the Cc “it? SirUnlined, This equation was derived from a series of laboratory fatigue tests on RCC beams samples using the rupture of the beams as the failure criteria. Then a conservative factor was incorporated into the equation to account for uncontrollable construction and environmental variables [2]. U.S Army COE Method The U.S. Army Corps of Engineers method for the design of RCC pavements uses a conservative version of its conventional concrete pavement design procedure. The conventional concrete pavement design assumes load transfer at joints and cracks results in a 25% maximum bending stress reduction. The RCC design however assumes no load transfer capability and therefore, no maximum stress reduction occurs when wheel loads are placed near cracks and joints. Stresses are calculated using the Westergaard free edge stress equations. Transverse contraction joint spacings are recommended to be between 30 to 60 feet (9 to 18 m) for airfields and 50 to 75 feet (15 to 23 m) for roads, streets, parking areas, and open storage areas if utilized to control unpredictable natural crack patterns [1]. In the Corps’ RCC design procedure, fatigue life of RCC pavements are modeled using behavioral data fi'om pavement test sections on which accelerated aircraft loads were simulated. The criteria for failure was defined as one-half of the slabs in the pavement tests section exhibiting one or more structural cracks [2]. 27 Comparison 201 note incorporate the 1 \either of these other of the pa mewm UKCo newness oteomentiona Eli is Similar a M uses the ts 5211936 track's “Nance l HELD PER In a P mm’nts f1. jlfse Siteg_ l Comparison and Contrast Both the PCA and Corps’ method for the design of RCC pavements do not incorporate the benefits of load transfer at discontinuities in the RCC pavement slab. Neither of these methods includes reasoning on transverse contraction joint spacing if the owner of the pavement would like to incorporate this feature. Both of these methods would result in thicker pavement sections than needed from a stress viewpoint [2]. The Corps of Engineers’ method does incorporate physical conventional concrete pavement test section data into their fatigue performance model. Although the mechanics of conventional concrete and RCC pavements are similar, this assumes that their fatigue life is similar as well. The PCA RCC design method instead incorporates RCC test data, but uses the beam rupture point as failure. Although this model may help in predicting fatigue cracking in RCC pavements, the criteria generally does not signal the end of the performance life of a pavement. FIELD PERFORMANCE OF RCC PAVEMENTS In a PCA study, Piggott [15] reviewed the performance of 34 in-service RCC pavements fi'om across North America and reported on the general condition of 18 of these sites. He found that, in general, RCC pavements tend to have a rougher surface than normal PCC or asphalt pavements. However, due to the introduction of high-density asphalt pavers, newer RCC pavements have been found to have ride qualities near those of new conventional PCC or asphalt pavements. This important quality should lead to more RCC pavements being used as roads which require a certain level of smoothness to drive. 28 Structur; [t5] attributes it tietor Mich 001 it that they are 'ILilIfll'} facilitie: retemerits coul. RCC p3. Imus years of 5 titles in than Widths which 0 problem Agai mistress in ma Orerall treble pat-em, it Sane SUrfat than 3‘19le te m ahfimatiw ”TIRE 0r Piggof RSEOnaHb‘ Per. “firms bui iTsm ' 939M: Structural failures were found to be very uncommon in RCC pavements. Piggott [15] attributes this in part to the high strength that RCC achieves over time. An other factor which could have influenced the general good condition of RCC pavements could be that they are over-designed. Since most RCC pavements are currently in use in military facilities, storage areas, and interrnodal terminals, the design traffic for these pavements could have been overestimated, resulting in an increased life of the pavement. RCC pavements that have shown extensive cracking have only done so after many years of service. These cracks have not been a factor in deterring its service qualities in many cases though. Although many pavements have shown large crack widths which would lead to higher stresses, faulting does not appear to be a major problem. Again, Piggott [15] attributes this to possible over-design of the RCC pavement thickness in many cases. Overall, Piggott notes that most of the RCC projects surveyed have provided a durable pavement to their respective users. While the RCC pavements did not provide the same surface quality of conventional PCC pavements, they were found to be more than adequate for service in low volume, high load areas and provided the users a low- cost alternative in these cases. FUTURE OF RCC PAVEMENTS Piggott notes in [26] that “RCC pavements can be successfully built to carry traflic on all but the highest class of multi-lane highway.” Many of the newer RCC pavements built in North America have enjoyed excellent surface smoothness due to high density asphalt pavers and the quality of the rolling after the initial placement. 29 Studies 5 3 payment 5 multiple WW nine admit tough RC C F mace texture Along ‘ menu} heent ilonh'id. 0R residential stre «reaction 2“ it: tha Events. Studies in Australia [19, 20] have been developed in order to test the use of RC C as a pavement surface material for residential and high-speed arterial roads. The use of multiple layers of RCC has been found to aid in the construction of smoother pavements with the adequate roughness and skid resistance for high-speed travel. In many cases though, RCC pavements were designed to be overlaid with a small amount of asphalt for surface texture consistency. Along with the Australian RCC pavement experiments, RCC pavements have recently been used as secondary highways (Williams Lake, BC), collector streets (Portland, OR and Edmonton, AB), internal roads and parking, (Spring Hill, TN), and residential streets (Alliance, KS and Fort. St. John, BC) with great success. With quality construction, more RCC pavements will eventually be built to support larger volumes of trafic that have historically needed smooth riding asphalt and conventional concrete pavements. 30 DESCRlPTlO.‘ Non-ties tillDt. heats-u lip \thrator All North America. EFitters Water AB “'35 CO nduct Eimonton The Wham: the ir l Austin. TX The RCC had- meems. biz-310 mOVe 1 “e an 390655 to ad 0% item 10% 30111 the . int» mmmjlirm - CHAPTER III - RCC Pavement Sites DESCRIPTION OF TEST SITES Non-destructive testing (NDT) deflection data from falling-weight deflectometer (FWD), heavy-weight deflectometer (HWD), and the Waterways Experiment Station 16- kip vibrator (WES) was utilized fiom six different RCC pavement sites located across North America. Testing for the first five sites was conducted by the U.S. Army Corps of Engineers Waterways Experiment Station, while the testing for the last site in Edmonton, AB was conducted by Construction Technology Laboratories, Inc. (CTL) and the City of Edmonton. The six sites are described in the following sections. Tables 1 and 2 summarize the inventory data for these sites. Austin, TX The RCC pavement sites in Austin, TX were generally used as low volume, high load pavements. The Central Freight site is a terminal site which was utilized by tractor trailers to move freight around the terminal. The Tuscany Way site in Austin, TX served as an access road to the trucks which serviced the Central Freight Line Terminal and other nearby locations [1]. Both the Austin, TX sites were comprised of a 7” (178 mm) RCC surface over a 6” (152 mm) lime-stabilized base and were constructed in April 1987. The RCC was a 50/50 mix of Type I Portland Cement and Class C fly ash producing a relatively weak surface layer with a flexural strength of 550 psi (3792 kPa). The test data used fiom both 31 of these sites \ luscant \Va} Table 1. RC 1 lin=354 in", % Location *1.— .tunin TX it. (triplet. M of these sites was originally collected in September 1991 using a HWD device. The Tuscany Way access road was tested again in January 1992 using a HWD device [1]. Table I. RCC Pavement Site Information (After Pittman [1]). lin=25.4mm RCC Base . Size Date Date(s) Test Base Location Area (sq yd) Const. Tested Device' Layer Layer Type b (In.) (In.) Egg] 90,000 Apr-87 Scp—9l HWD 7 6 LSB AuStin’ TX T s 91 HWD uscany _ ep- Way 14,670 Apr 87 Jan” HWD 7 6 LSB Ft'cmnpbe“: Ch63irr1ilal 66 500 J 1-87 Jan's” FWD 75 4 CLS KY e c ’ “ Aug-91 HWD ' Company PN69A 20,200 Oct-89 Aug-91 HWD 10 10 CGB PN69B 23,500 Jul-89 Apr'90 FWD 10 10 CGB Aug-91 HWD Ft' Dmm’ NY A 90 FWD pr- PN187 18,000 Aug-89 Aug-91 HWD 10 10 CGB PN203 3,700 Aug-89 Aug-91 HWD 10 10 CGB Bld Mar-85 was 260% 20,000 Jul-84 Feb-90 FWD 10 12 LSB Sep.9l HWD Bldg. Feb-90 FWD 38033 18,600 Aug-88 Sew] HWD 9 6 LSB Ft. Hood, TX Bldg. Feb-90 FWD 3850 63,900 Oct-87 Sew] HWD 8.35 6 LSB Wash Feb-90 FWD Rack 20,000 Sep-89 899-91 HWD 9 6 LSB Tank Feb-90 FWD Trail 5,200 Sep-89 Sep-91 HWD 9 6 LSB SpringHill, Zenith TN Road 13,200 Nov-88 Jan-91 HWD 6 -- CGB ' WES = WES l6-kip Vibrator FWD = Falling-Weight Deflectometer HWD = Heavy-Weight Deflectometer 32 " LSB = Lime-Stabilized Base CLS = Crushed Limestone Base CGB = Crushed Granular Base 5...— CI I 1.2.! . .Z.\Ct— w r~VI.~.:<-= . A. : 2.2.3.?- 56-54: akin-2.92%.: ===.=:~=L.~ 1:: 2:22.33 a; . V. v: .N Siskh can: waxes. “ERIE”: was: 99% mm 5% u 556:8 beooetzm woes—BS E can.» 053 3a «venom 5 £303 a EwEB ,3 6:8 Em< 2m + EoEoUVEBaB 6 Eu» 05:0 eon menace E Ems? . coo 3.0 cm? 022 aim NE of n. oov . eaom fiEoN .EEflwEm cow ovd $2 coon ER 0: 9: u now 28H xceh 00w oed $2 ooom :95 o: o: m mom xoum :83 . 33 3.2 . 2w 9. o 9 NR: 8 E: 2% o: m: u 8m 0an mam v2 oow ovd $0— ooom LE e: 9: m mom mmme .35 .eoom E 08 mmd 82 mo K ..3m 02 02 U chm N £2 -- 08 320 ~52 mnmm =N: _ w: w? U N; :22 -- ~88 .85 can who go Ema QR oz” of m Owe moNZm omw mmd mwa Emm aim 2N cm. m 0? Sim >2 omw mmd wwm ZEN Lem 3N of m 03 mooZm .EED .E 03 2.0 wwm .mmm aim 3N o2 m 03 <32; any >M 02. end 33 mm: aim mom _ _N m oov _8_Eoso £39.80 23 .E 0mm who 03— o E“ .3». NM: com 0 com 33 agomah . 292m 5 .522 0mm mm o 32 OS: __v\m NM: com U own 15:00 a fiweobm 8 e33— ._ 238$ 2335 35 EMS; 2303 2. :1 23.85 .555...— meeEEEoU ouawoeuut. ._ .532 .. .. .. no.5. .8339— UUM 3.855 0::— oaauoau < 99:30 .5235 it. by.— EoEoU ~ 09:. I 5: genie _ ”a? wmmvuneee a as vans _ .2: 5:5...— hefi aces—Ea... SEE—2.. :2 2.2.325 00: .N 63-h 33 Fort Cam phi The R nihars tehic maintenance Clerical Co. This t (102 mint cn lE‘pel Portl. [0 give an at constructed Fort Campbell, KY The RCC pavement site at Fort Campbell, KY was used to provide access for military vehicles including tanks, trucks, and other vehicles to motor pools and maintenance shops. These pavements were located in low-speed areas of the 63rd Chemical Company military installation [1]. This cross section was comprised of a 7.5” (191 mm) RCC layer resting on a 4” (102 mm) crushed limestone base for drainage. The RCC in this pavement utilized a Type I Portland Cement along with a Class F fly ash as filler in a 1.9 to] ratio by weight to give an average flexural strength of 760 psi (5240 kPa). This pavement was constructed in July 1987 and tested in January 1991 and August 1991 using an FWD and HWD respectively [1]. Fort Drum, NY The third RCC pavement site considered in this study is at Fort Drum in upstate New York. As with the RCC pavement site at Fort Campbell, KY, the primary purpose of this pavement was to provide low-speed access for large military vehicles. This site is sub-divided into four sections, PN69A, PN69B, PN187, and PN203, based on project numbers (PN) denoted during construction [1]. All of the projects at Fort Drum consist of a 10” (254 mm) RCC layer resting on a 10” (254 mm) crushed granular base. Although all of the different PN’s were constructed at different times during the summer of 1989, they all utilize the same RCC mix design. The mix design consists of Type I Portland Cement mixed a Class F fly ash filler in a 3 to 1 ratio. This resulted in a high average flexural strength of 820 psi (5654 kPa) for this 34 RCC nut. Sections seniors were then it lort Hood. TX The fourth R site's suhditided in 38033. and Buildint he Building nunthe used for washing ta link Trail. Projec RCC W‘Etttents [l The Builcii l? ( 305 mm] lim fimm SUNiVidet mm aggregate RCC mix. Sections PN69B and PN187 were tested using an FWD in April 1990 and all sections were then tested in August 1991 using an HWD [1]. Fort Hood, TX The fourth RCC pavement site considered in this study is at Fort Hood, TX. This site is sub-divided into five sections. The first three sections, Building 26027, Building 38033, and Building 3850, were access roads for military vehicles and were denoted by the Building number. The final two sections were located at a Wash Rack which was used for washing tanks and the access road to get to the Wash Rack which is noted as the Tank Trail. Projects at Fort Hood varied in mix design as well as thickness design of the RCC pavements [l]. The Building 26027 section consisted of a 10” (254 mm) RCC layer resting on a 12” (305 mm) lime-stabilized base and was constructed in July 1984. This section was further subdivided into two different mix designs, one used a 1.5” (38 mm) top size coarse aggregate and another used a 0.75” (19 mm) top size coarse aggregate. The section with the lager coarse aggregate used Type I Portland Cement in conjunction with a Class C fly ash in a 2 to 1 ratio by weight with a water-cementitious ratio of 0.34. The normal sized coarse aggregate section used the same constituents and ratios of binders, but in greater quantities and with a very small water-cementitious ratio of 0.23. Both of these sections attained an average flexural strength of 830 psi (5723 kPa) after 28 days of curing. This particuhr section was tested using all three deflection testing devices, the WES in March 1985, the FWD in February 1990, and the HWD in September 1991 [1]. 35 The Built mint lime stahilia The mix design 1 aggregate o ith 2 Cement aith a l Strength of Sign) and the MD '11 The RC ‘-|I fl 8.)) (.‘l: 1 October 1987, fill“ in a 2.5 to THE Strength 1.: 0-43 for RCC. Fem” 1990 The 1211 September 1 9! imbibed base The Building 38033 section was built as a 9” (229 mm) RCC layer over a 6” (152 mm) lime stabilized base. This particular section was constructed in August of 1988. The mix design for this RCC pavement consisted of a 0.875” (22 mm) top size coarse aggregate with a 0.40 water-cementitious ratio. The binder consisted of a Type I Portland Cement with a Type F fly ash filler in a 2 to 1 ratio resulting in an average flexural strength of 800 psi (5516 kPa). This section was tested using the FWD in February 1990 and the HWD in September 1991 [1]. The RCC pavement section located near Building 3850 was specified to be a 8.35” (212 mm) RCC layer on a 6” (152 mm) lime stabilized base when constructed in October 1987. The binder consisted of a Type I Portland Cement with a Type C fly ash filler in a 2.5 to 1 ratio resulting in an average flexural strength of 835 psi (5757 kPa). This strength was high when compared to the relatively high water-cementitious ratio of 0.43 for RCC. As with Building 38033, this section was also tested using the FWD in February 1990 and the HWD in September 1991 [1]. The tank washing area and access road to the wash rack were both constructed in September 1989 and are comprised of a 9” (229 mm) RCC layer over a 6” (152 mm) lime stabilized base. The mix design for this RCC pavement consisted of a 0.875” (22 mm) top size coarse aggregate with a 0.40 water-cementitious ratio. A Type I Portland Cement with a Type F fly ash filler in a 2 to 1 ratio was used for both sections. The average flexural strength for these sections was 800 psi (5516 kPa). This section was tested using the FWD in February 1990 and the HWD in September 1991 [l]. 36 Spring HilL l The 8} Road. This it min {ESle‘lj The S' patentent late mix design to TEpel Portia retaking am 536 We test: ”Mouton, , The E 1.71m AVenn 0% dc: :‘Q‘Sl and fit Spring Hill, TN The Spring Hill, TN RCC pavement site is part of the Saturn Plant on Zenith Road. This road provides access to tractor-trailers as well as normal sized vehicles to the main assembly building [1]. The Spring Hill, TN RCC pavement site consists solely of a 6” (152 mm) RCC pavement layer on a stifl roadbed soil. This site was constructed in November 1988. The mix design for this site used a 0.35 water-cementitious ratio with a binder combination of Type I Portland Cement and a Class F fly ash as filler material in a 2.67 to 1 ratio. The resulting average flexural strength of this mix was 600 psi (4137 kPa) after 28 days. This site was tested only once with an HWD in January 1991 [1]. Edmonton, AB, Canada The Edmonton, AB RCC pavement site is part of 112th Street between 167th and 171th Avenues in Edmonton, AB. This road is a two-lane city arterial road and was originally designed to be an RCC surface road for a short evaluation period (up to 5 years) and eventually was converted to an RCC base for an asphalt-surfaced pavement [21]. This site consists of a 8” (203 mm) RCC pavement layer resting on a 6” (152 mm) cement stabilized subgrade constructed in August 1992. The binder for this project consisted of Canadian Type 10 Portland Cement only. The resulting average flexural strength of this mix was 404 psi (2782 kPa). This site had joints cut at different intervals to test the effect on load transfer efficiency. This site was tested twice with an FWD in September 1992 and September 1993 [22]. Since this site was set up with cut joints and 37 lidnotin lt'ateru‘af DATA C D ll'atem'aj .‘ l‘ the term a-) O (J! .‘o 3;) .t -.J 7—4 This stud did not include the same information which the U.S. Army Corps of Engineers Waterways Experiment Station studies did, it was analyzed separately. DATA COLLECTION Data collected for every crack or joint during the U.S. Army Corps of Engineers Waterways Experiment Station study [1] and subsequently utilized in this study includes the following: 1. Project location 2. Date of tests 3. Pavement surface temperature 4. Crack or joint width 5. Crack or joint spacing 6. Deflections fi'om F WD/HWD testing at cracks/joints 7. Midslab deflections from FWD/HWD testing This study focused only on data from transverse cracks and joints while the U.S. Army Corps of Engineers Waterways Experiment Station study included many types of cracks and joints including both fresh and cold joints, longitudinal cracks, and others. Data for crack widths was determined using an optical scale lupe. This apparatus allows for magnification of the crack up to 7 times the normal size resulting in an accuracy of the crack width to the nearest 0.001 inches (0.0254 mm). The crack widths were measured at locations closest to the LTE testing location for the FWD/HWD where no Spalling, wearing, or excessive damage was done to the crack. In cases where joint sealant or excessive fines closed the surface of the crack, no measurements were taken [1]. To measure crack spacings during the U.S. Army Corps of Engineers Waterways Experiment Station study, a rolling measuring wheel was used resulting in an accuracy of 38 0.1 feet (30.5 mnti distances to the ne; Data from 1. Da 3. loi 3- Dc 4. .\t The Edmonton. _-‘ 56310115 which \A Wooded. ln additi material proper] DATA ANAL TWodi Conslszmc.“ ot‘ déflecrjom We Teenage“, 21 Va the mam b come- rant applied to the “be 3. 0.1 feet (30.5 mm). The resulting crack spacing was found by averaging the two distances to the nearest cracks for every crack tested using the FWD/HWD. Data fiom each joint in the Edmonton, AB [22] study included the following: Date of test Joint spacing Deflections from FWD/HWD testing at joints/cracks Midslab deflections from FWD testing PP’N?‘ The Edmonton, AB study focused on engineered joints at various spacings. With the few sections which were allowed to naturally crack in this study, crack spacings were not provided. In addition to the data on each joint or crack fiom both studies, cross-section and material properties were also found as provided earlier in this chapter. DATA ANALYSIS Two different backcalculation methods were analyzed in order to verify the consistency of these individual procedures. Thicknesses, loads, and the corresponding deflections were randomly selected for 10 different RCC pavement sites. These sites represent a variety of soil support conditions, mix designs, and locations. Afier analyzing the different backcalculation procedures, a typical range of elastic modulus of the concrete, radius of relative stiffness, modulus of subgrade reaction can be determined and applied to the RCC pavement site analyses. The deflection information is summarized in Table 3. 39 Table 3. De 1 h=1” mr Case Loc )0. Table 3. Deflection Data from Different RCC Test Sites. 1in=25.4mm llb=4.45N RCC Deflections (mils) Case . . Load , , N 0 Location Thlckness (lb) Distance from Load (1n.) (“‘J 0 12 24 36 48 60 72 Ft. 1 Campbell, 7.5 26648 14.3 13.1 10.4 7.8 5.9 4.2 3.2 KY 2 F"¥§°d’ 10 2590811.4 10.1 8.9 7.6 6.3 5.3 4.2 Spring 3 1mm 6 26224 7.9 5.9 3.7 2.2 1.2 0.6 0.4 4 Ft'NYD‘m‘” 10 9546 2.34 2.00 1.61 1.23 0.88 0.67 0.53 Ft. Drum, 5 NY 10 13992 3.48 3.03 2.34 1.75 1.30 1.00 0.73 6 Ft'NYD’m’L 10 19284 4.67 3.95 3.22 2.45 1.83 1.29 0.89 Ft. Drum, 7 NY 10 23844 6.59 5.62 4.45 3.34 2.38 1.65 1.15 8 EMXS‘O’L 8 6439 4.67 3.76 2.85 2.19 1.75 1.46 1.12 9 "Em/$0“ 8 9374 7.03 5.73 4.35 3.35 2.68 2.17 1.73 10 Edmgmn’ 8 12316 9.47 7.72 5.89 4.51 3.58 2.87 2.30 ERES Method In the ERES method of backcalculation, the first parameter to be calculated is the basin area (AREA). This parameter can be defined as the cross-sectional area of the deflection basin between the center of the FWD or HWD load plate and the outermost deflection sensor, normalized with respect to the maximum deflection (i.e., deflection at the sensor directly below the center of the load plate, 60) [23]. Due to this normalization 4O to the mu computed ‘1' t0. 8. 1 center 0t t deflection .tPfA : to the maximum deflection, AREA has units of length rather than area. It can be computed by using deflection data measured at sensors located at various radial distances “r” (O, 8, 12, 18, 24, 36, and 60 in, or O, 203, 305, 457, 610, 914, 1524 mm) from the center of the FWD or HWD load plate. This should be done once for each set of deflection data. Equation (5), fiom [23], can be used to calculate this parameter: F f ' 5 6 6 4 + 6 5—3 + 5 Eli + 6 6L3 + AREA ___ 0 0 K 0 (5) 624 636 (660 9 T +1 — +12 — 0 60 k 60 _ where: AREA = deflection basin area, in. 5, — deflection of the rlh sensor, mils. The next parameter that is calculated is the radius of relative stiflhess. This parameter characterizes the combined stifliiess of the slab-foundation system [23]. As with AREA, this should be calculated once for every set of deflection data. Equation (6), found in [24], is used to compute Z and is only valid for a load plate radius of 6 in (150 mm): 2.566 e: 1n{(60““zEA)}/(— 0.698) (6) 289.708 where: Z = radius of relative stifliiess, in. AREA = deflection basin area, in. 41 AI dimension mm senor 10c dffleetm Table 4, Radia “1512110: (in) Hoe/cl h—O 00 44/4 mil-’11:, C o are] Afier computing the radius of relative stiflhess for a given set of deflections. a non- dirnensional regression coeflicient must be calculated for each sensor. Equation (7) is a regression equation for which the regression coeflicients (a, b, and c) differ for each of the sensor location, as seen in Table 4. 8,. takes into account the decreasing pavement deflections as a fiinction of distance from the load plate. .. [454-61)] 6 = ae (7) r where: a,b,c = regression coefficients from [23] (see Table 4) I = radius of relative stiflhess, mm. Table 4. Regression Coefficients for 6,. (After [23]). Radial Distance, 1' a b c (in) 0 0.12450 0.14707 0.07565 8 0.12323 0.4691 1 0.07209 12 0.12188 0.79432 0.07074 18 0.11933 1.38363 0.06909 24 0.11634 2.06] 15 0.06775 36 0.10960 3.62187 0.06568 60 0.09521 7.41241 0.06255 After computing 8:, the next step in the ERES method is to calculate the elastic modulus of the concrete (E). A value for the elastic modulus of concrete is computed for each sensor’s deflection, noted here as E,. The reported elastic modulus values in this thesis are the average elastic modulus values calculated from each sensor for a given set 42 ofdeflect 0001(23] E v P ii. ‘0. In «0. of deflections. The elastic modulus for each sensor can be computed using equation (8) fiom [23]: 12(1- v2)P425r* ’— 5h3 r where: l?" O pozryzmv-e< II II (8) concrete modulus of elasticity based on 8,, psi Poisson’s ratio for concrete (assumed to be 0.15) applied load, lb radius of relative stiflhess, in deflection of the rth sensor. mils concrete slab thickness, in nondirnensional deflection coefficient at radial distance “r” The final step in the ERES backcalculation procedure is to compute the modulus of subgrade reaction (k). The modulus of subgrade reaction estimates the stiffness of the foundation under the RCC layer. It can be calculated using equation (9) from [25]: 3 k_ Eh where: “(WNW 12(1— v2)e4 (9) modulus of subgrade reaction, psi/in concrete modulus of elasticity, psi concrete slab thickness, in Poisson’s ratio for concrete (assumed to be 0.15) radius of relative stiflhess, in 43 ECOPP Pax'emer theoretic range 01 ECOpp ECOPP Method The second method for backcalculation is the ECOPP (Estimation of Concrete Pavement Parameters). This method was developed through regression analysis on theoretical, load-induced deflections of 288 pavement sections which represented a large range of dimensions and material properties. The first step in this analysis is to calculate the deflection basin slope factor (SF) using equation (10) from [26]: SF = M (10) 50 where: SF = deflection basin slope factor 50 = deflection of sensor at load plate, mils 824 = deflection of sensor located 24 inches (610 rmn) fi'om load plate, mils. Afier the deflection basin slope factor has been calculated, the next step in the ECOPP process is to calculate the radius of relative stiflhess (3) using equation (11) from [26]: l g = 11 [0.00401 148 + 0.102021(SF) — 0.0044331 1* log(SF)] ( ) where: 6 = radius of relative stiffness, in SF = deflection basin slope factor. calculate bacteria equation RDe = ( OJ COfilpttte C P’f‘"-‘~7r Afier the radius of relative stiffiiess has been computed, the next step is to calculate the modulus of subgrade reaction (k) of the foundation layers. For this backcalculation procedure, the normalized 636 deflection must be calculated first using equation (12) from [26]: N036=536*[9(::)0]*1000 (12) where: ND36 = normalized 836 deflection, in 836 = sensor deflection located 36 inches from the load plate, mils P = applied FWD or HWD load from load plate, lb. Using this normalized deflection, the modulus of subgrade reaction can then be computed using equation (13) for slabs thicker than 6 inches (152 mm) or equation (14) for slabs thicker than 8 (203 mm) inches from [26]. Regression coefficients for equations (13) and (14) are listed in Table 5. log(k) = 00 + 01*[log(ND36)]+ az * (8) + a3 * (%J (13) log(k) = ao + 01*[log(ND36)]+ az * (I?) + as * (é) (14) where: k = modulus of subgrade reaction, psi/ in ND36 = normalized 836 deflection, in I = radius of relative stifliiess, in r = radius of load plate, in L = RCC slab length, in an, a], a2, a3 = regression coeflicients from [26] (See Table 5). 