Application of DCP in Prediction of Resilient Modulus of Subgrade Soils

Transcription

1 Application of DCP in Prediction of Resilient Modulus of Subgrade Soils Louay Mohammad, Ph.D. Louisiana Transportation Research Center Louisiana State University 2006 Pavement Performance Seminar April 10, 2006, Ruston, Louisiana April 11, 2006, Alexandria, Louisiana April 12, 2006, Baton Rouge, Louisiana

2 Topics Background Objective Scope Methodology Experimental program Results

3 What is Resilient Modulus? M R Stiffness of Soils Under Dynamic Loading

4 Why Resilient Modulus was Selected? More realistic way to characterize moving wheel loads AC Base Subgrade

5 Deviator stress σ D Time Stress or load pulse Deviator stress Total ε Plastic Resilient ε R ε P MR= σ ε D R Strain Deformation of soil sample

6 Where is M R Used? Design of New & Rehabilitated pavements Forensic analysis of Pavement Failures Quality Control for fill or cut sections

7 Where is M R Used? Design of New & Rehabilitated pavements LADOTD 1993 AASHTO log W = Z S log ( SN + 1) R o 10 + log ΔPSI ( SN + 1) log M R

8 Sensitivity Example of M r to Flexible Pavement Thickness Design (1993) Change in overlay (in) Change in Mr (ksi)

9 Where is M R Used? Design of New & Rehabilitated pavements M-E Design Guide Traffic Climate Structure Thermal

10 Where is M R Used? Design of New & Rehabilitated pavements M-E Design Guide Input Material Level 1 Asphalt Concrete Measured DM Input Level 2 Estimated DM Input Level 3 Default DM PCC Measured EM Estimated EM Default EM Stabilized Materials Measured M R Estimated M R Default M R Granular Materials Measured M R Estimated M R Default M R Subgrade Measured M R Estimated M R Default M R

11 Where is M R Used? Forensic analysis of Pavement Failures Assessment of soil conditions develop an appropriate rehabilitation strategy

12 Where is M R Used? Quality Control for fill or cut sections pavement section is designed based upon a targeted resilient modulus

13 How is Resilient Modulus Determined? Direct Measurement Lab Test: AASHTO T 307 Undisturbed Disturbed, remolded and compacted Reverse Engineering in-situ DCP, Minincone nondestructive test (NDT) methods FWD, Dynaflect Prediction Soil properties SSV (DOTD) other Unconfined compressive strength CBR

14 Limitations: Direct Measurement Direct Measurement triaxial type of test Laborious and time consuming Require advanced testing equipment Skilled Personnel

15 Limitations: M R Predictions Models Static properties of soils SSV, etc: average value for each Parish Do not represent the dynamic behavior of traffic loading M R = [(53/5)(SSV-2)] - 2.5[(53/5)(SSV-2)] 2

16 Alternative In-situ Technique M R Prediction Models Dynamic cone penetration (DCP) fast, simple, and economical Geotechnical investigation

17 Objectives Develop M R prediction Models In-situ DCP test results Soil properties

18 Scope Thirty one sites Four common cohesive soil types in Louisiana A-4, A-6, A-7-5, and A-7-6 Field DCP tests Various moisture-density levels Laboratory test Mr from RLTT Soil physical properties Three tests per site

19 EXPERIMENTAL PROGRAM Field Activities Field projects DCP Soil Sampling Laboratory tests Repeated load triaxial test Physical properties Compaction and strength characteristics

20 Field Testing Layout Three sections of testing at each field project A, B, and C Each section contains nine points ' 3' 15' 3' 3' 15' 3' 3' Core Core Core DCP Mini- DCP Mini- DCP Cone Cone Mini- Cone Set A

21 Dynamic Cone Penetrometer Test: DCPI Stone Cement Treated Subgrade Compacted Embankment Depth Depth (cm) mm/blow 7 mm/blow Blow count Blow Count 1 mm/blow

22 Repeated Load Triaxial Test: M R Resilient modulus, Mr (MPa) 100 Silty clay Deviator stress, σd (kpa) Sample depth=0.8 m Confining stress σ c =41.3 kpa σ c σ c =20.7 kpa =0.0 kpa

23 Relationship B/W Mr and 1/DCPI R 2 = 0.82 Mr, MPa /DCPI, Blows/mm

24 Development of M R Prediction Models DCPI M R Moisture Content Density M R = f(dcpi, physical properties)

25 Resilient Modulus Prediction Models Direct Model: R 2 = M R = DCPI Parameter M r (ksi) DCPI (mm/blow) PI (%) Range γ d (pcf) Soil Property Model: R 2 =0.89 M R 1 = DCPI 1.32 γ d w W (%) LL (%) Silt (%) Clay (%) Passing #200 Sieve (%) 42-97

26 Measured vs. Predicted: Direct Model 1 M R = DCPI Predicted M r (MPa) Modeling Equality line Verification UCL- Upper Control Limit LCL- Lower Control Limit Measured M r (MPa)

27 Measured vs. Predicted: Soil Property Model M R = DCPI γ d 0.42 w Predicted M r (MPa) Modeling Equality line Verification UCL- Upper Control Limit LCL- Lower Control Limit Measured M r (MPa)

28 Measured vs. Predicted: Parish Map Parish Mr (ksi) Equality Line M r Lab-M r (ksi) Parish- M r (ksi) Range Lab M r (ksi)

29 Summary M R prediction Models were developed DCP = Good agreement Measured and predicted 1 M R = Improvement DCPI Current practice Tools 1 M R DCPI Design of New & Rehabilitated pavements Current & M-E Design Guide Forensic analysis of Pavement Failures Quality Control for fill or cut sections 1.32 γ d w

30 Report No. 417

31