45 Table 5. RC C lbickn mduiu 073‘ 0r. Table 5. Regression Coefficients for a., a2, a3, and a.. (After [26]). RCC Slab Thickness (in) a' “2 a3 a‘ 5 4.67014 -1.00366 -0.04159 -2.93908 6 4.08076 -0.99399 -0.02821 -1 .88585 8 3.19287 -0.99076 -0.01298 0.40057 10 3.21337 -1.00799 -0.01342 0.42543 12 3.20290 -0.99595 -0.01353 0.48629 15 3.14526 -0.98294 0.01243 0.56773 18 3.07694 -097320 -001 160 0.65605 20 3.03577 -0.96488 -0.01079 0.71466 The final step in the ECOPP backcalculation procedure is to compute the elastic modulus of the concrete (E). Unlike the ERES procedme, the elastic modulus is calculated only once using equation (15) from [26]: _12(1—v2)ke‘ E h3 where: B‘NW< {'11 ll (15) concrete modulus of elasticity, psi Poisson’s ratio for concrete (assumed to be 0.15) modulus of subgrade reaction, psi/1n radius of relative stiflhess, in concrete slab thickness, in 46 Backc ECO? rags Table 3'3- Can No. Backcalculation Procedure Comparison A backcalculation analysis was done using both the ERES method and the ECOPP method for the 10 deflection cases found in Table 3. The results fiom this analysis are listed in Table 6. Table 6. Comparison of Backcalculated Parameters from Different RCC Test Sites. l in=25.4 mm 1 psi/in=.271 kPa/mm 1 psi=6.89 kPa Radius of Relative Modulus of Subgrade RCC Elastic Stiffness (in.) Reaction (psi/in.) Modulus (106 psi) Case . Location M ECOPP ERES % ECOPP ERES % ECOPP ERES % Diff. Diff. Diff. Ft. > 1 Campbell, 29.12 30.00 2.93 258 248 4.32 5.16 5.58 7.38 KY 2 Fig?“ 34.12 37.65 9.36 223 197 -12.85 3.54 4.65 23.82 3 Smeifl‘lL 16.82 18.10 7.08 1,163 1,295 10.25 5.05 7.54 33.08 4 Ft'NDY‘m“: 26.26 27.83 5.63 672 629 -6.91 3.75 4.42 15.25 5 Ft'NDY‘m“: 25.26 27.35 7.61 712 638 -11.64 3.40 4.18 18.68 6 Ft'NDY'm" 26.36 27.55 4.31 679 650 -4.46 3.85 4.39 12.38 7 Ft°NDY‘“"" 25.44 26.69 4.69 632 605 4.40 3.10 3.60 13.87 8 mfg“: 21.94 25.26 13.17 276 257 -755 1.47 2.40 38.87 9 Emu/3°“: 22.34 25.60 12.72 260 242 -7.56 1.49 2.38 37.59 10 “mg”: 22.50 25.56 11.99 253 237 -6.92 1.48 2.31 35.86 47 Hattie .l/ttt lhe bactealcula all cases. 111 as seen in F In ( than those 0W predie 035th moc 10.000 and Ni [EVE b S if]: ’1'“ 7 ll], J] 6 {I}: [ll 5H1] .1 J J. 1;. 1“ .- I 3011 in Ml " ' l'flulllh' Nluclulul (pal) 3. ‘ [II-1. ‘11] ”Jr-'1' DU We 12. Elastic Modulus of Concrete Comparison The elastic modulus of the concrete differed greatly depending on the backcalculation procedure and in some cases resulted in a discrepancy of up to 40%. In all cases, the ERES method estimated a greater elastic modulus than the ECOPP method as seen in Figure 12. In Cases 8-10 in Edmonton, AB, deflections at the outmost sensors were greater than those of other RCC pavement sections and may have caused the ERES model to over predict the elastic modulus of the concrete. These sections also had a very low elastic modulus predicted from both methods (about 1.5"‘106 psi and 2.4"“ 106 psi, or 10,000 and 16,500 MPa) for the ECOPP and ERES methods, respectively) which may not have been addressed when the regression models for backcalculation were developed. f“ l psi=6.89 kPa RCC Elastic Modulus (psi) 1 2 3 4 5 6 7 8 Case Number [IECOPPMetbod cranes Method] Figure 12. Results of Backcalculated Elastic Modulus of the Concrete Comparison. 48 com tested CODCTI 11.11:. 169501 In Case 3 in Spring Hill, TN, the deflection for all sensors were extremely low in comparison to other RCC pavement sites. This pavement was also the thinnest slab tested at 6 inches (152 mm). Since the ERES method predicts an elastic modulus of the concrete for every sensor and the average is reported, an abnormal deflection could influence the results slightly. However, every sensor deflection in this case predicted a reasonably consistent, yet high, elastic modulus value. Although the results from this portion of the backcalculation analysis vary greatly, it is safe to suggest that the elastic modulus of the concrete for RCC pavement sections is similar to PCC pavement sections with an average elastic modulus of the concrete value of approximately 4,000,000 psi (27,500 MPa). Modulus of Subgrade Reaction Comparison The modulus of subgrade reaction results were in better agreement (Figure 13). In every case but one (Case 3-Spring Hill, TN), the ECOPP method predicted higher levels of foundation support than the ERES method. 49 filodulu- 0r Subgrade urns-Non (ll-Illfl-l 1,100.- .- . I 21!: O—* 001 K. 1111 N _ If) L400 - -.___-..._.____ __ --___ - - ,, l psi=6.89 kPa 1.200 '5 § § Modulus of Subgrade Reaction (psi/in.) § 1 2 3 4 5 6 7 8 9 10 Case Number [- ECOPP Method Ell-IRES Metlnd Figure 13. Results of Backcalculated Modulus of Subgrade Reaction Comparison. Since Cases 1-3, from Ft. Campbell, KY, Ft. Hood, TX, and Spring Hill, TN, are fiom three different locations and represent three separate base conditions (crushed limestone, lime-stabilized, and crushed granular base, respectively), the results should be expected to difl°er in comparison to each other. However, Case 3 in Spring Hill, TN produced an exceptionally high modulus of subgrade reaction values (1 163 psi/in and 1295 psi/in, or 316 and 352 kPa/mm) for the ECOPP and ERES methods, respectively) due to the low deflections and thin slab (6 inches, 152 mm) at this site. These values may have been out of the intended range for both methods which over estimated the modulus of subgrade reaction (in addition to the elastic modulus of the concrete from the preceding section). 50 AB teem difietent the 111001 methods diietene Cases 4-7 at Ft. Drum, NY (crushed granular base) and Cases 8-10 in Edmonton, AB (cement-stabilized subgrade) showed consistency within the methods with respect to difierent loads. In all of these cases however, the ECOPP method slightly over predicted the modulus of subgrade reaction in comparison to the ERES method. The results of both methods in these cases are relatively equal with almost every result within a 10% difl‘erence between the two methods. Using these two backcalculation analyses, it is suggested that most modulus of subgrade reaction values would range from 150 to 450 psi/in (40.7 to 108.6 kPa/mm). Values over 450 psi/in (108.6 kPa/mm) could have been over estimated by the backcalculation procedures and can be modeled using a lower modulus of subgrade reaction. Radius of Relative Stiffness Comparison Although similar in magnitude, the backcalculated radius of relative stiffness was found to be greater in every case using the ERES method for the set of deflections as seen in Figure 14. Since the radius of relative stiffiiess is a calculated value, it relies on other values (k, E, h, and v) to be determined. If any of these other values were estimated incorrectly, it could affect the calculation of the radius of relative stifliress adversely. 51 .4: l; a. an scmum of “ck-mo Sun-wo- (he-Iota) L4 -— O .35 o l l l l t 1 in=25.4 mm w {A 8 N U 20~ Radlus of Reletlve Stiffness (Inches) Case Number [IECDPPMetbodDERESMerod] Figure 14. Results of Backcalculated Radius of Relative Stiffness Comparison. In all cases except Case 3 in Spring Hill, TN, the radius ofrelative stifliress for these RCC pavement sites was found to be within a reasonable range (22-35 inches, or 559-889 mm) in comparison to PCC pavement sites. Case 3 was aflected by the extremely low deflections and resulting high modulus of subgrade reaction and elastic modulus of the concrete. However, both methods predicted a simikrr radius of relative stiffness resulting in only a 7% difference, which rmy add to validity of the results found in this case. LOAD TRANSFER EFFICIENCY BACKGROUND Load transfer across cracks or joints in RCC pavements is commonly quantified by a term called load transfer efficiency (LTE). Expressed as a percentage, LTE gives an indication of the eflectiveness of a crack in transferring load. Computation of load transfer efliciency based on deflections near a crack under an applied load is a very 52 useful metho approach is t l'se 1 proponionai ”E5 Wu us bad. Detle. tlbthom [Z useful method of determining the LTE. The load transfer efficiency computed using this approach is termed as the deflection load transfer efficiency (LTEs) [26]. Use of LTE5 assumes that the amount of load transfer across a crack is directly proportional to the relative deflections of the unloaded to loaded sides of the crack [28]. LTE5 was used in this study to characterize the ability of cracks and joints to transfer load. Deflection load transfer efficiency was computed in this study by using equation (16) from [27]: 5 U LTE8 = — x 100% . (l6) 5 L where: LTE5 = deflection load transfer efficiency, % 6U = deflection on the unloaded side of a crack or joint, mils SL = deflection on the loaded side of a crack or joint, mils. LTEa can be easily computed using field data fi'om a falling weight deflectometer (FWD) or heavy-weight deflectometer (HWD). An FWD is a device that applies an impulse load, using a 12 inch (300 mm) diameter circular load plate, to a pavement and measures the resulting pavement deflections through a series of sensors. Deflection data for computing LTEs is thus readily available when this device is used. The principal behind HWD testing is identical to that of FWD testing except that an HWD is best suited for higher impulse loads and thicker pavements. Figures 15 and 16 illustrate the meaning of LTE5 by considering the two extreme cases - 0% and 100% deflection load transfer efficiency, respectively. In these figures, a load P is shown to be applied to one side of a crack or joint. In the case of field testing, 53 *7 this load P \l'OUl‘ resulting detleet ligttre l5 that \\ b5n0deflecnor case scenario. at lib can et‘entuu. RCC llaVement lr Fllure l6. where ofthe Crack or lC Shared b‘ Nth si 011th PaVEmem of the dECOmim andergo. this load P would be provided by an impulse by means of the FWD or HWD. The resulting deflections from this load P are depicted in the figures as well. It is seen in Figure 15 that when there is no load transferred (0% LTEs), the unloaded side of the slab has no deflection and thus does not share in the carrying of the load. This is the worst- case scenario, as all the load must be carried by one side and increased deflections result. This can eventually lead to other distresses as well as increase fatigue damage of the RCC pavement in the vicinity of the crack or joint. The best-case scenario is depicted in Figure 16, where the LTE, is 100%. Here, it can be seen that the deflections on each side of the crack or joint due to the applied load P are equal. Thus, the load is being equally shared by both sides of the discontinuity, and the minimum amount of damage is inflicted on the pavement. In this case, the stress caused by this load is also shared by both sides of the discontinuity, thereby reducing the maximum stress a pavement system would undergo. 54 «re—— Figure l 5. 5 Crack/Joint 6v=0 Figure 15. Depiction of 0% Deflection Load Transfer Efficiency (After Bucb [29]). fl‘ ‘le C rack/Joint —— 5L = 5t: Figure 16. Depiction of 100% Deflection Load Transfer Efficiency (After Bucb[29]). 55 LOAD rim“ Load tr. reaction radius relationships bet lhis section has \Vatem‘at-s Exp: Waterways Ex; Data fro; EXDCrimem Stat; different pal'eme reaCllOm and Tad Most ofthe paw 10 Sat on C0518 mailer” 93ij LOAD TRANSFER EFFICIENCY TRENDS Load transfer efliciency data from the U.S. Army Corps of Engineers Waterways Experiment Station [1] and the Edmonton, AB [22] study was analyzed in order to study the eflects of pavement variables, such as such as elastic modulus, modulus of subgrade reaction, radius of relative stifiiress, etc., on load transfer efliciency. In particular, relationships between crack or joint spacing and load transfer efficiency were studied. This section has been divided into two parts: one for the U.S. Army Corps of Engineers Waterways Experiment Station and one for the Edmonton, AB study. Waterways Experiment Station Study Trends Data fi'om all five sites of the U.S. Army Corps of Engineers Waterways Experiment Station Study were grouped together in order to study the general trends of different pavement parameters such as crack spacing, crack width, modulus of subgrade reaction, and radius of relative stifliiess on load transfer efficiency in RCC pavements. Most of the pavements fiom these sites have been allowed to naturally crack transversely to save on costs related to joint cutting. The first part of this section will focus on natural transverse cracking RCC pavements while the second section will analyze jointed RCC pavements from the study in [1]. It should be noted that many variables such as thickness, age, and traffic were not analyzed in this study. Transverse Cracked RC C Pavements With RCC pavements that are allowed to naturally crack, many times the spacing between these cracks tend to be quite large in comparison to conventional concrete 56 pateme be distr ncteas nitb re ’he spa the eas Expert data in relatior lttu‘ee «.1 l. ( 'rssrls \Vldlh (10s.) (I?! ti 1;; Egan} pal. amen pavements. With these large crack spacings, volumetric changes in the concrete can not be distributed to many crack openings, but instead to very few openings. This tends to increase the crack widths in RCC pavements. Figure 17 shows a trend of crack width with respect to crack spacing. Again, the crack spacings in this study are the average of the spacing before and after a particular crack to the most nearby transverse cracks. In the case of this figure and others from the U.S. Army Corps of Engineers Waterways Experiment Station study, each observation on the figure represents an average of a set of data from different sites and subsections of the study. Although scatter is noticed in this relationship, a clear trend of increasing crack widths are noticed with increasing spacings between these cracks. 0.20 0.18 0.16 0.14 0.12 0.10 0.08 Crack Width (in.) e 0.06 0.04 0.02 / - ° 1 in=25 .4mm / t 0.00 1 ioot=.305 m , 0 l 0 20 30 40 Crack Spacing (ft) 50 70 Figure 17. Relationship Between Crack Width and Crack Spacing. In general, large crack widths are not conducive to high load transfer in rigid pavements structures. The trend from Figure 17 would suggest that smaller crack 57 fl spacings w0ul. slabs. This hi; Stresses on but} transfer also de surface for use: Large problem ir. can be limited h} Figures 1 SPBCingS and dell isetlubited \s‘hicl| lifter crack widtl cracks resulted in track itidtbs less 0.08 inches (3 mr 13310.08 lflCbes (- miles l3 mm W; 1““ W of LTE. “manna, high Crack Spacin Wings Over 40 Show crack Spa. Wings are more spacings would increase the chance for high load transfer between discontinuities in slabs. This high LTE would decrease stresses incurred by external loads and share the stresses on both sides of the discontinuity. It is commonly believed that high load transfer also deters faulting of cracks and joints thereby providing a smooth driving surface for users. Although Piggott [15] suggests that faulting does not seem to be a large problem in his RCC pavement visual field study, it is still a potential problem which can be limited by increased load transfer between slabs. Figures 18 and 19 show the relationships between crack widths and crack spacings and deflection load transfer eficiency, respectively. In both cases, a clear trend is exhibited which follows the theory explained in the above paragraph. In Figure 18, larger crack widths led to low levels of deflection load transfer efficiency while tighter cracks resulted in improved load transfer. A fairly clear delineation was seen in LTE5 for crack widths less than 0.04 inches (1 mm) and those between 0.04 inches (1 mm) and 0.08 inches (2 mm). Again, a clear delineation was seen in LTE5 as crack widths opened past 0.08 inches (2 rmn). However, the average LTE5 for crack widths greater than 0.08 inches (2 mm) was found to be around 51%. These cracks would fall into a dangerously low level of LTE5 which would increase the stresses experienced by the pavement under an external load near the crack. In Figure 19, lower levels of LTEa were associated with high crack spacings and vice versa. A similar difference in LTE5 was seen for crack spacings over 40 feet (12.2 m) in distance as with the crack widths when compared to shorter crack spacings. This data would tend to support the theory that larger crack Spacings are more detrimental to load transfer between cracks. 58 .s\s. c...- . .- TXL 3.: .— 1.11:, (°/.) Less Tim 0 04' Between 004" and 00!!“ Greater Than 0.08" Crack Width (inches) Figure 18. Relationship Between Crack Width and LTE5. 1007 , - - 1-..--. — 1 foot=.305 m LTE, (%) Greater Than 40' Less Than 30' Between 30' and 40' C rack Spacing (feet) Figure 19. Relationship Between Crack Spacing and LTEs. 59 As 1 Experiment 20. Points the figure) suggested ' .hneriean . urn - , It .. 30M 01.. X1 ._ '«i'i’a (9’. F 3.- 311 v be? l0- As the modulus of subgrade reaction (k) increased, data from the Waterways Experiment Station study showed that LTE5 also increased. This can be seen in Figure 20. Points that tend to disagree from the general trend (three points on the upper left of the figure) were tested at temperatures in excess of 90°F (32°C) which is over the suggested range of 50 to 85°F (10 to 29°C) for FWD testing as suggested by the American Association of State Highway and Transportation Officials (AASHTO). 8O 7 WV, 0 4’ LTE, (°/.) 8 D 1 psi/irrl=.27 kP mm 0 100 200 300 400 500 600 700 800 900 1.000 Modulus of Subgrade Reaction (psi/in.) Figure 20. Relationship Between Modulus of Subgrade Reaction and LTE5. Pavement temperature can affect the performance of transverse cracks through curling and thermal expansion/contraction mechanisms. Downward curling (at the cracks) and thermal expansion of slabs can occur at high temperatures, resulting in artificially smaller crack widths. This results in a greater potential for contact between opposing crack faces (and thus greater potential for aggregate interlock between slabs), which leads to higher load transfer efliciencies for such cracks. 60 A re order to on high load tr of scatter tt appear that and a .ighe modulus ar contradicts lit. k- 1.1 as. (was A relationship between the radius of relative stiflhess (l) and LTE5 was sought in order to provide a guide to designing RCC pavements for shorter crack spacings with high load transfer. The results of this focus can be seen in Figure 21. This trend has a lot of scatter to it and does not appear to be a true trend. From this graph however, it would appear that greater stiffness characteristics (Eh3) of the RCC would lead to lower LTEs and a higher resulting stress near the crack. If this were true, both a higher elastic modulus and thicker RCC slab would contribute to a loss in load transfer, which contradicts the common belief. LTE, (°/.) 30 20 1 in=25.4 mm 10 0 5 10 15 20 25 30 35 40 Radius of Relative Stiffness (in.) Figure 21. Relationship Between Radius of Relative Stiffness and LTE5. 61 Transverse Jo Joints Hood l\ to the purely in [30]. Nanni of ~10 to 600 90° 0. lion let?! (3.0 to alSO noted Unaffected Fr immigatt “lib this 1 21mgum 0 RCC pas- or “D POszbie 1 Transverse Jointed RC C Pavements Joints were sawed at some of the RCC pavements at Fort Drum, NY and Ft. Hood, TX to provide a better riding surface as well as a more aesthetic appearance than the purely firnctional naturally cracked RCC pavements of other sites. In a field study [30], Nanni found that saw cut joints in RCC pavements tended to range fi'om an LTE5 of 40 to 60% while naturally cracked RCC pavements had increased LTEa’s of 60 to 90%. However, he noted that when the average slab length is in the range of 10 to 30 feet (3.0 to 9.1 m), the LTEs remains unaffected by classification as a joint or crack. He also noted that the width and aggregate interlock of the crack or joint seemed to remain unaffected by the choice to saw cut at these spacings. From the Waterways Experiment Station study, joint spacing was analyzed to investigate its effect on LTE5 as seen in Figure 22. A large amount of scatter was found with this trend (R2=0.03) as well as a large range of LTEs. This may be due to the small amount of data available fiom the Waterways Experiment Station study in [l] on jointed RCC pavements. Each point found in Figure 22 represents one joint tested with the FWD or HWD instead of a set of data as with the naturally cracked RCC pavements. Another possible reason for the scatter is the lack of joint spacing data found on joints tested fiom Ft. Drum, NY with high LTEs’s. These data points could have helped form a better trend than the one presented in Figure 22. The Edmonton, AB study discussed in the next section deals primarily with saw cut joints and their effect on load transfer in RCC pavements. 62 ls'l‘Ea (o/O) 0 10 20 30 40 50 60 70 80 Joint Spacing (ft.) Figure 22. Relationship Between Average Joint Spacing and LTE5. Edmonton, AB Study Trends The study from the Edmonton, AB RCC pavement is a more controlled analysis on load transfer than that of the Waterways Experiment Station. It focuses on the effect of joint spacing on load transfer without great changes in RCC thickness, support conditions, and traffic that are associated with the Waterways Experiment Station study. While most of the data deals with jointed RCC pavements at different joint spacing, some of this study dealt with naturally cracked RCC pavements in order to test the effect of saw cutting on LTE5 on a gross basis. 63 T ranmrse J The l and 49.2 fee matings ten. :3 shows the With increase illt - Ava-naus- LTD-La ("/a) 1 Transverse Jointed RCC Pavements The Edmonton, AB study focused on four distinct joint spacings, 14.8, 21.3, 32.8. and 49.2 feet (4.5, 6.5, 10.0, and 15 m) to test its effect on load transfer. These joint spacings tend to agree to the range that most rigid pavements have designed for. Figure 23 shows the effect of joint spacing on LTEa. A clear trend of decreasing LTE5 is noticed with increased joint spacing although the levels of LTEa is fairly low overall. 1 foot=.305 m Average LTE, (%) ts 20 10 fi 14.8 21.3 32.8 49.2 Joint Spacing (feet) Figure 23. Relationship Between Joint Spacing and LTEgat Edmonton, AB Site. Radius of relative stiffness was found to have an effect on load transfer at the joints at the Edmonton, AB site. In Figure 24, an increasing radius of relative stiffness showed a decreasing load transfer efficiency. This tends to agree with the findings on natural cracks fiom the Waterways Experiment Station in the preceding section, but with less scattering of the data. lair—— «10’- ! la;“ (‘7. ) Figure 24. Rel From 1) reaction Was {0 palmems. H( e133; [Md “as mkhed as 5 In the F SlllTnegs lEhi‘) , SIM on the two lailflbles a is on the order when. 100 *1 —— m—w— — 7 — . , l l l 2 l 90 . * a, I L l ,0 ,7 - t g l a l 70 I; \ .. 4' l . l O . \ l 60 a. e 9 . a i A: l : 0 . 1 e l . ,9 50 - e ‘ III . ‘“‘ l 40 ‘4 l :T t I 30 T ,l ; I l | l 20 l in=25.4 mm 10 l 0 l A. . , 15 18 21 24 27 30 Radius of Relative Stiffness (in.) Figure 24. Relationship Between [ and LTE5at Edmonton, AB Site. From the Waterways Experiment Station study, an increased modulus of subgrade reaction was found to have positive effect on load transfer in naturally cracked RCC pavements. However, with the high amount of variation in the results from that study, no clear trend was established. From the Edmonton, AB study, no clear trend was again established as seen in Figure 25. In the previous section on naturally cracked RCC pavements, increasing slab stimiess (Eh3) was found to have a negative effect on LTEs. A brief analysis of slab stiffness on the RCC pavements from this site shows a good correlation between these two variables as seen in Figure 26. Since the elastic modulus value used in this analysis is on the order of 106 while the RCC thickness value is much smaller value, the elastic modulus is the controlling variable in the computation of the slab stiffness. 65 10. . to - in. AI\.V C... I .- -. N. “mow The ex... .- 100-“.-- - »; - , .L- ‘lT‘ ~ ~ ,, --.- -- l l l l 4 . _..___ l.— -——_J ¢ 0 -.__-_4 .__._ _.__4...__.__ _ . 70 ° O 0 O Jt. ..-_. 50 ¢ T LTE, 1%) 40 30 20 1 psi/in=.27 kPa/mm 150 200 250 300 350 400 450 Modulus of Subgrade Reaction (psi/in.) Figure 25. Relationship Between k and LTE5 at Edmonton, AB Site. 100 .,._..___ .__2_.. - _....__fi_ ..._ - ---- an ._ m- ,----.. . -- . ---. ”a. '1 -.--... I- - . . a 70 \~ 42 50 LTE, (%) 30 20 l lb*ineh=l .13 llsl*m 10 O 200 400 600 800 1.000 1.200 1,400 1.600 1,800 RCC s1.» Stifl'ness (10‘ Ib‘incbes) Figure 26. Relationship Between Eh3 and LTES at Edmonton, AB Site. 66 ll 1 noisier at small inure loaruudes studies in 1 similar not absolute 0 therefore ( Transt'ers It is generally thought that increasing thickness results in an increased load transfer at a crack or joint. A brief finite element analysis using ISLAB2000 shows a small increase in the calculated LTE5 when the RCC slab thickness is increased. Ioannides [27] reports that an increase in slab stifliress generates a lower LTEa. Other studies in [31, 32, 33] presented both numerical and experimental data that conclude similar notions on slab stiffiiess. It is noted however that as the LTE5’s are decreased, the absolute deflections will be reduced thereby increasing pavement life. The lower LTE5‘s therefore can be sustained over a longer period of time. Transverse Cracked RC C Pavements A few sections of the Edmonton, AB RCC pavement site were allowed to crack naturally as a comparison to the jointed RCC pavements with respect to load transfer. While no crack spacing data was available, the jointed RCC pavements averaged an LTE5 of 68% while the naturally cracked RCC pavements at this site averaged an LTE5 of 59%. This would tend to contradict the findings of Nanni [30] on load transfer in cracked versus jointed RCC pavements which were discussed earlier in this chapter. 67 OVERVIEW ”115 me influence the fa literature gm TESPCCI 10 lllt‘ e, IlliliOlms 0r era the willth 2 [fleet of Load Load tr SirEss When in Eliminator load Placemen 10mm Crack there a Regal higher man 1h. ctack. How - CHAPTER IV - RCC Pavement Design OVERVIEW OF DESIGN METHODOLOGY This method of RCC pavement design incorporates many different factors which influence the fatigue life of a pavement. These factors include load on the design axle, temperature gradient within the RCC slab, foundation support, and load placement with respect to the edge of the pavement. Other parameters, such as load transfer efficiency at the joints or cracks, slab length, and slab width, were also investigated for their impact on the nugnitude and location of the critical stress in a RCC pavement. Effect of Load Positioning on Tensile Stress Load transfer efliciency was found to only aflect the magnitude of the tensile stress when in the vicinity of the joint or crack. This range in most cases was approximately 5-7 feet (1.5-2.1 m) fi'om the discontinuity in the pavement. The critical load placement was found in virtually all cases to be midway between the transverse joints or cracks along the edge of the pavement as seen in Figure 27. In some cases where a negative temperature gradient (temperature at the bottom of the RCC slab is higher than the top of the slab) existed, the critical stress was due to a load at the joint or crack. However, this occurred so infrequently that it should not be considered in the design of RCC pavement thickness. In these few cases, the critical tensile stress occurred at the top of the RCC slab whereas the vast majority of scenarios resulted in a critical tensile stress at the bottom of the RCC slab. 68 --__._. - -- - ..l 2.- ___..._ ...- “--.- __.__._L._ ...... , -..“, ----_- -- +,--_, .. .l._.__,-._._ ._ _. .-. _.-.. .fle-.. -2..-,..._--__..__L _- ...I. -10.... ---_.- -. .-._._.__-- .----1 - ...-. - . _. ...--- .-- . .. _ «-..—$-1rafficw---_--.. .___-,_-----.---.H_-_..-._. Tf‘""““‘j’- g::.- ._ - ---_ $ tans-rem pm” f’jjf“Longitudinalf 12 " “ : “Joint--1;- -.-— __ Critical Load 1 foot "305 m Position < b 15’ Figure 27. Slab Dimensions and FEA Mesh for Critical Load Position at Edge. Effect of Slab Dimensions on Tensile Stress Slab length had an effect on the magnitude of the critical stress when the slab length was at an extremely short distance. When slab lengths were less than 15 feet (4.6 m), the critical stresses tended to decrease slightly. For slab lengths greater than 15 feet (4.6 m), critical stresses tended to remain constant. This supports the findings of Ioannides and Korovesis [27] who claimed that with a slab length (L) divided by radius of relative stifliiess (I) value greater than 5 under filll slab contact support conditions, the responses of the pavement approach infinite slab-fike conditions. For a common radius of relative stifliress of 36 inches (914 mm), the slab length would be required to be equal to or greater than 15 feet (4.6 m) to meet this criterion. This study showed that even with a low radius of relative stifliiess of 25 inches (635 mm), shorter slab lengths had little effect on the critical stress level. With the vast majority of natural crack spacings in RCC 69 pavements gre than 25 inches fatigue life (1851 Slab oi. location as slab restrictions on 1 much smaller rs. nructure. In all sheet on critical that a slab width all“ Pal‘ement mm OlTElzitit e (1&le below me finds in [:7] OlFElatitre 51]]:an kid to adl'eTSelt pavements greater than 15 feet (4.6 m) with a radius of relative stiffness (6) values greater than 25 inches (635 mm), slab length does not appear to be a major consideration in fatigue life design. Slab width appears to have a similar effect on the critical stress magnitude and location as slab length. However since practical restrictions from truck widths place restrictions on the slab width to minimum limits of 10-12 feet (3.0-3.7 m) in most cases, a much smaller range of values were examined for the effect in critical stress in a pavement structure. In all cases examined, increases in slab length over 12 feet (3.7 m) had no effect on critical slab stress magnitude or location. Ioannides and Korovesis [27] found that a slab width (W) divided by radius of relative stifliress (8) value less than 4 tended to affect pavement performance parameters such as load transfer. Again using the common radius of relative stifl‘ness of 36 inches (914 mm), the slab width would have to be decreased below 12 feet (3.7 m) to adversely affect the pavement structure according to the study in [27]. This corresponds well with findings from this study. With low radius of relative stiffness (E) values, slab widths would need to be reduced to an impractical level to adversely affect the critical stress magnitude. GUIDE TO RCC PAVEMENT STRUCTURAL DESIGN A “new” method for RCC thickness design is described in this section. This method incorporates axle load and configuration, foundation support, temperature gradients in the RCC layer, and the effect of load placement in the design. A flowchart ofthe design methodology can be seen in Figure 28. 70 .289:— =wfioa 2.2—.93.— DUfi 2.52.. ..e Essa—=02 .8.— tasu Be...— .an 9.53..— . mummocv—OEH MO 03351.0 W 5101.0 _ u—SE—Sbififl _ lV Hem a8 3325 .Sm 325—3.: - F - aces? 32.25 as... Stag—$5 8205.25. .3 20680 4 3m com 8385 238383. :33»??? + . mmobm AMOEV 8:82.: o:_a> :0: 03952? me 3:602 c: 5:83— a . $305.25. me A t % oceawpsm lV “om Em mommobm 29 m-Ehouaz .«o 3:602 ofiscfiom 33% 3.25 uhcaazm - Susanna ammon— ~26 =¢Sac==ch got— 3834 032. $609 mcozuoaom me a - w a . a _ 2E; L 71 This section describes the methodology and an example design of an RCC pavement thickness for the following conditions: RCC pavement to be used as an access road for a factory, servicing mainly trailers 100,000 allowable repetitions of 18 kip (160.1 kN), dual-tired axle for design life Average modulus of subgrade reaction of 250 psi/in (67.9 kPa/mm) Tires are assumed to be at a pressure of 100 psi (689 kPa) RCC flexural strength of 700 psi (4826 kPa) Assumed to be a positive temperature gradient 30% of the time, a negative gradient 15% of the time, and no gradient for the remainder of the time. Stress Ratio and Allowable Stress The first step in the design of the RCC slab thickness is to determine the stress ratio as a ftmction of load repetitions from Table 7. In this case for 100,000 repetitions, the corresponding stress ratio is 0.47. Table 7 is formulated based on the relationship between stress ratio and allowable repetitions in RCC pavements used by PCA [14] as outlined in Chapter 2. Alternatively, Tayabji and Halpenny [34] developed a similar relationship for RCC pavement fatigue relationships. Their fatigue relationship encompasses a larger range of stress ratios as seen in Table 8. A comparison of both relationships (Figure 29) shows that the PCA model for fatigue of RCC pavements is much more conservative than that of the Tayabji and Halpenny model. The Tayabji and Halpenny model will result in thinner RCC pavements. However, Table 8 is added for purposes of comparison to the PCA model and can be used in the same capacity. In the case of this example, the stress ratio would be approximately 0.65. 72 Table 7. Fat] \ 1 Stress Ra' Table 7. Fatigue Relationship (After PCA [14]). Stress Ratio . 1221:5132: Stress Ratio . 13:12:22: 0.41 465000 0.56 9700 0.42 360000 0.57 7500 0.43 280000 0.58 5800 0.44 210000 0.59 4500 0.45 165000 0.60 3500 0.46 130000 0.61 2700 0.47 100000 0.62 2100 0.48 76000 0.63 1600 0.49 59000 0.64 1200 0.50 46000 0.65 950 0.51 35000 0.66 740 0.52 27000 0.67 570 0.53 21000 0.68 440 0.54 16000 0.69 340 0.55 12000 0.70 260 ‘ Load stress divided by modulus of rupture 73 ..n f;2:Ls:::::::::\-:M Table 8. Fatigue Relationship (After Tayabji and Halpenny [34]). " Load stress divided by modulus of rupture 74 Stress Ratio ' 1:22:32: Stress Ratio ' £2222: 0.45 6800000 0.71 25600 0.46 5500000 0.72 20700 0.47 4400000 0.73 16700 0.48 3600000 0.74 13500 0.49 2900000 0.75 10900 0.50 2300000 0.76 8800 0.51 1900000 0.77 7100 0.52 1500000 0.78 5700 0.53 1200000 0.79 4600 0.54 980000 0.80 3700 0.55 800000 0.81 3000 0.56 640000 0.82 2400 0.57 520000 0.83 1950 0.58 420000 0.84 1575 0.59 340000 0.85 1275 0.60 270000 0.86 1025 0.61 220000 0.87 830 0.62 175000 0.88 670 0.63 140000 0.89 540 0.64 1 15000 0.90 435 0.65 93000 0.91 350 0.66 75000 0.92 280 0.67 60000 0.93 230 0.68 49000 0.94 185 0.69 39000 0.95 150 0.70 32000 Sure-II Rollo f I.) x 1 0f {[1 Figllr829- Cm [:5ng ea and the design TTI 0.9 0.8 0.7 0.6 05 Stress Ratio 0.4 0.3 0.2 0.1 — —<> .‘———“F——“ —-4 D ‘5 A“‘ ——-<>—--—-W—(>—-— HF. ———— L LAMA A AAAAAAA gsAs AMA; AAAJA‘ l .0900 l .OE+O3 l.0E+04 l.0E+05 l.OE-+06 l .0E+07 l .0E+08 1 05-0-09 Load Applications to Failure 1 .0901 1.0902 [opCADw‘gnCme oPCAExuapolated ATaynbjiDesignCulve ATayaljiEmIpohted] Figure 29. Comparison of PCA and Tayabji/Halpenny Fatigue Relationships. Using equation (17), the allowable stress can be determined from the stress ratio and the design modulus of rupture of the RCC for both fatigue relationships. 0mm.- = SR * MOR (17) where: W = allowable design stress, psi SR = stress ratio (found from Table 7) MOR = modulus of rupture of RCC, psi Using the values of SR = 0.47 and MOR = 700 psi, the allowable stress for this design is 329 psi (2270 kPa). This means that to achieve the desired 100,000 repetitions of the design load, the stress caused by this load cannot exceed 329 psi (2270 kPa). Based on 75 the Ta} an incr: Tempt dmde 9 pr ,__‘) ' I”) deflec forces 1113 n on {ht the Tayabji and Halpenny fatigue relationship, the allowable stress is 455 psi (3140 kPa). an increase of 38% over the PCA allowable stress. Temperature Gradients Three temperature difierentials fiom the top to the bottom of the RCC pavement slab of 0, ~15, and +15°F (O, -8.3, +8.3°C) were selected for use in the fatigue-based RCC thickness design For simplicity in modeling, all temperatures difl‘erentials in this analysis were assumed to change linearly with depth of the RCC pavement slab. Since the temperature differentials were held constant at the three values stated above, the temperature gradients (defined as the temperature differential across the depth of the slab divided by the depth of the slab) changed as the thicknesses of the RCC pavement slabs changed. These temperature gradients lead to a curling action of the RCC pavement slab. The subgrade in this analysis was assumed to be a Winkler foundation. A Winkler, or liquid, foundation models the subgrade as a set of independent springs. The deflection of each “spring” is proportional to the force at that point and independent of all forces elsewhere [35]. By assuming an RCC pavement to be a rigid plate supported by this type of foundation under a temperature diflerential, additional stresses will transpire on the slab that would not occur with exterior loads only. Downward curling occurs when the top of the slab is warmer than the bottom (positive gradient). In this case, the top of the slab elongates and consequently curls downward. The springs on the edges of the slab are in compression and push the slab upward, while the interior springs are in tension and pull the slab downward as seen in Figure 30. This results in compression at the top portion of the slab and tension in the 76 bottom portion of the slab. The maximum compression and tension occur at the extreme top and bottom of the slab, midway between the edges. The reverse situation (negative gradient) can be seen in Figure 31. The points if tension and compression are the exact opposite in this case. All of these situations assume that no curling is evident when no temperature difierential is present in the RCC pavement slab [35]. 77 1 Compression Tension Compression Figure 30. Downward Slab Curling due to Positive Gradient. \ Tension Compression Tension Figure 31. Upward Slab Curling due to Negative Gradient. 78 Lateral Load Location Most RCC pavement design methods based on stress calculations assume that the load positioning is either at the critical location at the edge of the pavement slab or at a point near the interior of the pavement slab, which would result in a lesser stress under the same load. In reality, the location with respect to the longitudinal edge (or shoulder) varies with each pass of an axle due to drifting of the vehicle. Over time. the accumulation of these passes result in a lateral load distribution as illustrated in Figure 32. Figure 32 represents a normal distribution with an expected value (E[x]) of 24 inches (610 mm) and standard deviation (cm) of 10 inches (254 m) from the slab edge on a 12 foot (3.6 m) lane width. The zero point on the x-axis represents the slab edge and all positive values along the x-axis represent distances fi'om the slab edge which are located on the mainline slab. This distribution is an amalgamation of many field tests [36, 37, 38, 39, 40, 41] which attempted to characterize the lateral load distributions of many different vehicle types on both concrete and asphalt pavements. From Figure 32, it can be seen that a vast majority of vehicles remained more than 24 inches (610 m) away fi'om the pavement edge, thereby reducing the stress induced by the axle load on the pavement slab. However, some axles drifi towards the slab edge causing an increased tensile stress. This drifiing towards the slab edge would underestimate the design stress if an interior slab loading conditions were employed. Conversely, if the edge loading condition were used for design, it would overestimate the design stress and result in a conservative slab thickness. 79 plnulbllt‘) Probability ~10 -20 -30 Location of Wheel Edge Left of Slab Edge (inches) Figure 32. Typical Probabilistic Lateral Load Distribution on 12’ Wide Lane. To assess the effect of lateral load distribution on RCC slab thickness design. a level of reliability must be chosen as defined in equation (18). R =(1-P[F])*100% (18) where: R = reliability, °/o P[F] = probability of failure The higher the level of reliability is, the lower probability that the level of design stress will be surpassed. For instance, if the chosen level of reliability is 70% (or 0.70), then the probability that a random wheel positioning would be closer to the edge than designed for would be 1 - 0.70 = 0.30 or 30%. The level of reliability chosen for any particular pavement design should be related to the use and expected life of the facility in question. 80 If a pavement is to be used as low volume roadway with little significance then the level of reliability should be low although in no case should a pavement be designed below the 50% reliability level. Conversely, if the pavement is to be traversed by vehicles requiring a better ride quality, then a higher level of reliability should be designed for. Figure 33 represents the level of reliability for the lateral load distribution data shown in Figure 32. A level of reliability of almost 100% is achieved if the edge loading condition is utilized. On the other hand the reliability level is only 50% when using E[x] = 24 inches (610 mm) from the slab edge. Level of Reliability 70 6O 50 40 30 20 10 0 -10 -20 -30 Location of Wheel Edge Left of Slab Edge (inches) Figure 33. Level of Reliability for Load Positioning on 12’ Wide Lane. The next step in this design would be to choose the level of reliability with respect to the lateral load distribution using Tables A-l though A-S and/or the corresponding figures in Appendix A In the example problem, the design load is an 18-kip dual-tire axle so Table A-2 or Figures A-7 through A-12 should be utilized. Table 9 gives a 81 summary of the levels of reliability for different distances from the pavement slab edge as well as a list of “c” values which are associated with certain levels of reliability. These “c” values can be used in conjunction with Appendix A to design RCC pavement thicknesses for certain levels of lateral load reliability. These “c” values in Table 9 were developed through a series of ISLAB2000 iterations to determine the effect of the lateral load placement on the critical tensile stress and are only approximate values. The “e” value is defined by equation (19). 0: (OR “024) (dodge-024) where: c : 0R 024 Cerise Table 9. Reliability Levels for Different Wheel Load Locations. “e” value, from Table 9 (19) tensile stress for reliability level in question, psi tensile stress for load placed at slab edge, psi tensile stress for load placed 24” (0.6 m) from slab edge, psi l in=25.4 mm Distance 01' Wheel Edge . . . , n n from Slab Edge (inches) Level of Reliability ( /o) c Values 24 50 0.000 17.25 75 0.065 ‘1 90 0.163 7-5 95 0.313 0-75 99 0.900 0 99.5 1.000 82 Determination of Pavement Thickness The final steps can now be taken from the example problem cited earlier in this chapter. The allowable stress found earlier using the PCA fatigue relationship was 329 psi (2270 kPa). This should be entered in a worksheet like the one found in Table 10 along with the modulus of subgrade reaction. In this case, the value of k was assumed to be 250 psi/in (67.9 kPa/mm). A level of reliability must be chosen based on the importance of the facility. Since this pavement is to provide service to trailers primarily, a level of reliability for the lateral load distribution will be assumed to be 95%. This corresponds to a “e” value of 0.3 l 3 according to Table 9. The next task is to input the critical tensile stresses for the specific design wheel load and modulus of subgrade reaction for the problem. These stresses should be entered for both the edge loading condition and the condition where the load is two feet (0.6 m) from the edge if the lateral load distribution is to be considered. In this example, these values can be found in Table A-2 of Appendix A and have been entered into the RCC pavement design worksheet in Table 10. A blank worksheet for fiuther computations of RCC thickness design can be found in Table 11. Afier this has been completed, the assumption of the percentage of time that the pavement will be under each temperature gradient must be decided. For this example, the problem statement assumes that a positive temperature gradient occurs 30% of the time, a negative gradient occurs 15% of the time, and no gradient occurs for the remainder of the time. While positive and negative gradients do exist virtually all of the time in concrete pavements, the assumption that no gradient occurs 55% of the time helps balance out 83 period areal 51E Iii periods when the assumed temperature difierential is greater than reality. These values are also entered into Table 10 near the top of the table. The next step is to use these temperature difierential assumptions to compute the weighted values of tensile stress for all thicknesses and in both loading position cases as seen in Table 10. After this, the difierential between the weighted tensile stresses for each loading position can be computed by subtracting the weighted stress when the load is two feet from the edge from the weighted edge stresses. These values were found for all the RCC pavement thicknesses and entered in the column in the upper portion of Table 10. To determine the design stresses for each RCC pavement thickness when considering the lateral load distribution, the “e” value must be used. The design stresses can be computed by multiplying the differential of the weighted stresses by this “e” value (in this case 0.313) and adding that to the weighted stress found for when the load is located two feet (0.6 m) fi'om the edge as seen in Table 10. The final step is to choose the thickness of the RCC pavement using the allowable stress value calculated earlier (329 psi, or 2270 kPa) and the design stresses for each RCC thickness found in Table 10. The design stress should not exceed the allowable stress, so the thickness should be greater than 6 inches (152 mm) according to Table 10. However, since the allowable stress is near the design stress for the RCC thickness of 6 inches (152 mm), a final design thickness of 6.5 inches (165 mm) was chosen. 84 Table 10. Worksheet for Example on Determination of RCC Thickness. 1 in=25.4 mm 1 psi=6.89 kPa 1 psi/in=0.27l kPa/mm ATncc "’1’ 15"F 45°F Differential , Weighted Between Design 0101 11;“ Stress Weighted Tensile n .er Value Stresses from Stress Gradient 0.55 0.3 0.15 . ' . (08!) Edge and .2 (1131) RCC Thickness (in.) Away (psi) 5 336.2 483.0 250.1 367.3 180.6 423.9 0 6 258.0 404.0 203.9 293.7 135.4 336.1 60 B 7 206.1 351.5 184.2 246.4 102.8 278.6 a g 8 169.8 315.8 176.4 214.6 84.4 241.0 7; 9 143.1 290.2 168.2 191.0 71.1 213.3 '53 11 107.0 255.5 152.8 158.4 53.0 175.0 4 13 84.0 231.7 149.7 138.2 40.7 150.9 15 68.3 212.8 144.9 123.1 32.8 133.4 5 528.0 659.0 399.0 548.0 k (psi/in.) 250 6 409.0 541.0 279.0 429.1 Lateral Load 95 g. 7 328.6 462.0 199.1 349.2 Rehablhty (°/°) If: 8 271.1 407.0 185.2 299.0 "c" Value 0.313 .22 .3 9 228.4 367.7 174.5 262.1 Anowablg 329 3 11 170.2 314.0 157.3 211.4 Stress (P51) 13 132.8 278.0 149.7 178.9 Design Th' kness (in) 6'5 15 107.3 250.5 144.9 155.9 '¢ 85 Table 11. Worksheet for Determination of RCC Thickness. 1 in=25.4 mm 1 psi=6.89 kPa l psi/in=0.27l kPa/mm ATncc 0"F 15°F -15°F Differential Between Desi n °/. Time Weighted Weighted 3. Tensile Under Stress Stresses Stress Gradient Value (psi) from Edge ( si) and 2' Away [3 RCC Thickness (in.) (psi) 5 q, 6 an ‘5'. 7 E E 8 i; 9 '8 11 3 13 15 5 k (psi/in.) 6 Lateral Load 0 Reliability .3,» 7 1%) g fl '1 .5 8 0 Value E 9 Allowable .3 11 Stress (psi) 13 Design Thickness 15 (in) 86 A catalog of designs can be found in Appendix C. These tables provide the final design thickness for 2,430 cases which are based on the following design inputs: Five axle weights and configurations Modulus of subgrade reaction (k): 100, 250, 400 psi/in(27.1, 67.9, 108.6 kPa/mm) Allowable stress (allowable): Ranging fi‘om 100 psi to 500 psi (689 to 3450 kPa) Lateral load reliability: 75, 95, 99% Percentage of time with positive temperature gradient: 0, 25, 50% Percentage of time with negative temperature gradient: 0, 25, 50% To use this catalog of design, the allowable stress must first be calculated from the from the stress ratio (Table 7) for the allowable repetitions desired and the expected modulus of rupture of the RCC pavement slab. Then using the other input parameters, the design thickness can be chosen fiom the appropriate table in Appendix C. The tables are organized by axle weight and configuration as well as the modulus of subgrade reaction. For cases where the design inputs are between the values listed in the table, a weighted thickness can be computed fi‘om the nearest thickness values to that of the desired design If the thickness values for the nearest design inputs vary greatly, the complete design method should be employed in order to find the correct design thickness. Limiting Subgrade Stress From a slab tensile stress standpoint, the load transfer efficiency at cracks and joints has little effect on the maximum tensile stress. The “maximum” tensile stress is almost always experienced when the axle is midway between the transverse joints or cracks, along the edge. When the load is at or near a discontinuity, the slab tensile stress is aflected by the load transfer efficiency but not to the point where the “maximum” tensile stress is exceeded. 87 One way that load transfer efiiciency aids in the pavement performance is by controlling the stress felt in the subgrade. When the underlying layers of a concrete pavement system are stressed, the probability of pumping in the underlying materials increases. This usually results in premature cracking in the corners of the slabs by creating a cantilever beam effect on the slab comer. The design method incorporates a design check against an excessive subgrade stress. This design check was developed using ISLAB2000 and modeling an axle load at the joint or crack. No temperature gradients were used when modeling these cases. As the load transfer efiiciency was increased, the subgrade stress was found to decrease. This efiect promotes the use of joints atdistances less than 30 feet (9.1 m) in order to keep the cracks tight and promote higher load transfer between slabs. A decrease in subgrade stress was also exhibited when the RCC thickness is increased. It is important to note that the use of retrofit dowels analytically reduced the subgrade stress for a given thickness and modulus of subgrade reaction. Typically, PCC pavements are designed to limit the subgrade stress to no more than 5 psi (34.5 kPa). From the ISLAB2000 analysis, the subgrade stress would always exceed this level under the 12 and 18 kip (53.4 and 80.1 kN, respectively) single axle loads unless it was retrofitted with dowels and may prove to be unpractical from a design standpoint. While no particular threshold is recommended to limit the subgrade stress to, this should provide a check for the designer in order to deem the design thickness as adequate for long-term protection of the underlying layers. It may also indicate the need for retrofit dowels for iii-service pavements that indicate low load transfer at the cracks or joints. The figures used to check against an excessive subgrade stress are found in Appendix E. 88 Mixed Traffic Designs When multiple axle designs are required for design of the RCC thickness, another step should be utilized in the design. To do this, the cumulative fatigue damage caused by the mixed traflic must be calculated. This step is modeled afier a Miner’s hypothesis model of cumulative damage from [14]. First, a trial RCC thickness must be chosen and then the tensile stress for each axle type and temperature gradient combination in the traflic mix must be found from Appendix A. Next, the stress ratio must then be determined from the stresses found and the assumed modulus of rupture of the RCC mix. From the stress ratio, the allowable repetitions for each axle type must be computed using Table 7. The number of allowable load repetitions for each axle design in the traffic mix should be calculated. After this, the fatigue consumption for each axle type should be computed using equation (20). N F, =—3°1—°“~“*100% (20) allowable, n where: F n = fatigue consumption for wheel load n, % N W n = expected number of load repetitions of wheel load n N “mug, n = allowable number of load repetitions of wheel load it The fatigue consumption for all of the axle types in the traffic mix should then be added. If the total fatigue consumption is greater than 100%, then the trial pavement thickness should be increased and the process should be repeated. If the total fatigue consumption is less tlmn 100%, the process can be repeated using a thinner trial pavement thickness to see what the thinnest slab thickness is possible which does not result in fatigue consumption over 100%. 89 To demonstrate this adaptation of the “new” RCC pavement design method. an example design is provided. Three axle types will be used in this example although more can be accommodated. Conunon design parameters for this example will include the following: Modulus of subgrade reaction = 250 psi/in (67.9 kPa/mm) RCC modulus of rupture of 600 psi (4140 kPa) Positive temperature gradient 25% of the time Negative temperature gradient 25% of the time No temperature gradient 50% of the time Lateral load reliability of 95% The design parameters that are specific to each axle type for this example are listed in Table 12. Table 12. Data to be Used in Mixed Traffic RCC Design Example. Design Axle Number 1 2 3 Design Axle (kips) 12 18 36 Single or Dual tired? Dual Dual Dual Single or Tandem Axle? Single Single Tandem Expected Repetitions 200,000 200,000 100,000 The first trial thickness used in this example is 8 inches (203 mm) as seen in Table 13. Using the temperature distribution and lateral load distribution for this case, the design tensile stresses are 183.0, 231.6, and 212.3 psi (1260, 1600, and 1460 kPa) for design axles l, 2, and 3 respectively. Using the modulus of rupture for this example, the calculated stress ratio would be 0.37, 0.46, and 0.42 for design axles 1, 2, and 3 respectively. Using Table 7 from earlier in this chapter, the allowable repetitions for 90 these stress ratios can be seen in Table 13. Using equation (20) and summing the results. the total fatigue consumption for all three axles is 182%, larger than the 100% limit for a satisfactory thickness design Since the total fatigue consumption for the first trial thickness was greater than 100%, the second trial thickness should be increased. In this example, the trial thickness was increased to 9 inches (229 mm) and the process was repeated as seen in Table 13. In this case, the total fatigue consumption was 43% and the thickness was deemed as the final design thickness. Table 13. Worksheet for Mixed Traffic RCC Design Example. l in=25.4 mm 1 psi=6.89 kPa Design Axle Number 1 2 3 Trial Thickness #1 (in) 8 DeSig“ Twine Stress 183.0 231.6 212.3 (ps1) Calculated Stress Ratio 0.37 0.46 0.42 Allowable Repetitions Unlimited 130,000 360,000 Fatigue Consumption (%) 0 154 28 Total F N (%) 182 Wish" tnimFih'iitiiiffiL. Trial Thickness #2 (in) 9 ”Sign Twin" Stress 165.0 206.1 190.9 (PSI) Calculated Stress Ratio 0.33 0.41 0.38 Allowable Repetitions Unlimited 465,000 Unlimited Fatigue Consumption (%) 0 43 0 Total F N (%) 43 Decision Total FN is below 100% Final design thickness is 9" 91 Comparison of Results to Other RCC Pavement Design Methods Five cases were analyzed using three separate RCC pavement design thickness methods (“new” method outlined in this thesis, PCA method, and the U.S. Army Corps of Engineers method) to make comparisons between the results of each method. The “new method” was analyzed both with and without temperature gradients using both the PCA and Tayabji/Halpenny fatigue relationships summarized earlier in this chapter. The five cases are outlined in Table 14. Table 14. Data to be Used in RCC Design Comparisons. 1 kip=4.45 kN 1 psi/in=0.271 kPa/mm 1 psi=6.89 kN Case No. 1 2 3 4 5 Design Axle (kips) 12 18 24 30 36 3mg"? °’ Dual Dual Dual Dual Dual Dual tued? Single 2:131“ dem Single Single Tandem Tandem Tandem 1’3““??“6 200,000 300,000 300,000 400,000 1,000,000 epetltions k (psi/in) 100 250 400 250 100 RCC MOR (psi) 400 500 800 700 600 Positive gradient 25% 30% 35% 40% 45% Negative gradient 10% 15% 20% 25% 30% No gradient 65% 55% 45% 35% 25% Lfigfi’jbfii‘t’yad 75% 95% 90% 99% 99.5% Case 1 represents an RCC pavement designed under moderate loading conditions with poor subgrade support. Case 2 in Table 14 is the same problem discussed in the example design problem. Case 3 incorporates conditions with good subgrade support and the load 92 distributed under a tandem axles instead of a single axle. The conditions for Case 4 represent average subgrade support conditions and level of design traflic. Case 5 represents the severe case of the five and should result in the thickest RCC pavement thickness. The results of the design method analysis are listed in Table 15. Table 15. RCC Thickness Results of RCC Design Comparison in Inches. 1 in=25.4 mm Case No. Design Method 1 2 3 4 5 New Methoda (with temp) 8.5 9 5° 7 12 New Method‘1 (no temp) 7 8 5c 6.5 9.5 New Methodb (with temp) 6.5 7 5c 5.5 8.5 New Method” (no temp) 6 6.5 5c 5 7.5 PCA Method 8 8 5 6.5 8 U'S' $31;ng 0f 7.5 7.5 4.5 6 8 ' Using PCA fatigue relationship b Using Tayabji/Halpenny fatigue relationship ° 5 inches (127 mm) is the thinnest RCC pavement section using this method In all cases, the “new” method using the conservative PCA fatigue relationship while incorporating temperature gradients resulted in the thickest RCC pavement sections. The same method and fatigue relationship were also utilized without a temperature difierential across the depth of the pavement, resulting in 7-20% thinner pavements. Using the “new” method with the Tayabji/Halpenny fatigue relationship and temperature considerations resulted in 21-29% thinner pavements than designs using the 93 same method using the PCA fatigue relationship. When temperature was not considered. the Tayabji/Halpenny fatigue relationship resulted in 14-23% thinner pavements than those cases designed using the PCA fatigue relationship. In general, the PCA and U.S. Army Corps of Engineers RCC design methods predicted similar RCC thicknesses with the PCA design method predicting a thickness of 0.5 inches (13 mm) greater than the U.S Army Corps of Engineers method in almost every case. Both the U.S. Army Corps of Engineers and PCA design methods resulted in similar thicknesses when compared to the “new” method at 75 or 90% lateral load reliability and no temperature considerations. This seems logical since the “new” method is based on the same fatigue relationships used in the PCA design method, which also does not incorporate temperature-related stresses. A significant increase in the design thickness can be seen by increasing the reliability of the lateral load location as seen in Figure 34. It is important to note that the lateral load reliability does not represent the reliability of the entire pavements design. Variability in other design variables such as load magnitude, modulus of rupture for the RCC, modulus of subgrade reaction, etc. could lessen the cumulative reliability level of the design. In Case 1 and 2, a 43% increase in thickness was required when the lateral load reliability went fi'om 50% to 99.5%. Other cases (except for Case 3) showed similar increases in the design thickness in order to attain higher levels of reliability. Case 3 did not require an increase in thickness because for all levels of lateral load reliability, the design required was less than the minimum thickness design attainable by this method. 94 RCC Thickness (inches) Casel Case2 Case3 Cased Cases [Lateral Load Reliabilityllsosa 137596 090% 095% 139996 l99.5%] Figure 34. Effect of Reliability on “New” Method Thickness Without Temperature. If the various temperature conditions are considered using the “new” method, an average increase in thickness of 18% resulted in comparison to when the temperature considerations were not employed. The impact of the lateral load placement reliability when temperature conditions are considered on the pavement thickness for all five cases can be seen in Figure 35. Again, no change in the design thickness for Case 3 was realized because the design required was less than the minimum thickness design possible using this method. If both the temperature gradients and the level of lateral load placement reliability are considered as in the “new” method, thicker RCC pavements will be designed. However, thicker RCC slabs with the same level of aggregate interlock factor will analytically produce higher load transfer across transverse cracks, resulting in an increased fatigue life. This will be discussed more in Chapter 5 of this thesis. 95 l l in=25.4 mm RCC Thickness (inches) Case 2 Case 3 Case 4 Case 5 [Ateral Load Reliabilityll50% E17596 090% 139596 12199-4 .99.5% Figure 35. Effect of Reliability on “New” Method Thickness With Temperature. 96 - CHAPTER V - Investigation of RCC Pavement Alternatives DESIGN ALTERNATIVES Although load transfer efficiency was found to have little effect on the critical tensile stress to be used in a fatigue-based design, a high load transfer aids in the reduction of faulting and Spalling potential at the crack or joint. From an initial design standpoint. two alternatives exist for increasing load transfer at a transverse crack or joint. The first alternative includes saw cutting joints at closer spacings than natural cracks spacings in order to decrease crack widths while the second alternative involves increasing the design thickness so that the stress can be distributed over a larger depth. Engineered Joint Spacing As discussed in Chapter 3, RCC pavements that are allowed to naturally crack in the transverse direction often crack at spacings much larger than those of conventional concrete sections (typically 15 to 30 feet, or 4.6 to 9.1 in). With thermal and volumetric changes of the RCC pavement, large crack widths result which are not conducive to effective load transfer. This low load transfer, in turn, results in high deflections and tensile stresses as a load passes over the crack resulting in premature fatigue damage as well as an increased potential for faulting. Water and other incompressrbles can easily enter the crack and underlying layers of the pavement resulting in pumping of the base material and locking up of the crack. When incompressibles enter the joint or crack of a rigid pavement, there is a loss of movement between the slab and stress concentrations 97 occur near the discontinuity. In combination with thermal expansions and/or exterior loadings, these stresses can surpass the strength of the concrete material and cause the joint or crack to spall. When pumping of the underlying materials occurs, a loss of support usually occurs on the leave side of the crack or joint. This loss of support can cause a cantilever effect of the concrete slab, thereby increasing tensile stresses and causing corner cracks. These phenomenon can be avoided if addressed in the design and maintenance of the pavement. If joints are engineered at spacings less than 30 feet (9.1 m), data from Chapter 3 and Appendix D has shown that crack widths have been fairly tight and the resulting load transfer efiiciencies have been above an acceptable level (generally over 80%). Joints are normally sawed after the RCC has begun the hydration process and as closely timed to alter the rolling process has occurred. This technology requires water to cool the saw blade since the RCC has already been rolled and achieved considerable strength. Another way joints may be cut involves the use of a vibrating plate (Figure 36) at a very early age before final rolling of the RCC pavement. This technology has produced effective joints in RCC bases in the United Kingdom as seen in Figure 37. To do this, the vibrating plate is used to cut a groove into the fieshly laid RCC pavement. A bitumen fill is then poured into the groove using a watering can and the pavement and joint are then re-rolled. The cracks then open up after a few days due to temperature and volumetric changes. This process has the benefit of lower construction costs over the saw blade cutting procedure with similar success in forming joints in RCC pavements. 98 Figure 37. Inducing of Joints Using the Vibrating Fin in an RCC Base. 99 Increased RCC Slab Thickness Analytically for a constant level of AGG, an increased thickness will not provide a substantial increase in load transfer across a crack or joint. Under a given load however. the increased thickness will reduce the tensile stresses near the crack or joint in a rigid pavement. Figures 38, 39, and 40 represent trends developed using ISLAB2000 to determine stress levels near the joint or crack for an 18 kip (8O kN) dual-tired axle placed at that joint or crack for modulus of subgrade reaction levels of 100, 250, and 400 psi/in (27.1, 67.9, and 108.6 kPa/mm), respectively. These figures can be used to determine relative stress levels for different RCC thicknesses as related to the level of deflection load transfer efficiency. From these figures, it can be concluded that tensile stress levels decrease with very high levels of LTEa. A breaking point can be seen in all three figures. At LTE5 levels below these thresholds, LTE5 has little bearing on the tensile stress near the joint or crack. For thicker pavements, this threshold point is at levels of LTE5 greater than 90%. For thinner pavements such as the 6 and 8 inch (152 and 203 mm) thick RCC pavements, this breaking point is at a level of LTE5 closer to 70 or 80%. This would suggest that the thicker slabs would tend to have less reliance on load transfer to reduce tensile stresses. By doing this, the thicker pavements provide a more reliable method of stress reduction than thinner pavements. If the thinner pavements lose load transfer due to aggregate attrition, opening of the crack due to temperature, etc., they will undergo a much larger level of tensile stress than desirable. 100 450 i — - ‘ ”‘ ' 1 g ’ ’ g' 1 ‘ 1 psit=6.89 kPa l g 400 350 300 250 100 Principal Stress near (.‘rack or Joint (psi) 50 0% l 0% 20% 30% 40% 50% 60% 709 o 80% 90% l 00% Deflection Load Transfer Efficiency (%) +6 inches +8 mics—:0 niches + l2 inches + l4 niches Figure 38. Effect of LTE5 on Tensile Stress near the Crack or Joint for k=100 psi/in (27.1 kPa/mm) and No Temperature Gradient. 400 350 300 250 200 150 Principal Stress near Crack or Joint (psi) 50 0% l 0% 20% 30% 40% 50% 60% 70% 80% 90% l 00% Deflection Load Transfer Efficiency (%) :- 6’ inches +81nches fe— 101nches -it_—_l 2—1110li—cs—: 1:61;th 1 Figure 39. Effect of LTE; on Tensile Stress near the Crack or Joint for k=250 psi/in (67.9 kPa/mm) and No Temperature Gradient. 101 400 350 1.3 O o N kl! O N O O G o 100 Principal Stress near (Track or Joint (psi) kli O Oo/o 10% 2090 30%) 400/3 5011/0 609,0 709/0 800' o 9090 1 00% Deflection Load Transfer Efficiency (%) +6 inches +8 inches + 10 inches + 12 inches + 14111che5 Figure 40. Effect of LTE5 on Tensile Stress near the Crack or Joint for k=400 psi/in (108.6 kPa/mm) and No Temperature Gradient. DOWEL BAR RETROFITTING AS A REHABILITATION ALTERNATIVE After an RCC pavement is constructed, few options exist for increasing load transfer across a joint or crack. An economical and reliable option is dowel bar retrofitting (DBR). This section aims to study the benefits of DBR as both preventive and corrective maintenance on transverse cracks in RCC pavements. Using the finite element computer program ISLAB2000, iterations were made to model cracks with aggregate interlock as the sole means of load transfer at various levels of AGG. The same cases were then analyzed with dowel bars inserted so that load transfer could be achieved through aggregate interlock as well as through the dowel bars. In both cases, load transfer 102 emciency and the critical tensile stress were computed in order to examine the immediate theoretical benefits of retrofitted transverse cracks in RCC pavements. To test the validity of these theoretical benefits, data from FWD tests of actual PCC pavement dowel bar retrofitted sites in both Michigan and Washington was utilized. The field tests were found to correlate well with the theoretical prediction of load transfer efficiency in most cases. Cases where the theoretical and field data did not match well were investigated for possrble explanations of disparity. In conjunction with long-term performance data on dowel bar retrofitted sites, this study should help pavement engineers develop a better understanding of the pr0per timing of dowel bar retrofits. Overview Load transfer across transverse cracks is critical to the maintenance of satisfactorily performing RCC pavements. In an ideal situation, both sides of a crack or joint an RCC pavement should share in supporting the load as it is transferred fi'om one side of the crack or joint to the other. By doing this, deflections and their ensuing damage to the pavement can be reduced [42]. Back in 1933, Benkelman [28] noted that when the crack faces of two slabs are held firmly together, the aggregate interlock can be expected to fimction permanently as a load transfer mechanism. While this may be true to a certain extent, wear of the aggregates and opening of the joints due to time and traffic can cause loss of load transfer capacity. However, this type of load transfer is highly dependant on material properties of the concrete, such as coarse aggregate type and size, mix design, and gradation [43]. 103 Low severity shrinkage cracks can also deve10p into structural fatigue cracks through a loss of load transfer. Raja and Snyder [44] note that abrasion of the crack faces and opening of the cracks over time can lead to intrusion of water and incompressibles into the cracks. This phenomenon can lead to a loss of load transfer, resulting in increased slab deflections, pumping, faulting, and Spalling at the crack. Loss of load transfer across transverse cracks results in increased intemal tensile stresses in the pavement and can lead to more fatigue cracking and loss of structural integrity in the pavement [45]. Just like many RCC pavements, many older PCC pavements have initially relied on aggregate interlock for load transfer at engineered joints. Transverse cracks that have formed between these joints have relied on aggregate interlock to maintain the adequacy of the pavement structure. With the United States Interstate Highway system deteriorating, a method for increasing load transfer at these joints and cracks was needed. A rehabilitation method of retrofitting dowel bars at these transverse cracks and joints was then utilized. Although dowel bar retrofitting has been in use since a 1975 German project, it was not until 1980 when the Federal Highway Administration (FHWA) awarded a research project to the Georgia Department of Transportation (DOT) to evaluate multiple devices aimed at restoration load transfer on 1-75, including dowel bars. After 10 years of performance, the dowel retrofitted sites were evaluated using an FWD. Surprisingly, many of the dowel bars were still performing at Optimum efficiency. While other dowel bars did not exhibit the same performance, the Georgia DOT was satisfied with the overall results of these first dowel bar retrofits [46]. In another study, Snyder et al. [45] also 104 concluded that retrofit dowels were most efiective device to increase load transfer in both joints and cracks of PCC pavements. Since the early 1980’s, other states have utilized dowel bar retrofitting with much success as well Over time, dowel bar retrofit construction has been refined to combat problems with overconsolidation of the concrete covering, Spalling of the sawed slots, and misalignment of the dowels. Importance of Dowel Bar Retrofitting Dowel bar retrofitting can be utilized as either a preventive or corrective maintenance tool for rigid pavements. Mamlouk et a1. [47] defines preventive maintenance as “the planned strategy of cost effective treatments to an existing roadway system and its appurtenances that preserves the system, retards future deterioration, and maintains or improves the functional conditions of the system without necessarily increasing structural capacity.” Therefore, DBR as a preventive maintenance measure is applied only to rigid pavements in a satisfactory condition in an attempt to eliminate or reduce future faulting, pumping, comer breaks, and Spalling. In the case of preventative maintenance, the dowel bar retrofit does not immediately benefit the pavement fi'om a structural standpoint after the rehabilitation in terms of stresses or load transfer efficiency. However, the deterioration of stress and load transfer emciency over time is widely believed to occur at a lesser rate, thereby increasing the structural life of a concrete pavement. While reliable deterioration models for DBR cracks do not exist at this time, field evaluation of dowel bar retrofits on a repeated basis has been occurring (most notably in Washington state [48,49]) for many years in PCC 105 pavements. This data will eventually be used in part to assess the long-term benefits of dowel bars versus aggregate interlock as a means of long-term load transfer. Laboratory testing of dowel bar retrofits in cracked PCC slabs is currently being conducted at the University of Minnesota using the Minnesota Accelerated Loading Facility Mime-ALF). The goals of these tests are to determine the effects of selected design and construction variables, such as joint face texture, repair backfill material, and dowel material and length, on retrofit dowel load transfer system performance as well as to determine the variability in Minne-ALF test results [50]. The results of these tests will eventually be able to provide relative comparisons of the benefits of aggregate interlock and various design features in dowel bar retrofits. Corrective maintenance is a necessary action conducted on a pavement to increase its performance to a satisfactory level. The defining distinction between preventive and corrective maintenance is the timing of the rehabilitation as seen in Figure 41. When the rehabilitation occurs below a given threshold, it is deemed as corrective maintenance. In general, there has not been clear threshold universally accepted by the pavement community for dowel bar retrofitting [51]. The corrective maintenance procedure for dowel bar retrofitting is similar to that of the preventive maintenance process except that faulted joints or cracks may need to be diamond ground. In corrective maintenance cases, dowel bar retrofitting usually provides increased load transfer efliciency and reduced stresses at the cracks or joints, thereby increasing the life of a rigid pavement. While the performance of dowel bars which were inserted in corrective maintenance cases may differ than that of preventive maintenance cases, corrective maintenance provides an initial increase in performance and extended pavement 106 life as seen in Figure 42. The performance of the pavement for any period of time can be maximized by increasing the total area under a curve similar to one from Figure 42 [52]. In both preventive and corrective maintenance scenarios, dowel bar retrofitting provides a mechanism by which cracks and joints may have prolonged load transfer and its associated benefits. Preventive Maintenance t 0 U , , , x L W 7, ,, , = g Conective Maintenance , g Arbiu'ary Performance Threshold 1 . E . a. b 0 Ta 5 .1 "u... or was.” ‘ l V I Figure 41. Distinction Between Preventive and Corrective Maintenance. 107 Performance Before Rehabiltation 0 3 E 1:- \ E \ a. \ g \ 3 \ it " Performance Without Rehabilitation \ \ \ \ \ ”Timeorfl'l‘raffi—ch — h Figure 42. Comparison of Performance Benefits of Rehabilitation Over Time. Current Methodology of Selection For PCC pavements, most agencies select cracks and joints that are candidates for dowel bar retrofitting on a project basis. Each project would include many miles of structurally sound concrete pavement with unspalled, primarily transverse cracks that extend to the shoulder. All cracks (or joints if originally undoweled) that match these criteria would be retrofitted with dowel bars provided that the spacing allowed such rehabilitation. Severely spalled transverse cracks are normally treated with full-depth patches since the concrete near the crack is not structurally adequate to hold dowel bars over time. As of this time, dowel bar retrofitting has not been widely used as a method of 108 rehabilitation of load transfer in RCC pavements. However. the criteria for such rehabilitation would be concurrent with that of PCC pavements. Cracks or joints can be selected for dowel bar retrofitting on the basis of falling- weight deflectometer (FWD) testing. An FWD is a device that applies an impulse load, using a circular load plate, to a pavement and measures the resulting pavement deflections through a series of sensors. When one side of a crack or joint is loaded by the FWD, the other side will respond as well. The amount of response on the unloaded side of the joint or crack is proportional to its load transfer capability. The response of the unloaded side to the loaded side is termed deflection load transfer efficiency (LTEa) as mentioned in Chapter 3 of this thesis. Analytical Modeling Finite element analysis using ISLAB2000 was conducted on RCC pavement sections representing ranges of variables which are typical of all rigid pavements including: o Modulus of subgrade reaction (k): 100, 250, 400 psi/in. (27.1, 67.9,108.6 kPa/mm) Concrete thickness (h): 6, 8, 10, 12, 14 inches (152, 203, 254, 305, 356 mm) Aggregate interlock factor (AGG): 100 to 1,000,000 psi (1.0 to 6,900 MPa) Temperature difierence: 0, -15, +15°F (0, -8.3, +8.3°C) Other material and geometric properties were set at constant values for this analysis. These constant values included the following: Slab length: 30 feet (9.1 m) Slab width: 12 feet (3.7 in) Elastic Modulus: 4,000,000 psi (27,600 MPa) Poisson Ratio: 0.15 Coefficient ofThermal Expansion: 4.4r10t in/in/°F (7.9*10*’ mm/mm/°C) Unit Weight: 150 lb/ft’ (2,400 kg/m3) Dowel Bar Properties 109 Diameter: 1.25 inches (32 mm) Elastic Modulus: 20,000,000 psi (138,000 MPa) Poisson’s Ratio: 0.20 Dowel Length: 18 inches (460 mm) Locations 12, 24, and 36 inches (0.3, 0.6, 0.9 m) from longitudinal edges as seen in Figure 43. 00000 All of the pavements in this analysis assumed no contribution to the load transfer occurred from a shoulder. This is usually the case when either an asphalt or gravel shoulder is present which is the case in the vast majority of older PCC pavements and almost all RCC pavements. 36” 24” DOWEL BARS l 2” 36” l in=25.4 mm Figure 43. Dowel Bar Retrofit Locations for ISLAB2000 Modeling. To simulate the load placed by an FWD while testing for deflection load transfer efficiency, a 18,000 lb (80 kN) axle load was placed on one side of a modeled crack as seen in Figure 44. The same matrix of parameters was analyzed with and without dowel bars in order to characterize the benefits of a simulated dowel bar retrofit. To simulate the dowel retrofit, two sets of three dowel bars were placed at mid-depth in the wheel paths at 12 inch (0.3 m) spacings. The elastic modulus of the patching mix that covers the embedded dowel bars was assmned to be the same as that of the rest of the concrete. 110 .—_..-‘._.. -_..— a» ‘ “..---m- _ .. H.” .-- .-.»A--- ..- —_~—.---_—< _-... -..-2 wowfl .__.-..._ _. _ _. _. -_. ".4 .... .— ..4 ._-..___._ .._.._ l foot=.305 m Figure 44. Slab Dimensions and FEA Mesh for Joint/Crack Load Position. d + 12‘ For all cases without dowel bars, the load transfer efliciency was analytically found for different levels of AGG. The same cases were then evaluated with dowel bars and the same levels of AGG. This method of analysis assumes that no loss of aggregate interlock was achieved by cutting slots for the retrofit dowels. The load transfer efiiciencies in both cases were then compared to assess the increase in load transfer due to dowels. Figure 45 provides an example for LTEs with respect to AGG for both before and alter a dowel bar retrofit for the case of h = 8 inches (203 m), k = 250 psi/in (67.9 kPa/mm), and AT = 0°F (0°C). Figure 63 shows the principal tensile stresses near the crack for before and after a dowel bar retrofit for the same levels of h, k, and AT as in Figure 45. Ill 100°~o~ *; :1; ‘1’? {’1' I] I 1.7 Bib-er] l- 80% | ll l _ X 7096 6090 i * A / Y I l ‘ i ‘ . VT ! 1 . i I . ——~—4» __ . a-AF..— _._ P’"— ...—— /° 4090 i /7l' 3096 f i 2096 ‘//X 109o Deflection Load Transfer Efficiency (%) Lit O 09. . l.E+02 l.E+03 l.E+04 i.E+05 1.E+06 Aggregate Interlock Factor (psi) [+ Aggregate Interlock Only + AGG and Dowel Bar Ream Figure 45. Example of LTE; Increase for Given Levels of AGG. In Figure 46, there is a distinct separation between the curve representing aggregate interlock as the sole means of load transfer and the curve representing aggregate interlock with the benefit of retrofit dowels to aid in load transfer. This break represents the point where dowels aid in increasing LTE): for a given crack or joint. In this case, the break is near an LTE); of 91%. If a dowel bar retrofit occurs when LTEs is greater than this level for h = 8 inches (203 m), k = 250 psi/in (67.9 kPa/mm), and AT = 0°F (0°C), these analytical trends would suggest that no inunediate benefit in LTEs, would be realized. This is important to note for any pavement owner that is considering using dowel bar retrofitting as a rehabilitation method. 112 300 1 1 fl 1 '1‘ | —*‘r—fi‘4~—- 4—-—~~-:4~« .-.- _x._, 1 1 1 1 1 1 1 1 3 1 ' 1 1 ' 1 l | 1 l 1 1 1 1 ' 1 1 1 1 250 1 f i 4: + 4 5 l 1 1 1 1 ' “‘4 i . . 1 1 1 1 1 1 :: 200 j 4 4 4 1 a 4 . g l psi=6.89 kHa 1 1 . 2 h 1 g 150 + 1n 1 1 I.) 3. 1 1 '8 1 '5 1 L 9" 100 50 0 t 1 0°/o l 0% 20°/o 30°/o 40°/o 50% 60% 70% 80% 90% l 00% Deflection Load Transfer Efficiency (%) [+Aggtegute Interlock Only +Atter Dowel Bar Retrofit] Figure 46. Example of Reduction in Maximum Tensile Stresses By Using DBR. Figure 47 shows the same data as in Figure 45, but instead avoids the immeasurable parameter of AGG to determine LTEs after a dowel bar retrofit from the initial LTE); for the case of h = 8” (203 mm), AT = 0°F (0°C), and various levels of k. From Figure 47 (and for other similar curves for difl'erent levels of h, k, and AT), it is seen that a significant improvement in LTEs can be achieved through dowel bar retrofitting. Increases of up to 85% can be achieved in LTEg, theoretically. Figure 47 also illustrates that the increase in LTEs is dependant on the modulus of subgrade reaction. Lower levels of subgrade support generally provide greater increases in LTEg. 113 1009'. 1 ”—7- -r - 4:- " 1 " “ ‘1 “‘ " 1 1 '1? :2: 1311‘ 1‘ F11 1 j 1 94% 3A 1 ‘ 1 92 °/o I 11 14 i 1 1 90°“ 1 ' 1 I /e/ ’ 88% 1+ I / / LTE), After Dowel Bar Retrofit (%) 1 w 86°/e " 84 o/O // 87 °/o MM ‘ “...—(f 80°/0 1 0%) 10°50 20°40 30% 40%) 50°/o 6096 7096 80% 909 o 100% LTEa Before Dowel Bar Retrofit (%) b-lsioo psi +k=250 psi -e-k=400 psi] Figure 47. Example of Final LTE5 for Given Levels of Original LTEg. Field Validation of Analytical Results To justify the analytical models which were produced using ISLAB2000, FWD data from actual dowel bar retrofit sites were utilized. FWD data fiom both before and after the dowel bar retrofits were utilized fi'om both Michigan and Washington state sites. This data is fiom conventional PCC pavement sites. However, the predicted gains in load transfer should be applicable to all types of rigid pavements, including RCC pavements, if proper construction and selection criteria are maintained. Michigan Dowel Bar Retrofit Sites In the state of Michigan, the dowel bar retrofit sites include the following: 114 - I-69, Eaton County, South of Lansing, MI I—75, Monroe County, North of Toledo, OH 0 M—14, Washtenaw County, East of Ann Arbor. MI With the exception of some cracks on the 1-69 site, the Michigan sites have been primarily preventive maintenance projects with the intention to increase the LTE: of tight transverse cracks over the design period of the dowels. The trend in Figure 48 was developed by analyzing a dowel bar retrofit with h = 9 inch (229 mm), the design thickness of the I-69 pavement. By using a backcalculation method (ERES method described in Chapter 3) to estimate the modulus of subgrade reaction to model for this site, it was determined that k = 250 psi/in (67.9 kPa/mm), modeled as a Winkler foundation, was appropriate. Each triangle in Figure 48 and the following figures represents a crack retrofitted with dowels and non-destructively evaluated using an FWD. The analytical trend corresponds well with the field-tested cracks at all levels of LTEs. Unlike the other Michigan sites, several cracks on the 1-69 site exhibited low levels of LTEs before dowel bar retrofitting to help substantiate the predicted trend for the entire range of LTE); values. The I-75 site was designed to be an 11 inch (279 mm) PCC pavement with an asphalt shoulder serving as the main north-south interstate route in Michigan. Backcalculation procedures resulted in a modulus of subgrade reaction value near 250 psi/in (67.9 kPa/mm) for use in modeling this site. The trend developed using these values in modeling the dowel bar retrofit is seen in Figure 49. As a preventive maintenance section, the expected values of LTEs after the dowel retrofit should be very near the initial high values of LTEs. 115 1009.1 7. 1: 1 1 ' 1 ‘ _ 1 1 1 1 1 1 1 1 1 1 1 1 1 ' 1 1 1 ‘ . 1 1 ; \° 1 1 O . g 90% 1 1 . 1 1 1 4 13‘. 3 1 19* 3 1 1 a a: . ' 1 . 1 ~ h 8599 T r 4 $— J; a ;- 1 ‘ 1 ' ‘ .. ' 1 g ; 1 1 1 1 3 80% . i 1* i b . 0 c 1 < 1 1 a? 75% i I 70% 65% 1 1 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% l 00% LTE; Before Dowel Bar Retrofit (%) Figure 48. Field Verification of LTE5 for DBR Site on I-69 in Michigan. 100%1 ._ - — A. e, A A- -. ., . 11 2x? a a V 90% f/ C: _ M a a =4 ' o I = 8570 I To 5 a 80% h o e: < a? 75% (— u—l 70% 65% 1 0°/e 1 0%: 20°/o 30°/o 40°43 50 9/0 60%) 70°/o 80°41 90 on) 1 00°/o LTE; Before Dowel Bar Retrofit (%) Figure 49. Field Verification of LTEs for DBR Site on I-75 in Michigan. 116 The data points and trend in Figure 49 demonstrate that little loss or gain in LTEs is developed initially following dowel retrofitting in these cases. The benefits of these dowel bar retrofits are based upon long-term preservation of load transfer across these cracks. As with the I-75 site, the dowel bar retrofitting on M-l4 in Michigan was a preventive maintenance measure. The PCC layer was designed to be 9 inches (229 mm) thick with an asphalt shoulder. A modulus of subgrade reaction was found to be near 250 psi/in (68.7 kPa/mm) again using backcalculation of mid-slab FWD tests. However, Figure 50 illustrates a very poor correlation between the cracks which were tested using the FWD and the predictive trend. In over 80% of the cases, the cracks exhibited lower levels of LTE5 after the dowel bar retrofittings than before any rehabilitation occurred. 100%1 -~- w , ,_ 1 *7 firms. , ' ~— 95°/6 A A {3 . :‘J 90% /)1 g M 3; .._4_.._..—a—--"‘ . A $5 “— ' ‘ ‘ u 85% ‘ , e '3 i an ‘1 1 A 3 A A a 80% I A 1 ’ h I o g A < mi" 75% 1'. d o 70% 65% ‘ 1 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% LTE; Before Dowel Bar Retrofit (%) Figure 50. Field Verification of LTE; for DBR Site on M-14 in Michigan. A visual observation of the M-l4 site in Figure 5] reveals a problem with spalling of the grout material covering the retrofit dowels, which may have lead to this 117 discrepancy. \Vithout a solid base of concrete for the dowel bars to bear against as loads cross these cracks, the retrofits could not provide added load transfer. An investigation of the rehabilitation construction found that there was a problem controlling the consistency of the grout material as it came out of the mobile mixer. The mix was originally very dry, so water content was increased on site. This resulted in a very wet consistency with many of the fines rising to the surface afier vibration. This reduced the air voids near the surface, thereby increasing delamination potential and decreasing freeze-thaw resistance. A petrographic investigation of cores taken from the site by the Michigan Department of Transportation revealed that the top 1 to 1 V2 inches (25 to 38 mm) of the grouting material had very few air voids and led to the premature delamination [47]. 1. t 5 . . 1T1“ - .:‘.‘E‘\‘? '1‘ . i ‘1! 1 53‘: ' . ’1‘ "l ;l .l 1 Figure 51. View of Spalling in Grout Covering of M-l4 DBR Site. 118 Washington State Dowel Bar Retrofit Sites In Washington State, the Washington Department of Transportation set up a PCC Pavement Rehabilitation Test Section on 1-90 west of Cle Elum, WA which included some sections which were retrofitted with dowel bars as a rehabilitation technique for transverse cracks. This site was a 9 inch (229 mm) thick plain jointed concrete pavement resting on a crushed stone base with an asphalt shoulder [48]. A modulus of subgrade reaction of 400 psi/in (108.6 kPa/mm) was assumed since no information of the level of stiffiiess of the subgrade was found. This assumption tended to correlate well with the tested transverse cracks as seen in Figure 52 although some scatter was noticed in the data However, in general, the predictive trend did forecast the increase in LTE5 due to the dowel bar retrofits. 1009/6 ..-v~~ _ ., — ,_ .- , . .-.A. A; .-. 7 A. . . . ~ ~ -- — - -~ , -.., ,, ,,,, - A // 90% ‘ ‘ A A ‘ ‘ A - A A c 3 A '— “A54 5' A A‘ ... 80% . .- 3‘ ‘ ‘ ‘ \o A A 2.. ‘ A ‘ ‘ A E 70%: E ‘l 3 a 60% I- C a t 50%: 3 S h 400/. 0 C < I: 30% i- —l 20% 10% 00/0 1 0% l 0% 20% 30% 40% 50% 60% 70% 80% 90 “/o l 00% LTE5 Before Dowel Bar Retrofit (%) Figure 52. Field Verification of LTEs for DBR Site on 1-90 in Washington. 119 Benefits of Selective Retrofitting Analytical trends, such as the ones in Figures 45-47, can help pavement owners set thresholds for levels of LTE5 in order to gain the maximum benefit of existing load transfer at a crack. Similar figures were developed for the entire range of modulus of subgrade reaction, pavement thickness, and temperature difference across the depth of the slab and can be found in Appendix B of this thesis. As discussed earlier, the break between the two curves in the example in Figure 45 represents the point where dowel bar retrofits start to have immediate impact on LTE5 analytically. Therefore, as a corrective maintenance measure, the threshold using FWD testing for LTEa would be set at a point less than this level (in this case LTE5 = 91%). Also in Figure 45, a distinct drop in the difference between aggregate interlock only and the curve representing both aggregate interlock and dowels occurs around LTE5 = 80%. Consequently, an ideal threshold value to trigger a dowel bar retrofit would be this level of load transfer. Figure 46 can also be used as a guide to set threshold values by limiting the maximum tensile stress analytically. When the LTE, is below a level of 75-80% the critical values of tensile stress in the concrete tend to remain near the same levels. Therefore no significant benefit can be gained by waiting for the crack performance to deteriorate past those levels. A pavement owner could set the threshold near LTE5 =75- 80% for this level of h, k, and AT (when testing the pavement using the FWD) to activate a dowel bar rehabilitation action. 120 By using the predictive LTE5 trends developed using ISLAB2000, a pavement owner could use non-destructive evaluations of potential rehabilitation candidates in order to make more efficient use of existing load transfer in transverse cracks in corrective maintenance situations. This will result in construction savings by reducing the amount of dowel bars placed if every transverse crack in a PCC pavement was retrofitted. From 1993-1997, the Washington Department of Transportation granted contracts for dowel bar retrofits ranging fi'om $34.40 to $42.33 per dowel placed with a total of over 300,000 dowels retrofitted over this period [49]. Assuming that one-third of the retrofits could be avoided, a potential savings of more than $3,000,000 over this period could be achieved. 121 - CHAPTER VI - Conclusions, Recommendations, and Future Research Needs CONCLUSIONS AND RECOMMENDATIONS Backcalculation Procedures FWD data from both Pittman [1] and Wu and Todres [22] was analyzed using two different backcalculation methods for pavement support and stiffness parameters. This was done in order to verify the consistency of the results of the two methods. If the results of the two methods were analogous, then backcalculated values from these two studies would be valid in comparison. Three parameters, k, E, and 6, were of particular of interest since they have an effect of the design of the pavement. o The results of the modulus of subgrade reaction comparison tended to correlate well. The ECOPP method predicted larger values in all, but one case. The ECOPP method predicted values 4-12% greater than the ERES method. When the modulus of subgrade reaction was very high, the ERES method tended to predict greater values than the ECOPP method of backcalculation. o A fairly large discrepancy was found between the results of the elastic modulus comparison. In some cases, the difference in the two values approached 40%. In all cases, the ERES method predicted greater values than the ECOPP method. In extreme cases (very low values of E), the discrepancy between the two methods was the greatest. For values within the normal range, 3-5"'106 psi (2.1-3.4"‘104 MPa), for rigid pavements, the values correlated fairly well. 122 The backcalculation results of the radius of relative stiffness were found to correlate well using both methods. The differences between both methods ranged from 2-13% difl‘erence in results. Since the radius of relative stiffness is a derived parameter dependant on E, k, v, and 11, large differences between the two methods in E and k will result in differences in the values of 8. This was the case as the greatest difference in t’ for these two methods correlated with the same cases where the values of E difl‘ered most. In general, the results of the two backcalculation methods correlated well enough to make comparisons between backcalculated values fi'om both studies mentioned above. Load Transfer Efficiency Correlations In addition to the above analyses involving FWD data, several other analyses were perfonned on field data from the studies in [l] and [22] to view the effects of various factors on load transfer in transverse cracks of RCC pavements. Findings related to these analyses are summarized below. An approximate relationship was found to exist between the crack width and the natural crack spacing of RCC pavements. As the pavements crack at larger spacings, the cracks tended to open more resulting in less aggregate interlock between both sides of the crack. The reduction in aggregate interlock resulted in lower load transfer efficiencies and increased tensile stress at the crack. Lower load transfer efficiencies tended to add to the potential for faulting of the pavement as well. 123 As with crack spacing, an increased sawed joint spacing was found to have detrimental effects on load transfer at the joints. This was true from data from both studies [1,22] examined in this thesis. An increasing modulus of subgrade reaction tended to improve the load transfer efficiencies of RCC pavements. This was found to be true for natural cracks, but the results from the jointed RCC pavements was found to be inconclusive. This is probably due to the lack of data available on joints in RCC pavements. An increasing radius of relative stimiess tended to decrease the load transfer efficiency for both cracks and joints in RCC pavements. Increasing thickness generally results in an increased load transfer at a crack or joint. A finite element analysis showed a small increase in the derived LTEa when the RCC slab thickness was increased. Trends from naturally cracked and jointed RCC pavements showed a decrease in load transfer when the Eh3 was increased. This correlated well with Ioannides’ [27] contention as well as from results of other numerical and experimental studies [31, 32, 33]. However, Ioannides notes that as the LTEs’s are decreased, the absolute deflections are reduced and can be sustained over a longer period of time. It is generally thought that higher concrete strength and the corresponding high values of elastic modulus provide higher load transfer efficiency. However, no clear trend from this field data was found between load transfer and concrete strength or elastic modulus. 124 RCC Pavement Design A fatigue-based RCC pavement design method was developed and summarized in Chapter 4 of this thesis. In developing this design process, many conclusions were established and are listed below. With the exception of a few cases with extreme negative temperature gradients, the critical tensile stress occurred when the axle was placed midway between transverse cracks, immediately next to the pavement edge. This load position, as well as a position 24 inches (0.6 m) from the longitudinal edge, was used in the development of the fatigue-based design method for RCC pavements. Load transfer emciency had no bearing on the level of the critical tensile stress of RCC pavements. The level of LTEa had an effect on stresses only within a short distance of 5-7 feet (1 .5-2.1 m) when a load was placed at the crack or joint in question. In virtually every case, this tensile stress was less than the stress when the load was placed midway between transverse cracks and at the pavement edge. The slab length did not have an effect on the tensile stress calculations using ISLABZOOO when it was larger than 15 feet (4.6 m). The tensile stresses remained unafl‘ected by a change in slab width as long as the slab width was greater than reasonable values set by lane delineation. The efl‘ect of the lateral load distribution was noted on pavement stresses and the resulting design. Using a typical distribution for the drift of an axle, a modification to the RCC design method was incorporated. The impact of the level of reliability was demonstrated with very high levels of reliability impacting 125 the design thickness by as much as 40% over cases where the average load location and the resulting stresses were employed. The impact of temperature differences across the depth of an RCC pavement slab was discussed and utilized in the RCC design method. By incorporating temperature gradients and their resulting tensile stress increases, RCC design thicknesses increased by as much as 30% in the design examples considered. The theory of using subgrade stress as a limiting factor in RCC pavement design was incorporated. Increased load transfer efficiency and thickness helped limit subgrade stresses and influence the decision to reduce joint spacing or retrofit dowels. Joint spacing was recommended to be below 30 feet (9.1 m) in order to maximize load transfer at the discontinuities and reduce the subgrade stress. RCC Pavement Rehabilitation Alternatives The idea of engineered joint spacing was studied in order to provide tighter cracks and higher load transfer efficiencies in RCC pavements. Trends from field data in Edmonton, AB showed that shorter joint spacings resulted in increased load transfer at the joints. Naturally cracked pavements at closer crack spacings also had similar results. If joints are utilized in RCC pavements, a recommended distance less than 30 feet (9.1 m) should be employed. This increased load transfer reduced the tensile stress as the loads pass over the crack or joint, thereby increasing the fatigue life of the pavement. This increased load transfer also aids in the reduction of faulting potential at the crack or joint. 126 Figures were introduced which analytically relate load transfer efficiency, RCC thickness, and the modulus of subgrade reaction to the tensile stress felt by the RCC pavement under an 18 kip (80 kN) design load. These figures can be used as a final check for the design of the RCC pavement thickness in order to efiiciently use the predicted load transfer efficiency at cracks or joints. Raw data on load transfer efficiency for the sites in this study can be found in Appendix D. The use of dowel bars was analytically proven to be beneficial in restoring load transfer in all types of rigid pavements. Figures for a range of different pavement parameters such as h, k, and AT were introduced which can aid a pavement owner in developing performance thresholds using an FWD to optimize the timing of a dowel bar retrofit. Methods for developing these thresholds using analytical modeling were also introduced. The use of dowels to decrease the subgrade stress was also introduced. FUTURE RESEARCH NEEDS The work performed in this study revealed a few areas where future research is warranted. These future research needs are listed below. More FWD tests on RCC pavements need to be conducted in order to better clarify the trends described in this thesis. These tests need to be systematically designed to include both jointed and naturally cracked RCC pavements. A life-cycle cost analysis needs to be utilized in order to better characterize the benefit of RCC pavements as a low-cost alternative to asphalt pavements over a design life. 127 Tests should be conducted on RCC samples in order to characterize the coefficient of thermal expansion for a scientifically designed set of mixes. It is apparent that the coemcient of thermal expansion for RCC mixes tend to have unique effects on the crack widths in RCC pavements. More reliable flexural strength relationships for RCC mixes should be developed by incorporating the water-cementitious ratio, aggregate properties, density, and other factors in order to better predict the fatigue of RCC. The effect of aggregate properties on the load transfer of cracks needs to be addressed in order to better understand its impact. Better temperature profiles need to be developed in order to characterize the magnitude of the temperature gradients in RCC pavements as well as the length of periods at which each gradient is impacting the pavement. This will help pavement engineers better incorporate temperature effects in the design of RCC pavements. Dowel bar rehabilitations of cracks and joints in RCC pavements should be constructed and monitored in order to evaluate the effectiveness of such rehabilitation. Long-term benefits should also be inspected using FWD testing in order to characterize the deterioration of load transfer over time in comparison to other sites which solely rely on aggregate interlock as a load transfer mechanism. Thresholds for subgrade stress should be investigated in order to better understand its impact to RCC pavement design. 128 APPENDICES 129 Metric Conversions for Appendices l foot=.305 m l in=25.4 mm 1 psi=6.89 kN 1 psi/in=.27 kPa/mm 130 APPENDIX A: TABULATED CRITICAL TENSILE STRESSES 131 n.3— m._m_ NNON mam m— _.oo_ «— mdfl ~— Néw. a PEN a When 5 c m Son $3 was 26 2 $8 3.3 w. _ a N: : ~me 0.0% $8 3: a 3.3 3.8 3.3 $2 a 22 3mm 02 m 32 a 2.: 0.8m :2 Z a e 33 «.33. 0.3. 83 m ...S»: ...SE 632. 5::— $5 8.. u s. an" n a. 8. u e. 2: u a. ”85.2.: .11.---Hmw111mwsmw111- 1mm_.w10ma:< 1 1|11§Muwmm 1 1 . one ...... .3 ...... e. a . .... as as: 3.5 $52: 93w an S! mm 132 500 § Critical Stress (psi) ‘é’ § 7 1'11 If .-,/ 7 I A 100 I. ..h.?~ - _ 1 .hanH’J'Hrn—i 5 6 7 8 9 10 ll 12 13 I4 15 Thickness (in.) bk=100pifn— -k-250pi/'n.' - 'k=400pi/'n.1 Figure A-l. Critical Tensile Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle Located 24” (61 mm) from the Pavement Edge with AT=0°F (0°C). 500 e / §3m ~fi?‘ \ '3 ‘~1_\ \ a ~..~ G §o~>\ \ 200 L..:~‘_ \ .‘P-?~~\ .°"1T"- ~“--—4 1'D-?.~~ \ 100 ."‘-:-‘~‘I‘:~-'\_ 'o.::—.-_-?J-.-nfi 0 5 6 7 8 9 10 ll 12 13 14 15 Thicknessfln.) 1—k=100pi/'n. — -k=250pi/'n.- - «=me Figure A-2. Critical Tensile Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle ed at the Pavement Edge with AT=O°F (0°C). 133 600 1 1 500 :.-..~ 400 N \ é 1~>\ a ..EIPN\ \ g 300 .Au?‘ \ 1 -«>..\ \ U I .fit?-~\K .5 I...JP¢.‘.?.~. _ - ¥ G 200 - ..lP. 9.1-417”" *7 100 0 5 6 7 8 9 10 ll 12 l3 l4 Thickness (in.) 1—k=100pi/'IL- -k'=230pi"n.- - - k=400pi/‘n1 Figure A-3. Critical Tensile Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle Located 24” (61 mm) from the Pavement Edge with AT=15°F (83°C). 600 500 \ . x \ s \ Q‘ \‘N a m n 1“ \ 8. ‘~ \\ \a ‘ is 5 "~.\.‘ ~\ a 3m . -:?.= ~ — .0- ~~ I .... .r. -"‘n_‘-—-‘- '3 ' ' "‘1: 7-74h\ '5 """ ‘1‘- 7.11% 200 100 0 5 6 7 8 9 10 11 12 l3 l4 Thicknessfin.) 1—lrsloopirn— -lr-2sopirrn - - -lr-4oopi/'n.1 Figure A—4. Critical Tensile Stresses for a 12 kip (53.4 er) Dual-Tired Single Axle ed at the Pavement Edge with AT=15°F (83°C). 134 600 500 400 300 Critical Stress (psi) 1‘2) 100 Figure A-5. Critical Tensile Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle —-——-1r—. . ——~— —4 Many—...- - __ . Wq-ue“ P-I ”‘1... “a- ”#1-- C— 159 H..- ‘HHH 10 Thickness (in.) 1——lr=loopirnt — -lr=2sopiirn.- - -lr=4oopri/in.] Located 24” (61 mm) from the Pavement Edge with AT=-15°F (-8.3°C). 600 500 Critical Stress (psi) § § § 100 Figure A-6. Critical Tensile Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle 15 "b- F‘w u‘ .‘u‘ ”‘4‘“ I‘ll-9 H-‘ 10 Thickness (in.) ii F—lr=loopirns— -lr=2sopirns- - -lr=4oopri/'n.] Located at the Pavement Edge with AT=-15°F (-8.3°C). 135 3 l4 v.03 mes: m.o~_ Oven mdmm mama mg 02: EmN— m— on— 59: N82 coon ownm ”Ru m.NN_ adm— Nwfl n— wdo— Ohm. o9: WEN 32m «cm: 5.02 Nd: 903 S W. Max.— mé: 5%. 9va ~83 QM; mdom Ywmm Swen a M 73. «.mfi New. $.33 oHov 93¢ 02% :5 more. a m. mini ~62 mien odmv odcv odmm odOm Oman 0.3m r em. 09% odnu ©va o.mOm 0.3m odww oat” odov came 9 odmm 0.3m oemv 0N3 odmo 05.3 c.5v mem odmc m v.2; oi; mem— od—N wN—N fivmm 9N0 Owe 0.5 m— o.Nm_ ho: m5: Emma 5.78 5.03 #2. 0.3 ado n— H. mm: wNfl o9: Neva Wmmm m.§~ Nam 92: 5.02 2 .0... wt:— N.mo_ mism— mshm Noam ms; mAS ~63 ..mm: a m... as: v6: 502 58m aw; omen <02 mac— mdom a we 52: New. "wom— m.mmm mgmm QEM >63 moon Ndvm b M. odmu o.moN 0mm. 0va oéov 0.02V EEK ©me 5.8m e an msmm ..omm odvm odmv ©me odfi Em?” Nomm 0.3m m £32. ...53 ...Sa .5)»: ...SE .5)»: £52— .:SE ...SE Ad: 8v": 3N": 2: u.— 3vux awn".— 8— H: gene— awn": 2: u.— 205—3.; ...—em? u Dug—.4 mefl u Oomph he: .I. Dug—Q “.1. 00¢ dues— oi< 0.9.5 Ev— a— :a .3.— cav nouuohm iottu .Nr< 933,—. 136 600 500 c 400 é 5 x 3}", 300 ’1‘ \ 1 ‘. 2 ~,1\ E it ~\ \ U ' ‘ -- 200 .11.? \ . _ ‘ _ ‘ o- ~.?.?J>\ """ - -‘- .~ ~ 5‘ 10° “ -T.T 1?. ' ' ' - 311?. 7‘7-'-‘--T o s 7 s 9 lo 11 12 13 14 is Thickness (in.) 1—k=100pi/'m. — -k=250pi/'n. - - - k=400pV'n1 Figure A-7. Critical Tensile Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle Located 24” (61 mm) from the Pavement Edge with AT=O°F (0°C). eoo \ .. \ \\ '. \ \‘ \ C 400 “ ‘\ ‘ é ‘. \\ \ 5 x g 300 ‘ ,\ \ g 5 ‘ . \~ \\\ 5 D. ‘ '7': ‘K s 200 "’ “.1 \ ..."-1:T~~\ ' ' ' ’ ' '1‘- =P..~\~1> loo ' ' ’ ' ' ’- 0 s 7 s 9 lo 11 12 13 14 15 Thickness (in.) [——lt=loopi/'n. — —k=250pi/il.’ - 'k=400pi/::1 Figure A-8. Critical Tensile Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle Located at the Pavement Edge with AT=O°F (0°C). 137 500\ \ § 1 I I Critical Stress (psi) ‘S S 100 5 6 7 8 9 10 11 12 l3 14 15 Thickness (in.) E—kBIOOpi/h— -k=250pi/il. - - -k=400pi/'IL1 Figure A-9. Critical Tensile Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle Located 24” (61 mm) from the Pavement Edge with AT=15°F (83°C). 800 1 700 \ N\\ “ ‘~‘ I‘\‘-~“ ~ \ A ~ ...\~\ 5 .“.\\ \ 7 +'-'-?.~‘1‘~\\ “0.. ~+ s .....~‘ \ " .--~~‘ \ G 300 A ."T-_ ~—~ 200 100 0 5 6 7 8 9 10 ll 12 l3 14 15 Thickness(in.) 1—lr=looprirrn— -lt=250pirrn- - -k=400pi/iq Figure A-lO. Critical Tensile Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle Located at the Pavement Edge with AT=15°F (83°C). 138 Critical Stress (psi) bl 8 S N 8 5 s 7 a 9 lo 1] l2 13 14 15 Thickness (in.) 1—k=100psi/‘n. — —k=250pi/'m. - - - k=4oopsirn.] Figure A-l 1. Critical Tensile Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle Located 24” (61 mm) fi'om the Pavement Edge with AT=-15°F (-8.3°C). 600 300 Critical Stress (psi) 5 6 7 8 9 10 ll 12 l3 l4 l5 Thicknessfln.) L—-k=100psi/'n. — -lr=2sopri/in - - - k=400psil'n.1 Figure A-12. Critical Tensile Stresses for an 18 kip (80.1 kN) Dual-Tired Single Axle Located at the Pavement Edge with AT=-15°F (-8.3°C). 139 fiwwnwn 9399 won .z S! pm I‘ll: itiil I. Elli! I111|l|i|i |i . hem—r u UUGH< 1| Fa u an: .33 2.2 53...? .5. ..N e ..e as: .285 .326 .3. 2.3. la... n owe—2.-. -- :5 $05.35. no: 140 Critical Stress (psi) § S L) O O Thickness (in.) 1—k=100psi/'n.— —k=250pi/'IL- - -lr=4oopii/'.r1 Figure A-l3. Critical Tensile Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=0°F (0°C). 600 Critical Stress (pal) § § § ‘ \ T. \ ‘~.\\ 1 1>.\\ 1 ‘._-~§‘ \ --.T-~.~~ \.\‘ : '1>.. § _ 1 -__—__-_~~ \‘ 1 .’--'T'T-..———_r~ 6 7 8 9 10 ll 12 13 I4 15 Thickness(in.) 1—k=100psi/'n.— -k=250pIi/'n.- - -lr=soopsi/'.r.1 Figure A-l4. Critical Tensile Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=0°F (0°C). 141 600 1 1 1 1 1 1 1 1 1 1 ... .1 1 1 . 1 1 1 1 A 400 . 1 : 1 1 3 ‘5 1 U ~¥ s 3.. .\ 1 - ‘.\ 8 ..1P..~~ \ 3 -.'1p~.~—~ '5 “natttéh‘ 200 ' '1“ "f - - ‘1 --.... 100 o . 5 6 7 8 9 lo 11 12 13 14 15 Tiickness(ha) [——lr=loopsi/'n.— —lr=2sopi/is- - -lr=4oop.i/'n.1 Figure A-15. Critical Tensile Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=15°F (83°C). 600 1 500 1 F|‘\ -§¥ I 1 400 ‘~ =5 5:: s E: ‘~.k \ 1 i s 1 - - £300 ‘.‘:T~— \\ 1 - .."*-T~"1~ \ é "Oeq‘h-T_?.~:.-:~1> .. '-1--.?.l':-_—_.—_...__ 5200 ' "- ..... 100 o -s 5 6 7 8 9 lo 11 12 13 14 15 Tiickness(ht) l—k-loopi/in— -lr=250pii/in. - - - k=400pi/iq Figure A—l6. Critical Tensile Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=15°F (83°C). 142 600 500 300 Critical Stress (psi) 200 - 100 D ‘- D \ — {w fi-‘HI—r ”‘5. —-‘- 1H ~II‘1- u H “*:=:g-T=:-—-a——« 10 15 Thickness (in.) 11 I4 [—k=100pifn. — -k=250pi/°n.' - - k=400pV'n] Figure A-17. Critical Tensile Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=—15°F (-8.3°C). 600 500 § § Critical Stress (psi) § 100 / ...-”Inllu'34I=:=hE:L£Ef-"W‘ ‘-"-‘_—. Fs F.\‘ ~.Q\\ O .\ ‘\ ~.\ Q. ~ 1» . r. ‘ ‘~\'\ ..0.?.?~‘~ \¥ vvvvv .‘n ...-:3: ?.?-‘-:-- ~ - - V‘— 5 6 7 8 9 10 ll 12 l3 l4 l5 Tlflckneas (in.) l—k=100pi/'n. — —k=250pirn. - - “k=400pi/hl Figure A-19. Critical Tensile Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=O°F (0°C). 600 500 Critical Stress (psi) § § § 100 Thickness (in.) [—k=100pIi/'n. — -k=250pi/'n. - - “k=400pi/il.] Figure A-20. Critical Tensile Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=O°F (0°C). 145 __ A ‘4‘ A___1 '? i a 7.‘ \ 1 1 3 30° -r~\alf\ , g - . l"-T.7.:>"\ ,5 1 ---..Tr.?:‘-r\‘u—-',___\___M 200 ’ ' ""‘fi-n—am‘ 100 Q i o . s 6 7 8 9 10 11 12 13 14 15 Thickness (in.) [—k=roopirn — —k=250p.v1n.- - - k=400pifn.‘ Figure A-21. Critical Tensile Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=15°F (83°C). 60° 1 Critical Stress (psi) § '7’ l .' I / l I . I § 1 I. """"" -T:_‘:::\\ § 5 6 7 8 9 10 ll 12 13 14 15 Thickness (in.) [—k=100psifn. — -k=250psil'n. - - - k=4oopi/in.| Figure A-22. Critical Tensile Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=15°F (83°C). 146 l 1 l 500 f r T 1 i 1 : | l I 1 l -1, 1 “a 1 ‘ 3 : s l ' g I ‘ E . 2: K- . 1:: \‘ - . U 200 x I“. ‘ - : k- C- <21” "‘0‘ a ”...: H‘s...“ H “—0 -‘ul- 1h. I... H 100 7 o 5 6 7 8 9 10 ll 12 13 14 15 Thickness (in.) l—k=100pifn.— -k=250pi/'n.' - - k=400pi/it.] Figure A-23. Critical Tensile Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=~15°F (-8.3°C). 600 500 § .7 ~\ Critical Stress (psi) § § I l ... 4 . _* ..___—___AL_.____ ... A4__.s WG—4.———_. -1 ——4 % ~ ~-‘,____-& “..— h' HH‘H‘H" 100 5 6 7 8 9 10 11 12 13 14 15 Thickness (in.) t—wroogm — -k=250pi/i1.- - -k=400pi/'n.] Figure A-24. Critical Tensile Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=~15°F (-8.3°C). 147 .281— o.u< Eon—3h Ev— en :a ...... cue Sufism .3220 .m-< 95:. no: wis— edm. 9mg ENVN menu _.No 0.2: Wm»: m— ..NE no: as»: mgmm @ch fiomm 0.0: o.m- >62 3 osf 5.5— _.mm_ :RN _.moN o.omm 0.9: vdfl o.No_ : m. ”of vNo— m.No_ 9.2m = m V . "1’7‘- 3 o . .~ 1: T.?-—1>7\ 'E e.---?::?.?.1h_ G 200 ' ' ' ’ ' 100 0 a 5 6 7 8 9 10 ll 12 13 14 15 Thickness(in.) F—k=roop1rn.— -"k=250pi/'n.- - -k=400pifaL.1 Figure A-27. Critical Tensile Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=15°F (83°C). 800 700 § Critical Stress (psi) § .. T‘k‘ \ '41-.~~4~- \ .. ‘hC4;~~‘_ 3m '..-.:“~~ 200 100 5 6 7 8 9 10 ll 12 l3 14 15 Tlfickness (in.) 1—k=100pi/'n. — -k=250pi/'n. - - - k=400pi/il.1 Figure A-28. Critical Tensile Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=15°F (83°C). 150 500 § Critical Stress (psi) § § 100 0 N1- a." — ".’_ ...-C‘ 1..- ” -DJI—o -' “I..." ...L 8 9 10 ll 12 13 14 15 Thickness (in.) 1—k=100pi/'n.— -k=250pi/'n.' - 'k=400pi/i1.1 Figure A-29. Critical Tensile Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle Located 24” (61 mm) from the Pavement Edge with AT=-15°F (-8.3°C). 600 500 § O'itical Stress (psi) § § 100 T L h. .\\ ‘5‘ i ‘5 any? - ~~-_~ p‘fifi—‘hqbut-"fiuui 5 6 7 8 9 10 ll 12 l3 I4 15 Thickness(in.) F—k'IOOpi/h.— -k=2sop.irn.- - 'k=400pi/'n.1 Figure A-30. Critical Tensile Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle Located at the Pavement Edge with AT=-15°F (-8.3°C). 151 APPENDIX B: DOWEL BAR RETROFIT GUIDE 152 10090 W96 80% 70% 60% 40% 30% 20% Deflection Load Transfer Efficiency (%) 8 l 0% 0% 1.E+02 l.E-*03 Aggregate Interlock Factor (psi) 1.E+04 1.E+05 1+Ag_yegdelrmbck0niy-I-AGG-llDowelBtRenofl1 l.E+06 Figure B-l. LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=100 psi/in (27.1 kPa/mm), and AT=O°F (0°C). 100% _1__12 1-11. L , ,__ A . we 2. 1...,— 7" 1 ,l—b—r‘ ::b§* F: mil/o a n/ g / ’3‘ 80% F E, /' 3' g 70%» 1A 53 60% I- :3 3 50% = / [— E 40% 5 30% 3 t” a 20% / 10% / 0% 1.E+02 l.E+O3 1.1904 LE+05 Aggregate Interlock Factor (psi) l-o—ermkom-o-Aoo-anomracmgl 1.E+06 Figure B-2. LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=100 psi/1n (27.1 kPa/mm), and AT=+15°F (+8.3°C). 153 l.E+O6 —— ——1~ —~——1 11! —~ * ' —~‘- ...—pave 1 1 1 /,DL: 111/ewe— 90% LA ”‘1 .1 _ V’T § 80% 1 / z W i 70% g 60% 1 E . /1/ g m. 1 1 g 409. j .3 5 30%: § M g 20% A / 10% 090 l.E+02 1.E+03 I.E+04 l.E+05 Aggregate Interlock Factor (psi) 1+Agyogu1nubck0nly+AGGuliDoweiBlRetmfl1 Figure B-3. LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=100 psi/in (27.1 kPa/mm), and AT=—15°F (-8.3°C). 100% a 90% 80% 70% 50% 40% Deflection Load Transfer Efficiency (%) 10% 1.E+02 ’_1 ”.1 ,-..1 ' ‘_1 _1,, ,k-.. “1 P" / 1.E +03 Aggregate Interlock Factor (psi) 1.E+04 l.E+05 l.E+06 [+WW0nry+Aoo-apowaraarzma] Figure B-4. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=250 psi/1n (67.9 kPa/mm), and AT=0°F (0°C). 154 ‘00?” “'11‘1‘1T“1_“7‘ “ '1" 2.1.11--- TE sw- ”90 1 HP? 1 #r" S 8090 1 3 / .2/1 E gee—«w // .2 70% / E 60% 1 E ||: .3 .2. 5090 h P '3 40% .S c 30% ‘3 e: 20% 5 111/v 10% V 0% 1.1902 1.E+03 1.E+04 1.E+05 Aggregate Interlock Factor (psi) wabckmry+mouoowraanma1 I.E+06 Figure B-S. LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h— ” (152 mm), k=250 psi/1n (67.9 kPa/mm), and AT=+15°F (+8.3°C). 100% 90% 80% 70% 60% 30% Deflection Load Transfer Efficiency (%) 10% I.E+02 .2,“ ._ I ”‘2‘” 4"}; A / 1. E+03 1.E+04 Aggregate Interlock Factor (psi) 1.E+05 1+mey+munomraama1 l.E+06 Figure B-6. LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=250 psi/in (67.9 kPa/mm), and AT=~15°F (-8.3°C). 155 Deflection Load Transfer Efficiency (%) 100% W916 80% 70% 60% 50‘? o 40% 30% l.E+06 ~lrw—ITIT— -"“7’ "VF—""T—fi ’ ' "1" ' Hr. ”—1 l 1 9"”‘5 bq=F—;;==-" 1 11 1 I 9"”1 i * fl 1 H / ' /( p‘ /% l.E+02 1.E+03 l.E+05 Aggregate Interlock Factor (psi) [+qu‘cmmomy,+mcanowiscma] Figure B-7. LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=400 psi/1n (106.8 kPa/mm), and AT=O°F (0°C). Deflection Load Transfer Efficiency (%) 100% r ._ 1- K \\ /Ir‘ } ,/ 10% l.E+02 1.E+03 Aggregate Interlock Factor (psi) 1.E+05 Fo—meckwy-a-Aooudnowmcma] l.E+06 Figure B-8. LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=400 psi/in (106.8 kPa/mm), and AT=+15°F (+8.3°C). I “No I 'd / 1 /a#”‘ wiry-1 1: i VC- t 8090 T Y ' Mpg/'1 8;“ T I g 600’ / 1 .3: ° * ' J I : a so. i l i L 1" I 2 ’° 7 T I * E" l /i/ g 30% 5 2°“ w W 10% e/ 0% ‘ 1.E+02 l.E+03 LE+04 LE+05 l.E+06 Aggregate Interlock Factor (psi) [+Wrmomy+mouoowmawl Figure B-9. LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=6” (152 mm), k=400 psi/1n (106.8 kPa/mm), and AT=-15°F (-8.3°C). 1 00% W96 80% Deflection Load Transfer Efficiency (%) 10% 1.E+02 l.E+03 1.E+04 l.E+05 l.E+06 Aggregate Interlock Factor (psi) wam-o-Aoonwowmama] Figure B-lO. LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=100 psi/1n (27.1 kPa/mm), and AT=0°F (0°C). 157 __0__ __ -4 or -7 , l «L. T: 8090 “n 1 70% J l l 60% 50% 1 T 1 4090 1| :r vl 30% 1 L1/ 20% Deflection Load Transfer Efficiency (%) una r/K, 0% 1 1 . l.E+02 I.E+03 LIE-'04 1.5105 1.1906 Aggregate Interlock Factor (psi) b—quummomya—Aoouoowsunmg] Figure B-1 1. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=100 psi/in (27.1 kPa/mm), and AT=+15°F (+8.3°C). impaT—-~—flu—T~ ~ —~——-V« «firmw- -~~—- - uw- figmq,, H. 3:5 ans #4“ I hwy” / an; 9‘ 70% \l 60% / ZZZ / 101%. ’1 Deflection Load Transfer Efficiency (%) \ 1.5m 1.E+03 1.5m 1.1905 l.E+06 Aggregate Interlock Factor (psi) [+Ag_greglehnerbck0nly +Aoo-wowmcnma] Figure B—12. LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), F100 psi/in (27.1 kPa/mm), and AT=-15°F (-8.3°C). 158 100% ”90 70% 60% 5090 40% 305% 20% Deflection Load Transfer Efficiency (%) 10% 0% 1.E+02 1.E+03 1.E+04 Aggregate Interlock Factor (psi) LE +05 [+Amaeimabckomy-o-Aoo-uoomlaamaj LE +06 Figure B-13. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=250 psi/1n (67.9 kPa/mm), and AT=O°F (0°C). 100% ”/0 Deflection Load Transfer Efficiency (%) 8 39 10% 1.E+02 5a \\ /:1 / j: / j / I.E+03 l.E+04 Aggregate Interlock Factor (psi) I.E+05 [+Ag_greg‘elflerbck0nly+AGGdDowelBlRmfl] 1.E+06 Figure B-14. LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=250 psi/m (67 .9 kPa/mm), and AT=+15°F (+8.3°C). 159 100% ‘ TV I T' ’ FF“— —”' “ll—T" ‘T l " H1“. 11 Wfiflfifiad 90% - “C A?” g 80% r—A ”H/ /w e 1’ 70% ~3- W s m v I- . f .2 M a 50% 2 I *- / '3 409. .3 / g 30% / a 20% “U V 10% ./ 0% , l.E+02 l.E+03 l.E+04 l.E+05 l.E+06 Aggregate Interlock Factor (psi) [+Agegnclmomy+AoG-noowemcamg Figure B-15. L'I'Es with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=250 psi/1n (67.9 kPa/mm), and AT=~15°F (-8.3°C). 100%) 1 7.,___ —fif ., _ .... ...fl & 2:: ll Wye 80% r 70%) 60% 40% Deflection Load Transfer Efficiency (%) 10% 0% 1.E+02 1.E+03 I.E+04 Aggregate Interlock Factor (psi) l.E+05 l.E+06 b—Ageg‘elncbckmy-fl-AGGIIIDowiBaRwofi] Figure B-16. LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=400 psi/in (106.8 kPa/mm), and AT=O°F (0°C). 160 Deflection Load Transfer Efficiency (%) 1.5m us+os 1.12m 1.5+os name Aggregate Interlock Factor (poi) [+Ww0nly +AGGdDowelB¢RenofiJ Figure B-17. LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=400 psi/in (106.8 kPa/mm), and AT=+15°F (+8.3°C). § g L- Q E (... 'u I 3 5 1.5902 I.E+03 1.E+04 LE+05 l.E+06 Aggregate Interlock Factor- (poi) b—Ageg‘embckmy—u-Aoomnomsumal Figure B-18. LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=8” (203 mm), k=400 psi/m (106.8 kPa/mm), and AT=15°F (-8.3°C). 161 100% 90% 70% Deflection Load Transfer Efficiency (%) 10% m. 1.5+02 1.5+03 ' 1.5+04 1.5+05 1.E+06 Aggregate Interlock Factor (psi) [+Am‘cmbckoqiy+AoG-uoommama] Figure B-19. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=100 psi/in (27.1 kPa/mm), and AT=O°F (0°C). 100% 1" ' " ‘ " ‘ ”“111 r " “'1‘ L— "1" * ‘ ' é;fl_um “)9. f 80% - _’ r-u-I V/ 70% 609$ Deflection Load Transfer Efficiency (%) \\ 10% 0% ‘ 1.E+02 l.E+03 l.E+04 LE +05 l.E+06 Aggregate Interlock Factor (psi) [+WWOfly+AfiGdDowlB¢Rmfil Figure B-20. LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=100 psi/m (27.1 kPa/mm), and AT=+15°F (+8.3°C). 162 309a Deflection Load Transfer Efficiency (%) l 0% 0% LE+02 LI \‘V J .—Jr— —+ —-——-— +- oft- IJE+03 IJE+04 Aggregate Interlock Factor (psi) IJE+05 [+Wehtabck0nly-O-AGGndDoweiBaRenofl] 1J5+06 Figure B-21. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=100 psi/in (27.1 kPa/mm), and AT=~15°F (-8.3°C). 10096 9096 8096 7096 Deflection Load Transfer Efficiency (%) 1096 096 lJE+02 r——<» fl ,———- fi— 4% 17‘1" .7 .V...H ,, flaw—“‘fi 1J3+03 1J5+04 Aggregate Interlock Factor (psi) LE+05 b—me-o-Aoodoomazmfl IJE+06 Figure B-22. LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=250 psi/m (67.9 kPa/mm), and AT=O°F (0°C). 163 100% 90% 80% 70°/o 50% Deflection Load Transfer Efficiency (%) 1.E+02 1.E+03 1.E+04 l.E+05 l.E+06 Aggregate Interlock Factor (psi) b—Agugumbckom-o-Aoouoowlaamg] Figure B-23. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=250 psi/in (67.9 kPa/mm), and AT=+15°F (+8.3°C). 100% 1_-.,____T - —-— — 1 1‘ -—— -- -- we »~ — —- - w/jfij/ i u 1 j. 11 ‘1 90% " VF/r‘”” M 80% / 70% / 30% Deflection Load Transfer Efficiency (%) \ 10% 0% 1.E+02 I.E+03 I.E+04 I.E+05 I.E+06 Aggregate Interlock Factor (psi) [+Wmabckmy+momoowmanmn] Figure B-24. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=250 psi/1n (67.9 kPa/mm), and AT=-15°F (-8.3°C). 164 1009. T -—7- W94: 70% 60% 50% 40% 30% Deflection Load Transfer Efficiency (%) 10% 0% l.E+02 "11— __._1-__.. 1.E+03 l.E+04 Aggregate Interlock Factor (psi) LE +05 [+Amaelnwbck0nly +AGG-IdDowelBchuoftJ l.E+06 Figure B-25. LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=400 psi/in (106.8 kPa/mm), and AT=O°F (0°C). 100% v 1“ ‘—TT"TF"TV 7‘11?" ‘ I .a 90% 7 A u/ \' 80% L / / E‘ 70. / j“; /o 2‘ r4“ ’ P/ / Fri 60% 3— ”H E E 50% '- / ‘- A '3 40% 3 V I: a 30°/e a 20% /r 10% w? L/p—dt’wflj 0% “l 1 l.E+02 1.E+03 l.E+04 l.E+05 1.E+06 Aggregate Interlock Factor (psi) [+Wmouy-o—Aoouoomaamn] Figure B-26. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=400 psi/1n (106.8 kPa/mm), and AT=+15°F (+8.3°C). 165 100“. "—— ‘fi [ 1 * 1*? ——- "1%wjuflpb / 90°. WM” wHP/ A // //1 .\’ 80% r :. grub”? / é 60°10 i 5 / g 509. *‘ / 1 .. C 40 AI 3 // E 30% /, 20% a M M” 10% e// 00/0 1 1.E+02 I.E+O3 I.E+04 l.E+05 1.E+06 Aggregate Interlock Factor (psi) hwmbam+mamwmumal Figure B-27. LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=10” (254 mm), k=400 psi/1n (106.8 kPa/mm), and AT=-15°F (-8.3°C). 100% ~ A . - — 90% 1 m / 70% 60% / 50% m. / 30% Deflection Load Transfer Efficiency (%) 10% l.E+02 1.E+03 l.E+04 l.E+05 l.E+06 Aggregate Interlock Factor (psi) EWWOfly-O-AGG-dmwlkfl Figure B-28. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=100 psi/m (27.1 kPa/mm), and AT=O°F (0°C). 166 70% 40% 7 1 // 30% 20% Deflection Load Transfer Efficiency (%) f 10% 1‘ 0% 1 1.1902 I.E+03 1.1904 l.E+05 1.E+06 Agg'egate InterlockFactor (psi) bwmm+mumwmunmnl Figure B-29. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=100 psi/in (27.1 kPa/mm), and AT=+15°F (+8.3°C). 100% — _r.t_1-T .h m- 11'1- wabwbs"; 90% / “M1“ W0 .. ”Mn-’4?” / r / 70% 40% // m / 20% Deflection Load Transfer Efficiency (%) \ 10% 0% 1 1.E+02 1.E+03 1.E+04 l.E+05 l.E+06 Aggregate Interlock Factor (psi) F—Agreg‘eWOnly-I-AGG-flDowelBIRenofil Figure B-30. LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=100 psi/in (27.1 kPa/mm), and AT=-15°F (-8.3°C). 167 100% ””0 l I l I 80% 70% 60% 50% ———4>— 4090 u. ......4, _-_._ 30% Deflection Load Transfer Efficiency (%) N96 109-o 0°26 LE +02 LE +03 LE+04 I.E+05 1.E+06 Aggregate Interlock Factor (psi) bAm‘eIficbckOnIy-O—AGGNDOMIBUMDRI Figure B-31. LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=250 psi/m (67.9 kPa/mm), and AT=O°F (0°C). I 0090 90% T“ T“ 'l in?” “r ‘ "Tm” " 1 wer— "L“ ”“" Ar H;"w W/o 70% // / / / Deflection Load Transfer Efficiency (%) ”96 10% / 1.E+02 1.E+03 1.E+04 1.E+05 l.E+06 Aggregate Interlock Factor (psi) [+Wmouy+mcdoomaunmrfi Figure B-32. LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=250 psi/m (67.9 kPa/mm), and AT=+15°F (+8.3°C). 168 10090 - T —1——j ~-—_ 1.fl-,._F’... ”96 / \\ X \ t :\: 8096 wt I V’WV g ,4 / g 709. «r .fi V if 60% / a V a 50% c / E 400/. j g :09. Z/ a 20 /o / .A/ 10% / ./‘ m0 1 l.E+02 I.E+03 I.E+04 1.E+05 I.E+06 Aggregate Interlock Factor (psi) [+W1mabckouy+AGG-noowmunma] Figure B-33. LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=250 psi/1n (67.9 kPa/mm), and AT=~15°F (-8.3°C). 100% 90% 80% 70% 60% 50% 30% Deflection Load Transfer Efficiency (%) 10% I.E+02 l.E+03 I.E+04 I.E+05 l.E+06 Aggregate Interlock Factor (psi) L—o—Agegumbekouy-o-Aocmnowemama] Figure B-34. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=400 psi/1n (106.8 kPa/mm), and AT=O°F (0°C). 169 100% 1~-——~r-- 9096 80% 7090 609/0 50% 40% 30% 20% Deflection Load Transfer Efficiency (%) 10% // 0°43 0", at” LE +02 l.E+O3 l.E+04 I.E+05 I.E+06 Aggregate Interlock Factor (psi) [+Wmmmiy+mouoowelaam&] Figure B-35. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=400 psi/in (106.8 kPa/mm), and AT=+15°F (+8.3°C). 100961 “We 7- e1r..._.1._ .w- \\ F—‘ _ n L a! J ,/ as Ll \ 80% 70% 60% 50% 40% 30% e \_ ‘\ 20% Deflection Load Transfer Efficiency (%) 10% LIE-+02 LE +03 I.E+04 1.E+O$ I.E+06 Aggregate Interlock Factor (psi) FO-WlnerbckOnIy-l-AGG-llboweleRetmfij Figure B-36. LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=12” (305 mm), k=400 psi/in (106.8 kPa/mm), and AT-=-15°F (-8.3°C). 170 Deflection Load Transfer Efficiency (%) ”am l.E+03 1.E+04 I.E+05 I.E+06 Aggregate Interlock Factor (psi) [+wmomy +Aoouoamlssnmnj Figure B-37. L'I'Ea with Respect to AGG Before and After Dowel Bar Retrofit for h=l4” (3 56 mm), k=100 psi/1n (27.1 kPa/mm), and AT=O°F (0°C). Deflection Load Transfer Efficiency (%) 1.5402 I.E+03 I.E+04 l.E+05 l.E+06 Aggregate Interlock Factor (psi) mem+mmoowlaxmal Figure B—38. LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=14” (356 mm), k=100 psi/in (27.1 kPa/mm), and AT=+15°F (+8.3°C). 171 100% l T[ "MT- '1 F ...Hi. I l 90., l a! g w::=¢ ,. I W ,J /p/ ' H g 80% l 2" PAW / I ‘ ii 2 70% 1'.’ E t 2’ “3 ‘o i a 6m : h a 50m / I- [— '3 40% .3 g 30% 3 a 20% ’M/ A 10% 0% 1 I.E+02 1.E+03 I.E+04 I.E+05 l.E+06 Aggregate Interlock Factor (psi) [4.5m immna-Aoouoowmcmn] Figure B-39. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=l4” (356 mm), k=100 psi/in (27.1 kPa/mm), and AT=-15°F (-8.3°C). 100% M4 ”9’. 70% 60% 50% Deflection Load Transfer Efficiency (%) 10% I.E+02 _u% LE +03 I.E+04 Aggregate Interlock Factor (psi) l.E+05 I.E*06 L—o—Aggummmiy-o-Accuoommama] Figure B-40. LTEs with Respect to AGG Before and After Dowel Bar Retrofit for h=1 4” (3 56 mm), k=250 psi/tn (67.9 kPa/mm), and AT==0°F (0°C). 172 10090 m°o 70% 60% 50% 4090 Deflection Load Transfer Efficiency (%) N96 10% 0% l.E+02 m4)— - » I.E+03 I.E+04 I.E+05 i.E+06 Aggregate Interlock Factor (psi) l+wmouy+mouoowaama1 Figure B-41. LTEs with Respect to AGG Before and Alter Dowel Bar Retrofit for h=l4” (3 56 mm), k=250 psi/in (67.9 kPa/mm), and AT=+15°F (+8.3°C). 100%-a —~~~ w w- ’22" 90% w”" ,/ A / }”’M g 80% / / V P*”/ a .. 60% / s / g 50% 1 In [- 3 40% .3 g 30% 10% M”? 0% 1 l.E+02 l.E+03 1.E+04 1.E+05 I.E+06 Aggregate Interlock Factor(psi) [+W1uabck0niy-o-Aoc-uoowuB-Rma] Figure B-42. LTE; with Respect to AGG Before and After Dowel Bar Retrofit for h=14” (356 mm), k=250 psi/1n (67.9 kPa/mm), and AT=-15°F (-8.3°C). 173 1009o i_.____.l —-4 | l l l —# *4 «J .J I 909° 80% l 709. T / \ \ >- - 0 - 4.. ..__.,p__._.——- o _—.____ W 3 o \\ 2096 Deflection Load Transfer Efliciency (%) 109a 0% 1 1.5+02 1.E+O3 ‘ 1.5+04 1.E+05 1.E+06 Aggregate Interlock Factor (psi) [+meckoqry+mmoownunmi] Figure B-43. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=14” (3 56 mm), k=400 psi/m (106.8 kPa/mm), and AT=0°F (0°C). 1W0 "" fl1r__l" _ _ '——""" ---. [‘1 "'1" VET—.F " l‘ ’" ”'T‘ ' ’ 'l'r 9096 8096 m. A DVW; / 6096 Deflection Load Transfer Efflciency (%) \ m, / lJE+02 IJE+03 lJE+04 IJE+OS iJE+06 Aggregate Interlock Factor (psi) [+Aggeg‘elflclock0niy-O-AGGndDouelBaRetmfa] Figure B-44. LTEa with Respect to AGG Before and After Dowel Bar Retrofit for h=14” (3 56 mm), k=400 psi/1n (106.8 kPa/mm), and AT=+15°F (+8.3°C). 174 90% 80% 70% 60% Deflection Load Transfer Efficiency (%) LE +03 I.E+04 1.E+05 1.E+06 Aggregate Interlock Factor (psi) [+AWeIflabckOufiy-O—AGG-IdDowechRetrofa] Figure B-45. LTE5 with Respect to AGG Before and After Dowel Bar Retrofit for h=14” (356 mm), k=400 psi/in (106.8 kPa/mm), and AT=~15°F (-8.3°C). 100% 98% 96% 94% 92% 86% LTE, After Dowel Bar Retrofit (%) 84% "F' _ ‘— 1 — _ "“ ' —'_—"' WM / 0% 10% N96 30% 40% 50% 60% 70% 80% 90% 100% LTE; Before Dowel Bar Retrofit (%) [+k=ioop.i +k=250pi +k=400psil Figure B-46. Analytical Determination of LTE5 After Dowel Bar Retrofit from Initial LTE; for h“ ” (152 mm) and AT=0°F (0°C). 175 10096 oz 90 /o ..___.__fir A _fi_~T- _ .3 v c a; 85% i Mia-r” / I 1 fl .3 mo / j 5 / g 75% ”VA/J as? , / E— 7090 / 65% w% 1 0% 20% 30% 40% 50% 60% 70%: 80% 9096 10096 LTE. Before Dowel Bar Retrofit (%) [+k=roop-i +k=zsopi +k=400psil Figure B-47. Analytical Determination of LTEs After Dowel Bar Retrofit from Initial LTE5 for h=6” (152 mm) and AT=15°F (83°C). 100% 95% 90% 85% 75% 70% LTE. After Dowel Bar Retrofit (%) 65% 60% LTE; Before Dowel Bar Retrofit (%) / _ / / W‘Jfl/ 20% 30% 40% 50% 60% 70% 80% 90% 100% [+k=100pi +k=250psi +k=400pni] Figure B—48. Analytical Determination of LTE5 After Dowel Bar Retrofit fiom Initial LTE5 for h— ” (152 mm) and AT=~15°F (-8.3°C). 176 LTE, After Dowel Bar Retrofit (%) 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% LTE. Before Dowel Bar Retrofit (%) [+131di +k=2sopi +k=aoopil Figure B-49. Analytical Determination of LTEa After Dowel Bar Retrofit from Initial LTE5 for h=8” (203 mm) and AT=0°F (0°C). 100% 95% 85% 75% LTE. After Dowel Bar Retrofit (%) 65% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% LTE. Before Dowel Bar Retrofit (%) [+L—=100p.i +k=250pi +k=400psil Figure B—50. Analytical Determination of LTEs After Dowel Bar Retrofit from Initial LTE; for h=8” (203 mm) and AT=15°F (83°C). 177 100% — "‘ — “‘ _~____ T" '“—““““r—"‘-“"“T-' ' * 1 / 95% l it 90% £0: 85 “/0 1 /A/ M s Tex—4H,“ //’ 1; 80% "’i *- 75.0 ‘ [rm .2 5 70% MM 65% 60%, 1 096 10% 20°/o 30% 4096 5096 6096 70% 80% 90% 100% LTE. Before Dowel Bar Retrofit (%) [+ktloopui +k=mpsi +k=400pei] Figure B-Sl. Analytical Determination of LTEa After Dowel Bar Retrofit from Initial LTEs for h=8” (203 mm) and AT=-15°F (-8.3°C). 100% 98% 96% 94% 92% 90% 88% 86% LTE, After Dowel Bar Retrofit (%) 84% 82% 0% 10% Z)% 30% 40% 50% 60% 70% 80% 90% 100% LTE; Before Dowel Bar Retrofit (%) [+k=100pni +k=mpi +15me Figure B-52. Analytical Determination of LTEa After Dowel Bar Retrofit from Initial LTEs for h=10”(254 mm) and AT=0°F (0°C). 178 LTEa After Dowel Bar Retrofit (%) 55% LTE; Before Dowel Bar Retrofit (%) [+k=100psi +k'250pai +k=400psi| Figure B-53. Analytical Determination of LTE; After Dowel Bar Retrofit from Initial LTEa for h=10” (254 mm) and AT=15°F (83°C). LTE. After Dowel Bar Retrofit (%) 100% ~ T « 85% 75% 65% 55% 0% 10% 27% 30% 40% 50% 60% 70% 80% 90% 100% LTE; Before Dowel Bar Retrofit (%) [+k=100pci +k=250psi +k——4oop.i] Figure B-54. Analytical Determination of LTEs After Dowel Bar Retrofit from Initial LTE; for h=10” (254 mm) and AT=-15°F (-8.3°C). 179 100% 9890 96% 949. 90% 8890 869a LTE, After Dowel Bar Retrofit (%) 84% 82% 80% 096 10% 20/ 30% 40% LTE; Before Dowel Bar Retrofit (%) 50% 60% 70% 80% 90% 100% Flaw"; -o_-k=2sop.i +k=400p3| Figure B-SS. Analytical Determination of LTE5 After Dowel Bar Retrofit from Initial LTE5 for h=12” (305 mm) and AT=0°F (0°C). 100% —- 9590 90% 85% 80% 75% 70% LTE. After Dowel Bar Retrofit (%) 60% 55% 50% *C—h hum—l ,____—._._ -1, ” l/ / 65% ~ / // 0% 10% 20% 30% 40% 50% 60% 70% 80% W96 100% LTE. Before Dowel Bar Retrofit (%) [+k=100p5 +k=250pi +k=400pi] Figure B-56. Analytical Determination of LTEa After Dowel Bar Retrofit from Initial L'I'Ea for h=12” (305 mm) and AT=15°F (83°C). 180 LTE, After Dowel Bar Retrofit (%) 10% 20% 30% 40% 50% 60% 70% 80% ”96 100% LTE. Before Dowel Bar Retrofit (%) [+k=100psi +k-250psi +k=4oog1 Figure B-57. Analytical Determination of LTEs After Dowel Bar Retrofit from Initial LTEa for h=12” (305 mm) and AT=15°F (-8.3°C). 100% 98% 96% 94% 88% 86% LTE. After Dowel Bar Retrofit (%) 84% 0% 10% N96 30% 40% 50% 60% 70% 80% 90% 100% LTE. Before Dowel Bar Retrofit (%) [+k=100pci +k=250pi +k=400pil Figure B-58. Analytical Determination of LTEa After Dowel Bar Retrofit fiom Initial L'I'Es for h=14” (3 56 mm) and AT=0°F (0°C). 181 10090 1r'—— —--—~Te—— -— —~_ —l - - j— _a / 9590 i j/ A 909 * L o\° o l l / V I E 85% g // a 8090 17"” g M // "9' 759/0 l 3 /?/ a 6590 1‘ Q 6090 55% °/ 5090 1 090 10% 20% 3090 40%: 50% 60%: 7096 8090 W90 10090 LTE, Before Dowel Bar Retrofit (%) [+k=loo psi +k=250 psi +k=400 psfl Figure B-59. Analytical Determination of LTEs After Dowel Bar Retrofit from Initial LTE5 for h=14” (356 mm) and AT=15°F (83°C). 100% 1 - 95% “1% 85% 80% 75% 70% 65% LTE, After Dowel Bar Retrofit (%) 55% / J / ”w / /A/ I / M a ,T/ 0°/0 10% 20% 30% 40% 50% 60% 7090 8096 “Win a 100% LTE; Before Dowel Bar Retrofit (%) [+k=iOOpsi +k=2sopi +k=400psi| Figure B-60. Analytical Determination of LTEa After Dowel Bar Retrofit from Initial . LTE5 for h=14” (3 56 mm) and AT=-15°F (~8.3°C). 182 450 §§§§ Principal Stress (psi) 'é’ 150 100 50 . s v \\ \\ l l l a M 11 090 1 0%» 2090 3096 40%) 50%: 6090 7090 8090 9090 1 00%» Deflection Load Tramfer Efficiency (%) [+Wwonry-o—Anuoomracmil Figure B-61. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=100 psi/1n (27.1 kPa/mm), and AT=O°F (0°C). 500 450 Principal Stress (psi) 3 § § "8’ E § § § C N M m x. x h —* 0°/0 1096 20%) 30% 40% 50% 60°/o 70% 80% 90% 10096 Deflection Load Transfer Efficiency (%) [+Awegacm0nry +Anaocmracmarj Figure B-62. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=100 psi/1n (27.1 kPa/mm), and AT=+15°F (+8.3°C). 183 350 u 6 Principal Stress (psi) -- N ‘3 8 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection Load Transfer Efficiency (%) [+Assns-Ic m Only -I—After Dowel a. Retrofit Figure B-63. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=100 psi/1n (27.1 kPa/mm), and AT=-15°F (-8.3°C). 400 350 Principal Stress (psi) 0% I0% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection Load Transfer Efficiency (%) |+Wehmmy+mormrauw| Figure B-64. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=250 psi/1n (67.9 kPa/mm), and AT=O°F (0°C). 184 450 350 Principal Stress (psi) N N u 8 ‘6 8 ..a M O 100 50 ‘1‘“ x A \N \\ “—13“ i 090 1090 2090 3090 40%) 50% 6090 7090 80% 9090 Deflection Load Transfer Efficiency (%) bwwma—mmmmn] 10090 Figure B-65. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=250 psi/1n (67.9 kPa/mm), and AT=+15°F (+8.3°C). 350 300 250 200 150 Principal Stress (psi) 100 50 0% 0 ”Orr—AA // 10% 20% 30% 40% 50% 60% 70% Deflection load Transfer Efficiency (%) [+Whnarbck0nly4—Anerbode-rnstrofir] 80% 90% 100% Figure B-66. Principal Tensile Stresses at Crack or Joint Before and After DBR for h" ” (152 mm), k=250 psi/m (67.9 kPa/mm), and AT=-15°F (-8.3°C). 185 400 3 50 300 A l f, 250 . g l :3 200 J 4 4“ $ '1: ' \\. .9 E '3 1 50 On 100 50 o 2 0% I 0% 20% 3090 40% 50% 60% 70% 8090 90% l 0090 Deflection Load Transfer Efficiency (%) [+Wwonry +Anaooweraanm$| Figure B-67. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=400 psi/m (106.8 kPa/mm), and AT=O°F (0°C). 450 400 3‘. kka 350 ‘é’ \\ ‘H I l / // ”é Principal Stress (psi) an M O "é M O O 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection Load Transfer Efficiency (%) [+Aggng‘eIntu-bokQfiy-I-AfterDowelBarRetrofit] Figure B-68. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=400 psi/in (106.8 kPa/mm), and AT=+15°F (+8.3°C). 186 350 fix N 300 \ . ...—T—a————-——-I*\r WA 250 —IT ' a . r. V a 200 1:: m '3 £- 150 U .5 100 50 0 1 090 1090 2096 3096 4090 5096 6090 7096 8090 9090 10090 Deflection load Tramfer Efficiency (%) F-o—Wu-bckomy-e—Aaaoomracmn] Figure B-69. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=6” (152 mm), k=400 psi/in (106.8 kPa/mm), and AT=~15°F (-8.3°C). 300 250 ' x \ § Principal Stress (psi) '5‘. O § 096 1096 2096 3090 4090 5096 6096 7090 8090 9096 10096 Deflection Load Transfer Efficiency (%) [+Ww0nry-e—Aauomracmn] Figure B-70. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=100 psi/tn (27.1 kPa/mm), and AT=O°F (0°C). 187 300 N 8 150 Principal Stress (psi) § 1 0‘1 ' NM ‘ N-\ x ,5 H e l 4MP \ I l 0% 1 0% 20% 3090 40% 50%» 6090 7090 8090 9090 1 00% Deflection Load Transfer Efficiency (%) [+Wrnmomy—a-Aaaoowelauamn] Figure B-7l. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=100 psi/1n (27.1 kPa/mm), and AT=+15°F (+8.3°C). 250 150 Principal Stress (psi) ‘l\ k : \ \ ’— 4 4% 0% 10°/o 20% 30°/0 40%: 50°/o 60% 70%: 8090 90% 100%: Deflection Load Transfer Efficiency (%) [+Aggregate1nterbck0niy +AfterDoweiBarRetrofit] Figure B-72. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), F100 psi/1n (27.1 kPa/mm), and AT=-15°F (-8.3°C). 188 300 250 r; 200 '1. e. V 5 o g 150 . L z 3 ‘3— a. '3 .5 5 100 50 0 1 090 1 090 20% 30% 4096 5096 60% 70%» 8090 90% I 0090 Deflection Load Transfer Efficiency (%) [+mery—a-moowuracw] Figure B-73. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=250 psi/1n (67.9 kPa/mm), and AT=O°F (0°C). 300 250 . l / Principal Stress (psi) 8 § 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection load Transfer Efficiency (%) Fo-wemiookOniy-e-AnabowolBu-Rmfit] Figure B-74. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=250 psi/1n (67.9 kPa/mm), and AT=+15°F (+8.3°C). 189 250 150 § Principal Stress (psi) 50 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection Load Transfer Efficiency (%) L-t-mery-a-mmacm] Figure B-75. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=250 psi/1n (67.9 kPa/mm), and AT=~15°F (-8.3°C). 250 \x 200 \ i V i 50 ...?" ...—a— 2 F 4‘" m E 100 50 0 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection load Transfer Efficiency (%) [+wmrockmy +Aacomraumn] Figure B-76. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=400 psi/1n (106.8 kPa/mm), and AT=O°F (0°C). 190 300 250 Principal Stress (psi) -- N ‘6 8 “2' _J 0% 50% Deflection load Transfer Efficiency (%) 70% [+Aggreg‘elntubok0nIy-O-AfiuDowelBarRmflJ 80% 9090 100% Figure B-77. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=400 psifrn (106.8 kPa/mm), and AT=+15°F (+8.3°C). 250 200 150 Principal Stress (psi) 0 \ \ \\ 0% 50% Deflection Ioad Transfer Efficiency (%) 70% [+erookmy-a-Anaoowerauamfir] 80% 90% I 00% Figure B-78. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=8” (203 mm), k=400 psi/m (106.8 kPa/mm), and AT=~15°F (-8.3°C). 191 l 60 s - fi , I "N | \ 1 40 :.-.~ 120 a 3 E l V I . 9 ' I n a 1— 6 100 1 i: ‘5 W E 3' 1 £- 30 T a U .5 6‘- 60 40 20 0 1 0% 1090 2090 30% 40% 50% 60% 70% 8090 90% 10090 Deflection Load Transfer Efficiency (%) [+11%an +Anaoowcraukeuorfl Figure B-79. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), F100 psi/in (27.1 kPa/mm), and AT=0°F (0°C). 200 180 9'— 160 140 i Haw—«... a— 4— yak E 120 .1 \ m 100 7 .9- U E 80 so 4o 20 0 0% 10% 20% 30% 40% 50% 60% 7w. 30% 90% 100% Deflection load Transfer Efficiency (%) L-o-Wmony-s-mmraamn] Figure B-80. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10”(254 mm), k=100 psi/1n (27.1 kPa/mm), and AT=+15°F (+8.3°C). 192 180 160 \\ \\ 140 \Nfl I at 4 _ .7: 120 ' ' iml—fi E. l 7 l I 5 100 , b 1 "’ l 3. so "5' .5 h h- 60 40 1 i 20 j i 0 1 090 I 090 2090 30% 40% 50% 60‘% 70% 8090 9090 I 0090 Deflection load Transfer Efficiency (%) to-mey-s—mooweraamn] Figure B-81. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=100 psi/in (27.1 kPa/mm), and AT=-15°F (-8.3°C). 180 160 v - \ ”° \ 5; 120 a \ V 4_._—— i too i i: m 3. so '8 .5 Ii: 60 40 20 0 0% I 096 20% 30% 40% 50% 60% 7096 80% 90% I 00% Deflection load Transfer Efficiency (%) [+wwmy-e-Aauoowerausmn] Figure B-82. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=250 psi/1n (67.9 kPa/mm), and AT=O°F (0°C). 193 180 160 I40 120 100 80 Principal Stress (psi) 60 71 l I ' N \ \\ 1%.“... $t==.______,_.__.==\ I . l , \ 1‘ l fr T 1 l l i 090 1 090 2090 3090 4096 5090 60%» 7090 8090 9090 10090 Deflection load Transfer Efficiency (%) [+Wmm0nry-e-Aaaooweraumn] Figure B-83. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=250 psi/in (67.9 kPa/mm), and AT=+15°F (+8.3°C). 200 180 I60 140 120 100 80 Principal Stress (psi) 20 I i .- —-__._—h _ '"x-L. ‘41 096 10% 2096 3096 4090 50%) 60%: 70°/0 8090 9090 100% Deflection load Transfer Efficiency (%) [+Aggregatelntu'bck0nIy-l-AfierDowechRdrofitl Figure B-84. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=250 psi/in (67.9 kPa/mm), and AT=-15°F (-8.3°C). 194 180 I i l : I60 4 TL gv r : : A l : l N 140 T l \ c l \ ii I \ n i B ‘—i— + 1 8 100 T 1 e l ' \ W 5 a 80 ? '5' 1 g 1 -.: l 5- 60 40 20 0 1 09 0 1 090 2090 30°/0 40% 5090 6090 7090 809 o 9090 10090 Deflection load Transfer Efficiency (%) [+AggegacW0ny-a-Asuooworacmn] Figure B-85. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=400 psi/1n (106.8 kPa/mm), and AT=O°F (0°C). 200 180 160 140 120 100 Principal Stress (psi) 20 0% 10% 20% 30% 40% 50% 60% 70% 80% 9090 100% Deflection load Transfer Efficiency (%) [+Wmonry-s-mnomraamal Figure B-86. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=400 psi/1n (106.8 kPa/mm), and AT=+15°F (+8.3°C). 195 180 160 Principal Stress (psi) 2: s 23 § ‘3 § N O 0% 10% 20% 30% 40% 50% 60% 70% 8090 90% 10090 Deflection load Transfer Efficiency (%) [+meokony-a-Aaaoowera-ma] Figure B-87. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=10” (254 mm), k=400 psi/1n (106.8 kPa/mm), and AT=-15°F (-8.3°C). 120 v N 100 a 80 5 n 11 n— f 15' 60 1. 1 .9.- u .5 6'- 4o 20 0 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection load Transfer Efficiency (%) [+Amumony +AfterDowelBar-Retrofij Figure B-88. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=100 psi/1n (27 .1 kPa/mm), and AT=O°F (0°C). 196 140 F —I n t : y“\ ‘5 E: E 80 7 f ‘ k l W l ii i .5- 60 ' U .5 a. fi- 40 20 0 1 fl 0% l 090 20% 3090 40% 5090 6090 7090 8090 9090 1 0090 Deflection load Transfer Efficiency (%) [+Asarosdchtubck0nly-I—Anuooweracamaj Figure B-89. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=100 psi/1n (27.1 kPa/mm), and AT=+15°F (+8.3°C). 1 40 1 l 20 \\ \ \ A F ~4F 4———-¢— \__._ z 3 §_h ' a 80 h m E 3 60 i 40 20 0 4 0% l 0% 20% 30% 40% 50% 60% 70% 80% 9090 1 00% Deflection load Transfer Efficiency (%) [+erookmy-e—Aaaooweracmn] Figure B-90. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=100 psi/1n (27.1 kPa/mm), and AT=-15°F (-8.3°C). 197 100 80 Principal Stress (psi) 20 .._.__< _____J... --. __ 0% 1 0% 20% 30% 40% 50% 60% 7090 80% 90% 100% Deflection load Transfer Efficiency (%) L—o—W hteriock'Qriy +AftchowelBarRetrofit] Figure B-91. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=250 psi/1n (67.9 kPa/mm), and AT=0°F (0°C). 160 140 120 § Principal Stress (psi) 8 8 20 ! i M . m N‘. - —n 4————o—t fl\\ 1 090 1 0%: 20% 30°/0 40°/0 50°/0 60°/0 70°/0 8090 90°/0 I 00% Deflection load Transfer Efficiency (%) Ee—AgmaemiookOniy-O-Anubowemckwofit] Figure B-92. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=250 psi/1n (67.9 kPa/mm), and AT=+15°F (+8.3°C). 198 Principal Stress (psi) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection load Tramfer Efficiency (%) |+Wwonry+mpoweraumnI Figure B-93. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=250 psi/in (67.9 kPa/mm), and AT=~15°F (-8.3°C). Principal Stress (psi) 120 20 0% 10% 20% 3090 40% 50% 60% 70% 80% 90% 100% Deflection load Transfer Efficiency (%) l-O-Aggregatelnta‘boerly-G-AfierDowechRetrofitl Figure B-94. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=400 psi/1n (106.8 kPa/mm), and AT=0°F (0°C). 199 160 120 100 80 Principal Stress (psi) 60 0% 1 0% 20% 30% 40% 50% 6090 7090 809-0 9090 1 00% Deflection load Transfer Efficiency (%) gwmm-a-momracmn] Figure B-95. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=400 psi/1n (106.8 kPa/mm), and AT=+15°F (+8.3°C). 160 140 120 A é Principal Stress (psi) N O S 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection load Transfer Efficiency (%) [+AggregdelntabckChly-G-AfierDowelBrRetrofl] Figure B-96. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=12” (305 mm), k=400 psi/1n (106.8 kPa/mm), and AT=-1S°F (-8.3°C). 200 8 I 80 L v ' j 4 ‘N 70 'T f: 60 % L H a I V r 3 so I h fl I In I I - I 3. 40 #1 '3 I .5 E 30 20 1 10 0 1 090 1 090 2090 3090 4090 50°/0 6090 7090 8090 9090 10090 Deflection load Transfer Efficiency (%) b-Aggreg‘elnterbckOnIy-l—AfierDomiBuRdrofij Figure B-97. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (3 56 mm), k=100 psi/m (27.1 kPa/mm), and AT=O°F (0°C). 120 100 80 Principal Stress (psi) 8 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection load Transfer Efficiency (%) L-o-Wwouy-e-moomracma] Figure B-98. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (356 mm), k=100 psi/1n (27.1 kPa/mm), and AT=+15°F (+8.3°C). 201 ., 1-0 1 ' l 100 L ' s: 80 \\ n 5 "‘" i: U) 60 3 G1 '3 .5 r- 0- 4o 20 0 1 090 1 090 20%) 3090 4090 50%) 6090 7090 809 0 9090 I 00% Deflection load Tramfer Efficiency (%) [+Aggegaehurockony-e-Aauomraama] Figure B-99. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (356 mm), k=100 psi/in (27.1 kPa/mm), and AT=-15°F (-8.3°C). 9° 7 8 O V : K‘— \ \ 70 a 60 L t* * 8. V a so 5 in a 40 3 .5 5 30 20 I 0 0 090 I 0%! 20% 3 08/0 40% 50% 6090 7090 80°/0 90°/0 I 0096 Deflection load Transfer Efficiency (%) mema—mmracmn] Figure B-lOO. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (3 56 mm), k=250 psi/1n (67.9 kPa/mm), and AT=O°F (0°C). 202 100 80 Principal Stress (psi) S 20 0% 1 0% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection load Transfer Efficiency (%) [+mery-s-moomracma] Figure B-lOl. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (3 56 mm), k=250 psi/in (67.9 kPa/mm), and AT=+15°F (+8.3°C). I 20 a: __.__ _.__ “H 100 ‘ a~ I‘\ N N A ,1: 80 3. V h M 60 '3' Q- '5 .5 5'1 4o 20 0 0°/0 I 0%: 20°/o 3096 40% 50% 60%) 70%: 8090 9096 I 00% Deflection load Transfer Efficiency (%) L—o-Amumbokony-a—mmmcmnl Figure B-102. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (3 56 mm), k=250 psifrn (67.9 kPa/mm), and AT=-15°F (-8.3°C). 203 90 j j l | r A A 80 Wir— I i f I -r\ 1 7O \ 3 : i L 1 fl .1: 60 t t I 5 | I ' ‘ I 3 so % i X 5 {D E. 40 “5 E I B- 30 20 10 0 t 0% 1 0% 20% 30% 40% 50% 60% 70% 80% 90% 1 00% Deflection Load Transfer Efficiency (%) L+WWCth+AflaDm~elBtR¢rofiJ Figure B-103. Principal Tensile Stresses at Crack or Joint Before and Afier DBR for h=14” (356 mm), k=400 psi/1n (106.8 kPa/mm), and AT=O°F (0°C). 120 100 80 Principal Stress (psi) 8 20 NW A ' “r“ N H—u—a A a— § 096 l Oo/o 20%: 309/0 400,0 5 09/6 609 o 7096 809 a 9096 l 00% Deflection Load Transfer Efficiency (%) [+Awognolmerbck0nly-o—Aneroowelackmfir] Figure B-104. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (356 mm), k=400 psi/1n (106.8 kPa/mm), and AT=+15°F (+8.3°C). 204 140 120 ‘_—* ‘ 100 t 4 l i 80 Principal Stress (psi) 20 090 1096 2096 3096 4096 5096 6096 7096 809o 9096 Deflection Load Transfer Efficiency (%) bWeinteriookOnly-I—AflerbowelBuRmOHJ Figure B-lOS. Principal Tensile Stresses at Crack or Joint Before and After DBR for h=14” (3 56 mm), k=400 psi/1n (106.8 kPa/mm), and AT=-15°F (-8.3°C). 205 APPENDD( C: CATALOG OF RCC DESIGN THICKNESSES 206 0.0 0.0 0.0 Wm Wm 0.0 0.m 0.0 0.0 02V 0.0 m0 0.0 0.0 0.0 m0 Om 0.0 0.0 000 0.0 06 0.0 0.0 0.0 0.0 0.0 m0 0.0 000 0.0 m.» 0.0 0.0 m6 0.0 m0 0.2. ms Sn 0.0. 0.: 0.: 0.0 0.0 0.0 0.0 0.0 0.0 00N $9. 232.6 200.... .30" 2.3039 03325 $0 .5820 30.8.— 0.0 0.0 m0 m.m 0.0 0.0 0.m m.m m.m 0mm ms ms 2. m0 m0 0.x. 0.0 0.0 m0 000 0.0 0.0 0.0 ms 2. 0.0 m0 0.0 0.0 03 0.2 0.2 0.: 0.02 0.02 0.0. 0.0 0.0 m.» 00a +0.2 +0.2 +0.2 0.2 0.2 0.2 0.: m0. 0.02 02 . -.0.m 0.0 -. m.m .- 0am.-- -- 0m. , own. , .- 0.m Bowie - - m.m 0.0 0.0 0.m m.m m.m 0.0 0.m 0.m 0mm 00 0.2. 0.0 0.0 0.0 0.0 m.m m.m 0.0 09m 0.0 0.0 0.0 00 0.5 ms 0.0 m0 m0 0m~ 0.2 0.: 0.: 0.0 0.0 0.0 0.0 2. ms 00" +0.2 +0.2 0.2 0.2 0.2 0.2“ 0.0_ 00 0.0 02 .000 $3 $0 $3 $0 $00 $0" :0 09.0 .092 095 .002 03.5 .092 095 .002 0.20 .002 09.9 .002 095 .002 BE saw—ho II 7.66 an!!! p801 1mm 7.96 barman DWI Imam -/.SL mumps 9W1 mom I l l n i as... 8E. ...: 23 2.2 235 .5. N. a E 3.3500: :28: cum ._.0 2%... 207 0.0 0.m 0.0 0.0 0.0 0.0 0.0 0.0 0.m 000 Om 0.0 0.0 0.0 0.0 m.m 0.0 0.0 0.0 000 0.0 0.0 0.» Wm 0.0 0.0 0.0 0.0 .0.0 gm 0.0 0.0 0.0 m0 0... m... 0.0 0.0 0.0 00" 0.2 0.: 0.: 0.0 0.0 0.0 0.0 ms 00 00N +0.2 +0.2 002 m0 md 00— o. o. o m o m o m o m 3 o. -..o... .. K... m. 3 o... 9. ... m. 9. on on 8». o... o. o. 3 Ce 3 n. ... n. 3. no. 3 ca 3 m. 2. 3 3 no 8. +92 0... 3. o... no. no 3 o. o... 8. +0.2 +0... +3. 32 .6... o... 8. 7.66 mumps 7.56 ‘JEIW . 8 P301 [“3131 P301 ["9131 as... 3.". ...... .33 2.2. 9.95 .5. a. a .... 83252.3 5.0... 8.. .90 2...; 11¢... o. o. q. o.“ o m om On m.m Wm 0.m 0.0 0.0 0.0 0.m 0.0 000 m. n. m0 m0 0.0 0.0 Om 0.0 0.0 0.0 0.0 00" m M 0.2 3 o... o... o. o. 3 0.0 o... 8. h. m. I- +0.2 m2.— m.: 0.: m0 m0 0.0 m... 0.0 02 No. m. C +0.2 +0.2 +0.2 +0.2 +0.2 0.2 0.2 0.: 0.0 00— 62.00 2.0" $0 ..\900 30" ..\e0 300 $3 $0 .— 09.0 .002 09.0 .092 09.0 .092 09.0 .092 09.0 .092 09.0 .002 09.0 .092 09.0 .092 09.0 .002 iAMH-w .0 $00 3309.0 95.3.— e\omn 0:30.20 25:...— .x.0 2.9.09.0 02:3.— 208 0.m 0.m 0.0 0.m 0.0 0.0 0.m 0.m 0.m 000 0.m m.m 0.0 0.m 0.m m.m 0.m 0.0 0.m 0mm W m. 0.0 m0 m0 m.m 0.0 0.0 0.m Wm m.m Sn W m was m... 0.0 m0 m0 0... m.m 0... m0 0m~ ,M m. m... 00. 0.0. 00 0.0 0.0 0.5 0.0 m... 00" m m. +0.2 +0.2 0.2 0.3 m... 0.0. m-m 0.2 0.0 02 o 0.m 0.m 0.m 0.m 0.m 0.m 0.m 0.m 0.m 0mm m.m m.m m.m 0.m 0.m 0.m 0.m 0.m 0.m 000 m m. o... 3 no oe 0.0 o... o. o... m. emu W m 0.0. 0.0 0.0 ms 0.5 0.0 0.0 0.0 0.0 00" b m1 +0.2 0.2 m... 0.2 0.0. 0.0 0.0 0.0 m... 02 m m. +0.2 +0.2 +0.2 +0.2 +0.2 0.2 0.2 0.: 0.0_ 00. o 0....“- . . -00 0.:m . 0w. 04». 0.m .00 . olwi ......... 0w- . 0mm m.m Wm m.m 0.m 0.m 0.m 0.0 0.m 0.0 000 m n. 00 00 0.0 m.m Om m.m 0.m 0.m 0.m 0m~ W m 00 0.0 m... 0... m0 m0 0.0 0.0 0.0 00" W W. +0.2 0.... 0.: 0.: 0.0 0.0 ms 0... 0... 00. o“ .n... +0.2 +0.2 +0.2 +0.2 +0.2 0.2 0.2 m0. 0.0 00. $00 $3 $0 $00 {emu $0 $00 {emu ..\e0 09.0 .002 09.0 .002 09.0 .092 09.0 .002 09.0 .092 09.0 .002 09.0 .092 09.0 .002 09.0 .092 sac—”fie 209 “image”. 20.8.. e\o0 2.30.9.0 93:3.— .83 28.220 30.8.. as... 8v". ...... .83 2:. 805m .5. 2 a .3. 8.850.... 5.8.. 8.. ..-“. 0......n L m0 0... 0.0 0.0 m0 0... 0.0 0.0 0.0 02. 0.5 m... 0... 00 0.0 .6 0.0 .0 05 000 W n. .6 0.0 0.0 0.0 ms 0.0 0... N. 0... 000 W m 0.0 .0 0.0. 0.0 m.» 0.0 m... 0.0 m0 00.m h M. 0.: m... 0.2 0.0 0.0. 0.0. 0.0 0.0 0.0 0mm W m. 0.... m... ...; m... 0.2 0.2 0.0. m0. 0.: 00m 0.. 0.0 0.0 0.m ... .0 0.m 0.. On 02. 0.0 m0 0.0 W. 0.0 0.0 0.0 ... m0 000 W m. .0 0.5 .00 0.0 m0 0... ... 0.0 0.0 000 W m m.» 0.0 0.0 0... 0.0 0.5 0.0 m0 0... 000 h m. .0 0.0. 0.0. 0.0 m.» 0.0 06 mg 0.0 03 m m. 0.2 m2 m2 0.0. 0.0.. 0... .0 0.0 0.0 00" o ... o. n o... 0.. o. ... o. 0.. WM 8.. 0.0 m0 0... Om 0.0 m0 0.. ... 0.0 000 m m. o. ... o. 0.. ... o. ... o. .... 8.. W m 0.0 0.0 0.0 m... 0.0 0.0 0.5 0... m... 00" .w .01 0.2 m... m... 0.0 m0 0.: 0.0 0.0 a.» 00m 75. m. +0.2 +0.2 +0.2 0.3 m2 m2 m0. m0. .0. 02 T [I an $00 {emu $0 $00 $3 ..\..0 .300 1 09.0 .002 09.0 .092 09.0 .002 09.0 .03.. 09.0 .092 09.0 .092 09.0 .002 09.0 .002 09.0 .wuz .330 $00 2.2090 95.8.. $0" 2509.0 «>58.— _. .30 2509.0 250.....— ...<... 8.". ...... .83 83.. 8.9... .5. .. .... 8.. 8.8.52... .98.. on... ....u as; 210 ......a ...-u. ...... .83 8...... m5.... .5. .. .... .... 8.850.... 5.8.. 9.... ..-u 2...... .... .... ... ... o... .... ... ... o... a... o. .... o. o... o... ... o... o... .... 8.. m n. o. ... o... .... o... o. ... ... o. .... w m o... .... o... o. ... o... o. o. ... 8n n m o... ...: o... ... o... ... o... o... ... 0.. m m. .... ...: .... o... o... o... ... ... o... 8. ... ... .... o. ... ... .... o. o. 8.. o... ... .... ... o... o. ... ... .... .... m. n o. ... ... ... .... o. o... o... .... 8. w m o. o... o. ... .... o... o... o... o. .... m .or o... o... o... .... .... ... o... o... .... 8. on. m. .5... +0... +o... o... o... o... ..o. ..o. o... .... .-.... .o... - ... gowg o..- o. o. o. o. 8.. ... .... o... o. ... ... o. o. o. .... m m. .... o. o. ... o... .... ... ... o. 8. m m o... o... o... .... .... o. o... .... ... ...a m m. o... o... ...: .... .... .... o. ... ... 8. ./c.. m. +9.. .5... o... o... o... o... o... ... o... .... .8... 8.. .x... ...... x... ...... .x... 09.0 .002 09.0 .002 09.0 .007. 09.0 .002 09.0 .92 09.0 .002 09.0 .002 09.0 .097. 09.0 .002 emu—awn ...... 28.5.5 20.8.. ...... 28.5.... 20.8.. .x... 58.5.... .30...... 211 Om 0.0 0.0 0.0 0.0 0.0 0.0 0.m m.m 000 0.0 0.0 0.0 0.0 0.0 0.0 On Om 0.0 000 m0 0... ms 00 0.0 0... 0.0 0.0 00 000 m... 0.0 0.0 0... 0.0 0... m0 m0 0... Sn 0.0 0.02 0.0. 0.0 0.0 0.0 m... m... 0.0 0m~ 0.0 0.0 0.0 m0 Wm 0.0 0.m 0.0 On 000 0... 0... 06 0.0 0.0 0.0 0.0 0.0 0.0 000 m0 0.0 0.0 0... m... 0.0 0.0 m0 0.0 0m~ 0.2 0.: 0.: 0.0 0.0 0.0 0.0 m... 0.0 00" .8... 28.0.20 2.2.8.. .- .....a 22.8. ...... .28. 8...... 8...... .5. .. .... .... 8.8.5.2.... 8.8.. 00.. ....0 2...... .2... 58.0.20 2.28.. - - -- 8.1.5.080 85.8...- “A66 KIEIN'EPH P301 ["9131 “AS6519” . II mm [mm ... ... .... .... .... ... .... .... .... .... m... n. ... ... ... ... .... .... .... ... ... 8. W m o... o... o... .... o. o... .... .... o... .... .w. m. o... ...: ...... .... .... o... o. .... .... 8. w... m. .6... +0... ...: o... .... ..o. ...J. o... ... .... .8... .2... .2... .8... 8m. .2... .8... 2... .2... 09.0 .002 09.0 .092 09.0 .002 09.0 .097. 09.0 .002 09.0 .007. 09.0 .007. 09.0 .002 09.0 .37. 30h”... 212 3 3 3 3 3 3 3 3 3 ..mn 3 3 o... 3 3 3 3 3 3 8». m n. m... m... m... 3 3 2 3 no 3 cm” W m 3. m... o... 3 3 3 ms 3 o... :8 a m. +3. +3. 3. 3. 3. 3. 2.. 3. S: ..m. m m ----+o.m_ +o.m_ ....o..-m_ .52 I +0.2--- -+o..m_,- l....o..m_ll 0.2, 0.: 8— 3. -- w, 3 .- 3 _ 3 3.- - 3 3 3 3 cm». 3 3 3 3 3 3 3 3 3 8.. m m. 2 3 3 3 o o m o o m m m m m on" m... m 3. o... 3 3 o... 2. no 3 no 8” h m. +3. 3. 3. 3. o... o... m... m... 3 on. m m. +3. +3. +3. +3. +3. +3. 3. 3. o... 8. . 3 3- 3 3 3 3 .3 3 on m.m m.m ed on 06 oh 06 ca eon m n. 0.5 We m6 m.m m.m o6 oh 0% o.m emu W m 0.2 od Ow 2. cs OK. 0.0 0.... ob an W M. +0.2 02 0.2 m... 9.: od ow ms ms cm. ..x/g. m. iho.m...-.- -ifim..-.éi+fi.i ifimezihomm ..----mw—-.-..-.im.....w..l.Elma! %-_.mo,.l!-@wl+ $3 ..\..m~ .xé {can ..\..m~ .xé $3. Rama ..\..c 3:0 .92 3:0 .woz EEG .woz v.20 .wuz EEG .32 3:0 .woz EEG .92 320 .3 Z BBQ .92 .39.”. |i | ...... b .x... 29.26 .33.3.. 1&3 2.0.2.5 2...»... :3 E32? 3...»... ......2. 8.". ...... .33 2:. 52...: ...: 3 a .... 8.85.92... :38: 8: ..-u 0...: 213 ......2. an". ...... .83 2:. .52.: ...: 3 a .... 8.85.9.5. $.qu Do: ....u 2...... 9m 9m 9m 9m 9m 9m 9m 9m 9m :3. 9m m.m m.m 9m 9m 9m 9m 9m 9m 9mm 9o 90 9c m.m m.m m4». 9m 9m .m.m 8n n... ms ms m6 m6 m6 m.m 9e 9e cmu 9: 92 md m.» 9m 9m 9.. 9n 9.. 8" +92 +92 92 02 9: 9m 9m 9m 9m 9m m.m m.m 9m can We no 9m cma 9o. Ow 9w 8" +92 3.. 95 :2 +92 +92 ma 8— J3 3 3 3 ..mm 9m 9m 9m 9m 9m 9m 9m 9m 8m 90 9e 90 m6 m.m m.m 9m 9m emu 9m 9w ms 9» 90 9o m.m m.m 8" +92 m2 9o. 9: 9a 9m 95 m... :2 +92 +92 +92% +92 +92 8— .x.mn ..\..c .xém {emu 3: .x... 3.5 .woz EEG .woz 3:0 .32 EEG .woz EEG .92 3:0 .92 3.5 .97. 32G .32 an... "flu it- :i iiizinillliiiilliill -ii. .....Efib .xém 23.—=20 32.25 $233.0 3...»...— .\e.. 2.0.19.0 95.8.— MFR”! %66 £1! %56 hyuqvmu %SL ‘1!l!q3!l°‘tl P301 [”9131 p801 [mun pvo'l Imwl 214 .3... 23.2.0 2......— _ $mn 23.—:26 02:3.— _ a x... 29.3.6 2...... ., J! as... 2...". ...... ....3 9...... .52.? ...v. 3 .. .8 8.85.9.5. $.95 US. ....u 2...... %66 hmqm %S6 £1!l!q3!l°ll v.91. Kama . u DWI Ivan mm [mm P301 ["3131 9m 9m 9m 9m 3m m.m 9m 9m 9m 3 9o m.m m.m m.m 3" ms m6 Do We 8" ma 9a 9m 9m 3m 9m 9m 9m 9m :3 m.m 9m 9m 9m 3m We 90 9o m.m 8" 9» ms 95 9.. 3— nd— mé. 2: 9m 2: -.- 3 3 3 3 .3 9m 9m 9m 9m 3 9m 9m 9m 9m 3N 9o m.m m.m m.m 8" +92 92 9o. ma: Ow ms 9.. Do We 3— +92 +92 +92 +92 +92 9: m2 ha 9» 2: H“ --.. - . $3 $m~ $9 $3 $3" $¢ $3 $3" $9 69.0 .92 3.5 .32 EEO .woz EEO .woz 3.5 .32 2:0 .37. EEG .woz EEG .wuz 22¢ .92 215 ...... 28.2.5 2...... $mN 25.—3.0 02:3.— _ $¢ gum—.20 03:2:— _ m.m «6 9m 9m m.m m.m 9m 9m m.m 3.. 9o 90 We m.m 90 9o 9m m.m 9o :3. H n. 3 3 n. 3 no 3 3 3 3 cm... W M 9» m.» m... 95 ms 2. 90 We 95 3 .M m. 90— 9o. 98. 0w 9a 9a ms m... 9m 3n W m. Tlm .2 o .2 o .2 m w. o m m m m m o m m.m 90 9o 9m 9.. 9h 95 9o m.» Ow Wm 9.. 92 m... 9: 9o +92 3 3 4.13 ,. - 3 3 -1- 8.. 9m m.m 9m 9m 9m m.m 9m 3m W m. 90 Do We m.m 9e 9e 9m :3 mm M 9m 9m 9.. 90 9h 9.. m.m 3N W. m. n... Do. We. m.» m... m.» 9.. 8n .7» m. +92 +92 9.: 92 92 9: 9o 32 $3 $3 $e $3 $3" $= $3 3.0 .97. 3.0 .32 3.0 .37. 3.0 .92 3.0 .32 3.0 .92 3.0 .32 3.0 .32 3.0 .woz ifiahb ......2. 8.". ...... ....3 8...... 58...... ...M S. .. .... 8.85.2.5 ....8: no: ....-u 2...... 216 9m 9m 9m 9m 9m 9m o m 9m 0 m 8v 9e 9e m6 Wm 9m 90 m m m.m m m 3n 9.. 9.. ms 90 m6 m6 0 0 90 o o 3 9a 9a 9a m... m6 m... o h 9.. o .. 3N 92 92 m... 90. md md o w 9w 0 w 8N flan" ‘ $.3 2.9.1.20 02:8.— .=_._a emu". ...: .33 2.2 52.5... .5. 8 a 3. 8.3.3.2.. :38: 8.. ...-u as: m.m m.m mg». 9m 9m 9m 0 m 9m o m 3m 9e 9o 90 Wm m.m 90 o m 9n n m 3 9» m... m6 m6 m6 90 o e 90 o o 3N .0: We. no m.» 9.. 9m o .. 9.. o 5 :3 9e 9o 90 9m m.m 9.». o m 9m 0 m 3 9m 9w 9.. 90 9o 90 m m m.m m m 3" 9o. md Ow 9» m... m... m 0 m6 m 0 :3 +92 +92 m2 92 9.: 9m m m 9w o w 3— +92 +92 +92 +92 +92 92 +92 92 m6. 8. $3 $m~ $5 $3 $3" $m~ $9 3.0 .92 3.0 .92 ......0 .32 .....0 .woz 3.0 .woz .....0 .37. ......0 .92 .3th .x... 23:36 32.8.. 7.66 £1!I!¢I'!Ptl %S6 51!]!q3flatl %SL BREW}! pvo'l ImWI P301 ["9131 P301 [”3131 217 on m m 3 on ow o6 o.“ ow o m 8a m m 3 o.» on on 3 ow o.“ 3 cm... W m m o 3. 3 o.» no no 3. on o o 8». m M o w 3 o.» 3 S as no 3 m o cm" 0. m. 2: o.: 2: as ea 2 ms 2 m 5 8a m m o o 0.0 0.0 m.m m.m Wm od 06 o m can N m. m N. o6 as. 00 m6 m6 Wm Wm o o awn W M Q: m 0 od 9w m6 m6 0.5 m6 m o 8N .M m. +0.2 +o.m_ m.N_ 0.2 ma: md 05 Wm o m cm— W M. fun“. kn] .xam 25:35 25.8.. {I :3 £2.56 258.. o 0 ed o6 ow ow ow ch 06 o m :3 W m. o N. o.» 3 od 0.0 3 on 3. m m emu m m Do. o6 ow ms cs 0.5 m6 m6 0 0 can W m. 3.2 +92 0: QB o.o_ od Wm ms m h :3 ow W 3.2 +o.m_ +o.m_ +o.m|_|L. +0.2 02 +o.2 0.: m o SMulllt: .xé .xém {own o\o¢ $3 $3 .\e¢ :95 .32 195 .32 19.0 .32 19.6 .woz 19.0 .32 EEG .woz 69.0 .92 19.6 .92 :95 .92 than ...SE 8.1. E: .33 2.2 52...; .5. a... a ..e .3852: :38.— uum .fié 0.3; 218 ..\..¢m 23:39.0 03:3...— ..\.3~ 33.395 2.3.8.. .35.... 33.": 3:3 333 3....V .5353 3c. 33 3 .3. 8.852.: 33.3.. 8.. 3.6 3.33.3 .33 35.3.3.6 32mm... - .. 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 333 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 33.. m m. 3.3 3 3 3.3 3.3 3.3 3.3 3.3 3.3 33.. m M 3 3 3 3.3 3 3 3.3 3.3 3 333 h m. 3 3.3 3 3 3 3 3 3 3 333 m m. 3.3. 3... 3... 3.3 3 3.3. 3 3 3.3 33" 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3 33.. 3 3.3 3.3 3 3.3 3.3 3 3 3 33.. m m. 3.3 3.3 3 3.3 3.3 3.3 3 3.3 3 333 w m 3 3 3 3.3 3 3 3 3 3 3 3 3 333 b m. 3.3 3 3 3 3 3 3 3 3 3 3 3 333 m m. 3. 3. 3. 3. 3.3. 3.3. .Illliln .flnflflfluflflfl 3.3 3 3.3 3 3 3 3.3 3 3 3 3.3 3 3 3 3 3 3 3 333 J n. 3.3 3 3 3 3 3 3 3 3 3.3 3 3 333 m m 3 3 3 3 3 3 3 3.3 3 3 3 3 333 m m. 3. 3... 3... 3.3 3.3 3 3 3 3 3 3 3 333 m M. +3.3. +3.3. +3.3. 3.3. 3.3. 3. 3 3. 3.3. 3 3. 33. $33 $3 .333 $9.0 .97. .395 .32 3.9.0 .37. 69.9 .32 .395 .32 “fl $1-1! 3.32.2.6 219 -/.66 human pm Imam “/396 byuqmou DWI mam 0.3 3.3 3.3 0.3 93 3.3 Q3 0.3 Q3 emv 3.3 Cd 0.1 3.3 3.3 0.1 Q3 3.3 3.3 81 3.1 3.0 0.3 0.0 0.0 3.1 3.3 0.1 .06 1mm 1.3 o.w ow 3.1 0.5 0.x. 0.1 3.1 3.3. can od— 33 33 ed 3.x 3.x o... 3.5 3.3. cm" o3 3.3 0.1 06 0.3 3.3 3.3 0.3 0.3 0.3 1mm 3.1 3.0 0.3 Q1 9.1 ob 3.3 3.3 3.3 gm 3.x o.w o.w OH cs 0.3 06 0.1 3.1 an" 3&— 0... 3.0. 0.3 od 3.« 3.3. 3.3. 3.5 can +03— +o.3_ 0.3. 3.: 3.3 0.3 cm— “/oSL hummer! DWI ["31"] 3.3 3.3 3.3. 3.3 3 3 33 3 3 3.3 --3 -- 533.. 3.3 3.3 3.3 0.3 0.3 o3 o3 o3 o3 awn o3 0.3 3.0 3.3 3.3 o3 o3 o3 3.3 83.. o.» 3.3 3.3 3.3. 3.3 3.3. o3 0.3 od cm" 0... ad. 3.0 3.3 ow o.» 3.3. 03. 0.3 can +0.3. +0.3. o... 3..3. o... H 3.2 3.3 3.3 3.3 cm— ..\3=m .53 .x... $13 $3 33... $3. .33 x... 19.6 .92 19.6 .37. 19.6 .37. 19.6 .37. 19.6 .32 19.6 .92 19.6 .97. 19.6 .92 19.6 .97. fish—ho .333 35.335 2.3.8.. e\omN 31319.6 03:8.— ......2. 333". 3:3 3.3.. 3.3 3.33:3... 3.3. 33 3 ..3. 8:35.32: 33.3.. 8.. .....0 3.33.3 .33 35.3331 33...... l I 220 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 33.. 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 33.. m n. 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 333 w m 3 3 3 3.3 3.3 3 3.3 3.3 3.3 333 m. m 3.3 3.3 3.3 3 3 3 3 3 3 333 m m. 3. 3.3. 3... 3.3. 3 3 3 m3 3 .1333 3.3 J- 33 3.3 3.3 3.3 3.3 3.3 -3.-3 3.3 l 33.. 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 333 J n 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 333 m m 3 3 3 3 3 3 3.3 3.3 3.3 333 h m. 3. 3.3. 3 3 3 3 3 3 3 333 m M. +3.3. +3.3. 3.... 3.... 3... 3.3. 3 3 33. . Tlnfll-rulfln ---l | .i -- -- Iii--1 ...... ---i-i --.- “flnflnwl-flLfllll 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 33.. 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 333 m n. 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 333 W m 3 3 3 3.3 3.3 3.3 3.3 3.3 3.3 333 m m. 3... 3 3.3 3 3 3 3 3 3.3 333 m m. +3.3. +3.3. 3. 3.. . 3.3. 3 3.3 3 3 33. .333 .333 $3 $33 .333 $3 .333 $33 :3 19.6 .32 19.6 .32 19.6 .32 19.6 .97. 19.6 .32 19.6 .97. 19.6 .37. 19.6 .92 19.6 .37. ...Amuwawb $3335.33... 2.3.8.. II.“ $33.5...fi33. --I--il|..\u335.3333333326- 33333 333"... 3:3 333 33. .535; 3.3. 33 3 .3. 8385...... 33.3.. 3.. 3.6 3.33.. 221 APPENDIX D: RAW LOAD TRANSFER DATA 222 Table D-l. LTE Tests on Transverse Cracks (After Pittman [6]). . 31.9 . Std Dev 8.48 9 16.3 0.026 Central COV 8% 11% 51% 126% Freight Sep-91 # of Tests 75 76 76 5 Maximum 126 93 84 0.068 Minimum 101 35 7.3 0.003 Range 25 58 76.7 0.065 Mean 71 62 30.5 0.051 Std Dev 0.31 15 11.1 0.064 Austin, COV 0% 24% 36% 125% TX Sep-91 # of Tests 25 25 25 18 Maximum 71 85 53 0.300 Minimum 70 33 13.6 0.013 Tuscany Range 1 52 39.4 0.287 Way Mean 60 55 30.5 0.059 Std Dev 0 13 11.1 0.069 COV 0% 24% 36% 1 17% Jan-92 # of Tests 24 24 25 18 Maximum 60 83 53 0.320 Minimum 60 32 13.6 0.013 Range 0 51 39.4 0.307 Mean 43.5 24.3 79.6 0.244 Std Dev 1.97 12.1 28.2 0.132 COV 5% 50% 35% 54% Jan-91 # of Tests 78 78 78 78 Maximum 46 83.5 155.6 0.550 w 2.: 3:2; 32:3 . ge . . . Gaggeu, 52:11:13 Mean 112 77 81.0 0.134 Std Dev - 9 28.3 0.096 COV - 11% 35% 72% Aug-91 # of Tests 86 86 86 86 Maximum 127 89 135.6 0.425 Minimum 96 22 24.7 0.008 Range 31 67 110.9 0.418 223 Table D-l. LTE Tests on Transverse Cracks (After Pittman [6]) (cont'd). Std Dev - 17 26.0 COV - 22% 59% 69% 21:9? Aug-91 # of Tests 40 42 41 24 Maximum 116 93 106.3 0.175 Minimum 98 11 8.4 0.001 Range 18 82 97.9 0.174 Mean 71 40.4 29.6 0.067 Std Dev -- 19.8 21.7 0.028 COV -- 49% 73% 41% 12:66ng Aug-91 # of Tests 24 25 25 18 Maximum 74 88.1 86.55 0.110 Minimum 65 9.3 8.4 0.018 Range 9 78.8 78.15 0.093 Mean 93 78 31.8 0.043 Std Dev - 14 12.9 0.036 COV -- 18% 41% 82% Ft' £3” PN69B Aug-91 # of Tests 95 95 95 52 Maximum 112 91 76.3 0.150 Minimum 79 25 13.2 0.008 Range 33 66 63.1 0.1425 Mean 65.6 88 -- -- Std Dev 4.13 6 -- -- COV 6% 7% -- - Apr-90 # of Tests 16 16 -- -- Maximum 72 96.2 -- -- Minimum 61 70.2 -- -- Range 11 26 -- - PN187 Mean 87 83 34.34 0.024 Std Dev -- 9 14.58 0.017 COV -- 11% 42% 68% Aug-91 #ofTests 15 15 15 12 Maximum 91 90 58.85 0.055 Minimum 81 62 16.00 0.008 Range 10 28 42.85 0.0475 224 Table D-l. LTE Tests on Transverse Cracks (After Pittman [6]) (cont'd). Std Dev 0 0 . . Ft. Drum, COV 0% 0% 0% 0% NY PN203 Aug-91 # of Tests 1 1 1 1 cont'd Maximum 91 80 29.8 0.010 Minimum 91 80 29.8 0.010 Range 0 0 0 0 Mean 60 64 51.5 - Std Dev 0 14 17.8 - COV 0% 22% 35% -- Mar-85 # of Tests 129 129 129 -- Maximum 60 88 89.6 -- Minimum 60 18 10.5 -- Range 0 70 79.1 -- Mean 61.8 30 -- Std Dev 2.02 18 -- Bldg. COV 3% 60% #DIV/O! -- 2 6027 Feb-90 # of Tests 27 27 129 -- Maximum 64 73 67 -- Minimum 60 1 13.2 - Ft. Hood, Range 4 72 53 .8 -- TX Mean 95.21 86 28.20 0.107 Std Dev 0.78 7 12.90 0.105 COV -- 8% 46% 99% Aug-91 # of Tests 42 45 45 45 Maximum 98 97 57.45 0.375 Minimum 95 53 5.25 0.005 Range 3 44 52.20 0.37 Mean 75 88 27.60 - Std Dev 0 9 9.70 -- Bldg. COV 0% 10% 35% -- 3 8033 Feb-90 # of Tests 30 30 30 -- Maximum 75 99 51.00 -- Minimum 75 48 14.80 -- Range 0 51 36.20 -- 225 Table D—l. LTE Tests on Transverse Cracks (After Pittman [6]) (cont'd). é“ Location Area Date(s) Statistic Tested =fi Std Dev 3.028 4 Bldg. COV 3% 5% 38033 Sep-91 # of Tests 50 51 cont'd Maximum 116 93 Minimum 1 10 75 Range 6 18 Mean 63 32 Std Dev 0 18 COV 0% 56% 27% -- Feb-90 # of Tests 42 42 42 -- Maximum 63 84 41.30 -- Minimum 63 14 15.40 -- Bldg. Range 0 70 25.90 -- 3850 Mean 85.95 54 23.90 0.064 Std Dev 5.7 21 15.10 0.042 Ft. Hood, COV 7% 39% 63% 66% TX Sep-91 # of Tests 64 66 66 66 cont'd Maximum 96 92 103.40 0.175 Minimum 84 12.8 7.45 0.003 Range 12 79.2 95.95 0.1725 Mean 67 63 -- - Std Dev 0 21 -- -- COV 0% 33% -- ~- Feb-90 # of Tests 1 1 11 -- -- Maximum 67 92 -- -- Minimum 67 36 -- -- Wash Range 0 56 -- -- Rack Mean 88.5 40 40.50 0.094 Std Dev 1.12 9 20.20 0.080 COV 1% 23% 50% 84% Sep-91 #ofTests 13 13 13 13 Maximum 89 65 79.05 0.275 Minimum 86 28 15 .40 0.020 Range 3 37 63.65 0.255 226 Table D-l. LTE Tests on Transverse Cracks (After Pittman [6]) (cont'd). T Crack Crack Location Area Date(s) emp Spacing Width Tested (%) ft. in. Std Dev 607 945 -- -- COV 0% 4% -- -- Feb-90 # of Tests 11 11 -- -- Maximum 67 100 - -- Minimum 67 85 -- -- Ft' 30¢ Tank Range 0 15 -- .- cont'd Trail Mean 101 93 32.20 0.073 Std Dev 0 4 11.60 0.053 COV 0% 4% 36% 73% Sep-91 # of Tests 20 20 20 19 Maximum 101 99 63.50 0.175 Minimum 101 85 9.70 0.018 Range 0 14 53.80 0.1575 Mean 55 88 15.2 0.005 Spring Std Dev 0 0 0.0 0.000 Hill, TN Zenith COV 0% 0% 0% 0% (Saturn Road Jan-91 # of Tests 1 1 1 1 Plant) Maxnmum 55 88 15.15 0.005 Minimum 55 88 15.15 0.005 Range 0 0 0 0.000 227 Table D-2. LTE Tests on Transverse Joints (after Wu and Todres [39]). l4. 8 6. 5 9271 78 78 9405 80 80 12284 79 12490 80 7066 61 6550 54 14. 8 6. 5 9770 60 60 9493 58 57 12554 60 12490 60 6891 70 6320 67 2 14. 8 6. 5 9461 66 68 9247 65 65 12014 67 12379 64 603 5 5 1 643 1 73 14.8 6.5 9175 55 55 9461 72 72 12054 59 12570 71 6217 77 6614 88 14.8 6.5 9358 77 77 9540 87 87 123 79 77 12609 87 6788 81 6416 76 3 14.8 6. 5 9223 79 79 93 82 76 76 12102 79 12546 77 6495 79 6416 79 14.8 6.5 9413 80 80 9382 80 79 1241 1 80 12546 80 6273 79 6558 79 14. 8 6. 5 9326 79 79 9540 80 80 12371 80 12665 80 6201 82 63 13 72 4 14.8 6.5 9167 80 81 9286 73 73 12213 82 12387 74 6424 83 63 13 79 14.8 6.5 9382 84 83 9318 79 79 12347 83 12482 79 228 Table D-2. LTE Tests on Transverse Joints (after Wu and Todres [39]) (cont'd). l 1992 1993 '1 Joint Average A A Slab Spacing FCC Load LTE; verage Load LTE, verage No. ( ft.) Thickness (lb) (0/ ) LTE; (lb) (0/ ) LTES (in.) ' (%) ’ (%) 12276 76 12157 76 6201 73 6257 69 5 21.3 6.7 9167 73 73 9318 70 70 12229 73 1241 1 71 63 84 75 63 52 77 21.3 6.7 9342 74 75 93 82 76 76 12355 75 12514 76 6368 61 6297 41 21.3 8.5 9223 65 65 93 58 46 46 12173 67 12316 50 61 14 50 6217 41 6 21.3 8.5 9025 52 52 9207 47 46 12086 54 12419 51 6305 42 6471 78 21.3 8.5 9302 45 45 9445 78 78 12268 48 12657 79 7082 42 6313 66 21 .3 8.5 9873 49 48 93 89 67 67 12149 53 12213 68 6241 70 6424 64 7 21.3 8.5 9136 68 69 9405 67 66 12189 69 12538 68 6416 54 6225 63 21.3 8. 5 9397 56 56 9239 62 62 12308 58 12363 61 6360 55 6297 72 32.8 8.3 9239 56 56 9294 70 71 1218] 56 12244 69 6209 65 6273 64 8 32.8 8.3 9120 62 63 9255 64 63 12141 61 12292 63 6907 68 5932 38 32.8 8.3 9516 66 66 9001 42 42 12268 65 12205 45 229 Table D—2. LTE Tests on Transverse Joints (after Wu and Todres [39]) (cont'd). mm 1992 Slab No. Joint Spacing (ft) Average RCC Thickness (ilk) 1993 Load LTE, Aggy Load LTE5 Aggy (lb) (%) (%)“ (lb) (%) (%)“ 32.8 8.3 9429 75 74 9199 42 42 12387 73 12292 45 6487 54 6225 75 32.8 8.1 9175 57 56 9318 76 75 1 1967 59 12141 75 6209 67 6217 60 10 32.8 8.1 9072 68 67 9271 60 60 12062 67 12332 61 6463 72 6344 67 32.8 8.1 9421 73 72 9358 68 68 12340 73 12482 68 6558 68 6162 56 49.2 8.0 9080 70 70 9199 59 59 l 1800 71 12070 61 6067 59 61 14 23 1 1 49.2 8.0 9001 60 60 8977 28 28 1 1935 60 12062 32 6463 62 6336 61 49.2 8.0 9421 62 62 9255 64 63 12347 62 12419 65 6186 60 6368 62 49.2 7.5 9207 65 64 93 74 63 63 12022 67 12316 64 6257 66 61 14 51 12 49.2 7.5 9080 67 67 9152 52 52 12070 68 12276 53 6408 68 6328 64 49.2 7.5 9326 71 70 9239 67 66 12292 71 12379 67 230 Table D-Z. LTE Tests on Transverse Joints (after Wu and Todres [39]) (cont'd). 231 1992 1993 Joint Average Shh Spacing FCC Load LTEis Avmg’ Load LTEls Avmge N“ (ft) 11.333, (lb) (%) LTE“ (lb) (%) LTE“ (in.) (%) (%) 6075 84 6408 77 49.2 7.5 9183 84 85 9397 80 79 12260 85 12435 81 6090 82 63 28 79 13 49.2 7.5 9056 83 83 9231 81 81 12086 84 123 63 82 63 76 81 6336 66 49.2 7. 5 93 02 84 82 93 02 67 67 12260 82 12490 68 AVERAGE 69. 1 AVERAGE 67.2 COV 15.2% COV 19.4% APPENDIX E: GUIDE FOR LIMITING SUBGRADE STRESS 232 l4 10 Subgrade Stress (ps1) on AAAA 0% 1 0% 20% 30% 40% 50% 60% 70°/' 80% 90% l 00% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) 1-4—Undowebd k=100 paifn +Undowebd k=250 psifai +Undowebdk=4mpsifn *O-“Dowebdhlm psi/ii *Dowebdmpsifm *Dawebdkflmpsi/h Figure E—l. Subgrade Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle on a 6” (152 mm) RCC Pavement Before and After Retrofit Dowels. 16 14 §‘_\\\ 10 N Subgrade Stress (ps1) N 111 " 11 1L 1 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) —o—Undowebdk=100pil‘m —l—Undowebdk=250 psi/ii +Undowebdk=4m pil'n *Dowebdk-lmpsil'n *Dowebdk'IZSOpsi/‘m "lb-“WWW: Figure E-2. Subgrade Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle on an 8” (203 mm) RCC Pavement Before and Afier Retrofit Dowels. 233 Subgrade Stress (ps1) 0% 1 0% 20% 30% 40% 50% 60% 70% 80% 90% 1 00% Deflection Loud Transfer Efficiency Before Retrofit Dowels (%) +Undowebd k-lfl) psi/ii +0Mk=250 psi/ii +Undowebdk-400psi/ii ~+~Doweuk=roo psi/ii «~ka para ~+~oowebab4oopiru Figure E-3. Subgrade Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle on a 10” (254 mm) RCC Pavement Before and After Retrofit Dowels. 12 \ 10 W Subgrade Stress (psi) N b 0 i) 4 1) 0 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) -—.—W k=100 pm. +um k=250 par. +Undowebdk=4mpsilii “-fi-“Dowebdbloo psi/ii m'O'-'-Dovt1vobl:lk=250psi/ii *t—"Dowebdk—W psi/ii Figure E-4. Subgrade Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle on a 12” (305 mm) RCC Pavement Before and After Retrofit Dowels. 234 14 10 Subgade Stress (psi) .. M 4 M 4 ; ¢ 3 3 t C 3;. 2 1' i 0 1 0°ro 1 090 2096 30% 40%) 50% 60%) 7096 8090 9096 1 0096 Deflection [and Transfer Efficiency Before Retrofit Dowels (%) +Undowebdk=lw psi/m +Unbwebdk=250 psi/is —t—Undowobdk-4mpsilm *DowebdblOOpsi/‘m *WFZSOpai/‘n WDowebdeioopai/‘m Figure E-S. Subgrade Stresses for a 12 kip (53.4 kN) Dual-Tired Single Axle on a 14” (3 56 mm) RCC Pavement Before and After Retrofit Dowels. Subgrade Stress (psi) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% I 00% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) -—3—UM k=100 an +Undo~vebdk=250peilii +Undowabdk=4mpilil *Dovnbdbloo psi/ii *Dowebdmpsifu *Dmbdmpsifn Figure E-6. Subgrade Stresses for a 18 kip (80.1 kN) Dual-Tired Single Axle on a 6” (152 mm) RCC Pavement Before and After Retrofit Dowels. 235 Subgrade Stress (psi) 0% 1 0% 20% 30% 40% 50% 60% 70% 80% 909 6 100% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) —O—Undowebd k=100 psi/ii +Undowebdk=250 psi/ii +um¢ar=4wpw “*DoweblelOOpsi/h *DowebdFZSO psi/ii WMHOOpsi/‘m Figure E-7. Subgrade Stresses for a 18 kip (80.1 kN) Dual-Tired Single Axle on an 8” (203 mm) RCC Pavement Before and After Retrofit Dowels. Subgrade Stress (psi) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) -—o—Undowebd k=100 psi/ii +Undowebdk=250pst +Undowebdk=400psilh *Dowehdhloo psi/ii *Dmvebdmpsi/h *Dowobdk—MO psi/ii Figure E-8. Subgrade Stresses for a 18 kip (80.1 kN) Dual-Tired Single Axle on a 10” (254 mm) RCC Pavement Before and After Retrofit Dowels. 236 Subgrade Stress (psi) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) +Undmvebdk=lwpifm +Undowebdk=250pailil +Undowebdk=400pilh +Dmvebdk=lmpsifn *Dowebdmpaifn *Dowebdk‘MOOpai/h Figure E-9. Subgrade Stresses for a 18 kip (80.1 kN) Dual-Tired Single Axle on a 12” (305 mm) RCC Pavement Before and After Retrofit Dowels. Subgrade Stress (psi) 0% 10% 2096 30%) 4096 50% 60% 70% 809'- 90'/o 1 00% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) —O—Undowebdk-lwpsifn +mefi +Undowebdk=4mpaifn *Dowebdk-IMpsi/i +Dowfidb250psi/‘n *Dowebdkr—Mpsi/‘n Figure E-lO. Subgrade Stresses for a 18 kip (80.1 kN) Dual-Tired Single Axle on a 14” (3 56 mm) RCC Pavement Before and After Retrofit Dowels. 237 7 T 1 : 1 s I 6 k 1 \Hx 3\ N 5 _ '3 ; ; ; A A H\\. e- ' ' - - - *3... . gm 7 g 4 I a MM% m A A A A 4 A A A . '§ ' ' ' ' ' ' ' '7 a 3 1 .D 5 (n 2 1 0 1 0% l 0% 20% 30% 40% 50% 60% 70% 80% 90‘! o 1 00% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) —o—Undowebdk-100paifn +Wk—250pv. +Undowebdk=400pilii *Dowebdblmpsifm *Dowebdk-ZSOpsifm *Dowebdkafmpsil'm Figure El 1. Subgrade Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle on a 6” (152 mm) RCC Pavement Before and Afier Retrofit Dowels. 1 5 _ .__ i 'fl .— é .— '— t 5. h 3 ; ¢ ¢ ¢::- V1 3 0 '5 - El) .Q 5 2 1 0 0°16 10%) 2096 30% 4096 50% 60°16 7096 80% 90%) 10096 Deflection Load Transfer Efficiency Before Retrofit Dowels (%) —0—Undowebdk-lm par. +Undowebdk-250psifai +Undowebdk=4mpsilii *Dowebdki'loo psi/ii *Dowebdk-QSOpsi/‘n wombat-400w- Figure E-12. Subgrade Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle on an 8” (203 mm) RCC Pavement Before and After Retrofit Dowels. 238 Subgrade Stress (psi) .- N 5» 3- -‘ LA N l) W M ‘5 u. M .O u. O 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) +Wk=1w pi/ii +Undowebdb'250psil'ai +Undowebdk-400pail'n WDowebdk-loo psi/ii *Dowebdk-ZSOpsi/‘m *Dowebdk-Mps'fm Figure E-13. Subgrade Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle on a 10” (254 mm) RCC Pavement Before and After Retrofit Dowels. Subgrade Stress (psi) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) +Undoweiedk=100 par. +Undowebdk=250pilii +Wk‘4mpil'n ‘r'O-‘rDowehdki-lm psi/ii *Dowebdk-QSOpsi/‘m *DowebdkflOOpsi/h Figure E-14. Subgrade Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle on a 12” (305 mm) RCC Pavement Before and After Retrofit Dowels. 239 Subgrade Stress (psi) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90°» 100% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) —0—Undowebdk=100pailh +uwk=2503ya +UMk-Aoopirn} ~o~ooweuk=mopirn +Wk=2sopwh -+-Doweledk-4mpirn ! Figure E—15. Subgrade Stresses for a 24 kip (106.8 kN) Dual-Tired Tandem Axle on a 14” (3 56 mm) RCC Pavement Before and After Retrofit Dowels. Subgrade Stress (psi) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) —0—-Undowebd k-lOO pi/‘n +Undowehd k-2JO [pi/ii +Umiowehd k-400 psi/h *Dowebdk-lw psi/ii *Dowebdk-ZSOpsi/‘m *O—Dowebdk=400 psi/i: Figure E-16. Subgrade Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle on a 6” (152 mm) RCC Pavement Before and After Retrofit Dowels. 240 Subgrade Stress (psi) h) 0% 20%: Deflection load Transfer Efficiency Before Retrofit Doweb (%) 1 0% -—a—Undowebdk=lw psi/ii +Undowebdk=250 psi/ii +Undowebdk-400pi/ii *DowebdklOOpai/ii *Dowebdk-QSOpai/ii *DowedeOOpsi/‘m ,_ ' ‘1 “'4 ““WL-io-B“ N I‘ I k m e e e e e : : 3.; 30% 40% 50% 60% 70% 80% 90% 1 00%: Figure E-17. Subgrade Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle on an 8” (203 mm) RCC Pavement Before and After Retrofit Dowels. 0% 90%: 20% 30% 40% 50% 60% 70% 80% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) 10% +Undowebdk=1m psi/’- +Undowebdk=250peilh +Undowebdk=400psilii *Dowebdk-wOpsifn "O‘DowebthSOpsi/h *Dowebdkfloopeifn 100% Figure E-18. Subgrade Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle on a 10” (254 mm) RCC Pavement Before and After Retrofit Dowels. 241 Subgrade Stress (psi) 0% 1 0% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection Load Transfer Efficiency Before Retrofit Dowels (%) +Undowebd k=lw psi/ii +Undowebdk=250psilii +Undowebdk=400psilii *DowebdblOOpsi/ii +Dowebdk=250psilii *Dowebdkmimpsifn Figure E-19. Subgrade Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle on a 12” (305 mm) RCC Pavement Before and Afier Retrofit Dowels. i=— %. F. A It - I I | "1- 0% 10% N96 30% 40% 50% 60% 70% 80% 90% 100% Deflection load Transfer Efficiency Before Retrofit Dowels (%) —O—Undowebd k=100 peifm +Undowebd k=250 psi/ii +Undowebdk=400peilh ~+~Dowebdk=100psifn *Dowebdkfiwixifn WeDowebdkflOOpei/h Figure E-20. Subgrade Stresses for a 30 kip (133.4 kN) Dual-Tired Tandem Axle on a 14” (3 56 mm) RCC Pavement Before and After Retrofit Dowels. 242 Subgrade Stress (psi) 0% 10% 20% 30% 40% 50% 60% 70% 80% 9096 100% Deflection load Transfer Efficiency Before Retrofit Dowels (%) —0—Undowebdk=1wpsifn +Undowebdk=250 psi/ii +Undowebdk-4w psi/h *Dowehdk-lm Fifi: *Dowebdk=250 pei/ii *Dowebdk-flll psi/ii Figure E-21. Subgrade Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle on a 6” (152 mm) RCC Pavement Before and After Retrofit Dowels. Subgrade Stress (psi) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection load Transfer Efficiency Before Retrofit Dowels (%) +Wk=m par. +Uuiowebdk-250pil'm +Undowebdk=4mpsifn *Dowebd k=100 psi/ii +Dowebdk=250 psi/ii *Doweled k=400 psi/ii Figure E-22. Subgrade Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle on an 8” (203 mm) RCC Pavement Before and After Retrofit Dowels. 243 Subgrade Stress (psi) 0% 1 0% 20% 30% 40% 50% 60% 70% 80% 90% l 00% Deflection load Transfer Efficiency Before Retrofit Dowels (%) +Undowebd k=lm psifm +UndowebdFZ50 psi/ii +Undowebdk=4Npeilm +00webdk=100 psi/ii *DowebleZSO psi/“m *Dowebdkflmpsi/‘n Figure E-23. Subgrade Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle on a 10” (254 mm) RCC Pavement Before and After Retrofit Dowels. Subgrade Stress (psi) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 1 00% Deflection load Transfer Efficiency Before Retrofit Dowels (%) -+Wk=1mp.irn +1133)“ k=250 para +Undowebd k=400 pi/il *Dowebdk-lm psi/ii *DounbdeSOpei/ir m-t----Dowebdk=400psi/ii Figure E-24. Subgrade Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle on a 12” (305 mm) RCC Pavement Before and After Retrofit Dowels. 244 Subgrade Stress (psi) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Deflection load Transfer Efficiency Before Retrofit Dowels (%) +Undowebd k=100 psi/ii +Undowebd k=250 psi/ii +Undowebdk=400 psi/ii *Wmoo psi/it *WWSOpei/m *Mwoopva Figure E-25. Subgrade Stresses for a 36 kip (160.1 kN) Dual-Tired Tandem Axle on a 14” (3 56 mm) RCC Pavement Before and After Retrofit Dowels. 245 REFERENCES 246 - REFERENCES - . Pittman, D.W. Development of a Design Procedure for Roller—Compacted Concrete (RC C) Pavements. Ph.D. dissertation. University of Texas, Austin, TX. 1993. . Pittman, D.W. Construction of Roller-Compacted Concrete Pavements. Transportation Research Record 1 062, TRB, National Research Council, Washington, DC, 1986, pp. 13-19. . Roller-Compacted Concrete Pavements; Design arnd Construction, TM 5-822-7. Standard Practice Manual for Concrete Pavement — Appendix D. U.S. Army Corps of Engineers, Department of the Army, Washington, DC. . Suggested Specifications for Roller Compacted Concrete Pavingfor Municipal and Industrial Applications, CP014.01P. Canadian Portland Cement Association (CPCA), Ottawa, ON, CA, 1995. . Stewart, RR, and CJ. Schexnayder. Construction Techniques for Roller- Compacted Concrete. Transportation Research Record 1 062, TRB, National Research Council, Washington, DC, 1986, pp. 32-37. . White, T.D. Mix Desigrn, Thickness Design, and Construction of Roller- Compacted Concrete Pavement. Tiansportation Research Record 1 062, TRB, National Research Council, Washington, DC, 1986, pp. 1-6. . Munn, W.D. Roller Compacted Concrete Paves Factory Roads. Highway & Heany Construction, Sept. 1989. . Piggott, RW. Roller-Compacted Concrete for Heavy-Duty Pavements: Past Performance, Recent Projects, and Recommended Construction Methods. Transportation Research Record 1 062, TRB, National Researdn Conmcil, Washington, DC, 1986, pp. 7-12. . Cortez, ER, and 1A Gerlach. Roller-Compacted Concrete Pavement Construction at Fort Drum, New York. Transportation Research Record 1 282, TRB, National Research Council, Washington, DC, 1990, pp. 8-17. 10. Al-Abdul Wahhab, 111., and I.M. Asi. Optimization of Roller-Compacted Concrete for Local Application Transportation Research Record 1 458, TRB, National Research Conmcil, Washington, DC, 1994, pp. 1-7. 1 1. State-of-the-Art Report on Roller-Compacted Concrete Pavements, AC1 325.10R- 95. AC! Manual of Concrete Practice, American Concrete Institute, Detroit, MI, 1 995. 247 12. Tayabji, SD, and PA. Okamoto. Engineering Properties of Roller-Compacted Concrete. Transportation Research Record 1 1 36, TRB, National Research Council, Washington, DC, 1987, pp. 33-45. 13. Nanni, A. Properties and Desigrn of Fiber Reinforced Roller Compacted Concrete. Transportation Research Record 1226, TRB, National Research Council, Washington, DC, 1989, pp. 61-68. 14. Structural Design of Roller-Compacted Concrete for Industrial Pavements. 18233.01. Portland Cement Association (PCA), Skokie, 11., 1987. 15. Piggott, RW. Roller Compacted Concrete Pavements - A Study of Long Term Performance. PCA R&D Serial No. 2261. Portlarnd Cement Association (PCA), Skokie, IL, 1999. 16. Ragan, S.A. Evaluation of the Frost Resistance of Roller-Compacted Concrete Pavements. Transportation Research Record 1 062, TRB, National Research Council, Washington, DC, 1986, pp. 25-32. 17. Pigeon, M., and V.M. Malhotra Frost Resistance of Roller-Compacted High- Volurne Fly Ash Concrete. Journal of Materials in Civil Engineering, Nov. 1995, pp. 208-211. 18. Dolen, T.P. Freeze-Thaw Durability of Roller-Compacted Concrete (Part II). A Paper Prepared for the 1996 Spring Convention of the American Concrete Institute, Denver, CO, Mar. 1996. 19. Petrie, RE, and SC. Matthews. Roller Compacted Concrete Pavements: Recent Australian Developments arnd Prospects for the 905. Proceedings of the 15'” Australian Road Research Board Conference, Melbourne, Australia, 1990, pp. 65- 87. 20. Jameson, G.W., Sharp, K.G., Bailey, DP, and RA. Warwick. A Field Trial of the Use of Roller Compacted Concrete irn High-Spwd Trafiic Applications. Proceedings of the 15'” Australian Road Research Board Conference, Melbourne, Australia, 1990, pp. 143-168. 21. Todres, HA, Jacobs, A.P., Hang, P., and TJ. McLaughlin. An Experimental Section of Roller Compacted Concrete Arterial Road in Edmonton, Alberta. PCA R&D Serial No. 1980. Portland Cement Association, 1992. 22. Wu, C.L., arnd HA Todres. Improving Roller Compacted Concrete Pavement Technology: Pavement Evaluation and Thickness Design. PCA R&D Serial No. 2016. Portland Cement Association, 1995. 23. Smith, KD., HT. Yu, M.J. Wade, D.G. Peshkin, arnd MI. Darter. Performance of Concrete Pavements, Volume I - Field Investigation. Report FHWA-RD-94-1 7 7. FHWA, U.S. Department of Transportation, 1997. 248 24. Hall, KT. Backcalculation Solutions for Concrete Pavements. A Paper Prepared for Long-Term Pavement Performance Data Analysis. Report SI-IRP-P-ZO. SI-IRP, National Research Council, 1992. 25. Darter, M.I., K.T. Hall, and C. Kuo. Support under Portland Cement Concrete Pavements. NCHRP Report 372. TRB, National Research Council, 1995. 26. Wu, CL, and HA. Todres. Non-Destructive Evaluation of Roller Compacted Concrete Pavements at Fort Drum. New York. PCA R&D Serial No. 2012. Portland Cement Association, 1994. 27. Ioanrnides, AM, and GT. Korovesis. Aggregate Interlock: A Pure-Shear Load Transfer Mechanism Transportation Research Record 1286, TRB, National Research Council, Washington, D.C., 1990, pp. 14-24. 28. Benkelman, AC. Tests of Aggregate Interlock at Joints arnd Cracks. Engineering News-Record, Vol. 111, No. 8, Aug. 24, 1933, pp. 227-232. 29. Bucln, N.J. DeveIOpment of Empirical-Mechanistic Based Faulting Models in the Design of Plain Jointed Concrete Pavements. Ph.D. Dissertation. Texas A&M University, College Statiorn, TX, 1995. 30. Nanni, A, Ludwig, DA, and J .E. Shoenberger. Physico-Mechanical Properties and Load Transfer Efficiency of RCC Pavement. American Concrete Institute (AC1) Materials Journal, Vol. 93, No. 4, pp. 356-361. 31. Teller, L.W., arnd HD. Cashell. Performance of Dowels Under Repetitive Loading. Public Roads, Vol. 30, No. 1, April 1958, pp. 1-24. 32. Kilareski, W.P., MA. Ozbeki, arnd D.A Anderson. Fourth Cycle of Pavement Research at the Pennsylvania Transportation Research Facility — Vol. 4: Rigid Pavement Joint Evaluation and Full Depth Patch Designs. Report FHWA/PA-84- 026. The Pennsylvania Transportation Institute, The Pennsylvania State University, University Park, PA, December 1984. 33. Ozbeki, MA, W.P. Kilareski , arnd D.A. Anderson. Computer Simulation arnd Field Evaluation of Transverse Joints in Rigid Pavements. Proceedings of the Third International Conference on Concrete Pavement Design and Rehabilitation, Purdue University, West Lafayette, IN, Apr. 23-25, 1985, pp. 577-586. 34. Tayabji, SD, and DJ. Halpenny. Thickness Desigrn of Roller-Compacted Concrete Pavements (Discussion arid Closure). Transportation Research Record I 136, TRB, National Research Council, Washington, D.C., 1987, pp. 23-32. 35. Huang, Y.H. Pavement Analysis and Design. Prentice-Hall, Inc., Englewood Cliffs, NJ, 1993. 36. Shankar, P.R., arnd C.E. Lee. Lateral Placement of Truck Wheels Witlnin Highway Larnes. Transportation Research Record 1 04 3, TRB, National Research Council, Washington, D.C., 1985, pp. 33-39. 249 37. Benekohal, RF., K.T. Hall, and HW. Miller. Efl‘ect of Lane Widening on Lateral Distribution of Truck Wheels. Transportation Research Record 1286, TRB, National Research Council, Washington, D.C., 1990, pp. 57-66. 38. Taragin, A. Lateral Placements of Trucks on Two-Lane Highways and F our-Lane Divided Highways. Public Roads, Vol. 30, No. 3, Aug. 1958, pp. 71-75. 39. Miller, El, and 6N. Steuat. Vehicle Lateral Placenents on Urban Roads. Transportation Engineering Journal. ASCE, Vol. 108, No. TES, Sept. 1982, pp. 495-508. 40. Taragin, A. Transverse Placement of Vehicles as Related to Cross Section Design. Proceedings, HRB, National Research Council, Washington, D.C., Vol. 23, 1943, pp. 342-350. 41. Shankar, P.R Lateral Placement of Truck Traffic in Highway Lanes. Ph.D. Dissertation. Urniversity of Texas, Austin, TX, 1984. 42. Poblete, M., R Valenzuela, and R Salsilli. Load Transfer in Undoweled Transverse Joints of PCC Pavements. Transportation Research Record 1207, TRB, National Research Council, Washington, D.C., 1988, pp. 39-49. 43. Frabizzio, M.A, N. Buch, and J.E. Hiller, Impact of Coarse Aggregates on Transverse Crack Performance in Jointed Concrete Pavements, AC1 Materials Journal, Vol. 97, No. 3, May-June 2000, pp.325-332. 44. Raja, 2.1, and MB. Snyder. Factors Affecting Deteioration of Transverse Cracks in Jointed Reinforced Concrete Pavements. Transportation Research Record 1307, TRB, National Research Courncil, Washington, D.C., 1991, pp. 162-168. 45. Snyder M.B., MJ. Reiter, RT and MI Dater. Rehabilitation of Concrete Pavements, Volume I — Repair Rehabilitation Techniques. Report FHWA-RD-88- 071. FHWA, U.S. Department of Transportation, 1989. 46. Larson, RM., D. Peterson, and A. Correa Retrofit Load Transfer — Special Demonstration Project SP-204. Report FHWA-SA-98-047. FHWA, U.S. Department of Transportation, 1998. 47. Marniouk, MS., M. Backer, and E. Owusu-Antwi. Dowel Retrofitting — A Viable Method for Preventative Maintenance. Presented at the 79" Annual Meeting of the Transportation Research Board, TRB, National Research Council, Washingtorn, D.C., 2000. 48. Piece, L.M. Portland Cement Concrete Pavement Rehabilitation in Washington State: Case Study. Transportation Research Record 1449, TRB, National Research Council, Washington, D.C., 1994, pp. 189-198. 49. Piece, L.M. Design, Construction, Performance and Future Direction of Dowel Bar Retrofit irn Waslnington State. Sixth International Purdue Conference on 250 Concrete Pavement Design and Materials for High Performance. Vol. 2, Indianapolis, IN, Nov. 18-21, 1997, pp. 87-100. 50. Embacher, RA and MB. Snyder. Minne—ALF Project Overview and Results of Tests of PCC Slab Specimens Containing Retro-F it Dowels. Paper Prepared for the Intennational Conference on Accelerated Pavement Testing, TRB, Reno, NV, 1 999. 51. Zaniewski, J .P., and MS. Marrnlouk. Preventive Maintenance Effectiveness — Preventive Maintenance Treatments, Report FHWA-SA-96-027, FHWA, U.S. Department of Transportation, 1996. 52. Galehouse, L. Innovative Concepts for Preventative Maintenance. Transportation Research Record 1 62 7, TRB, National Research Council, Washington, D.C., 1999, pp. 1-6. 251