Effect of Cementation on Cone Resistance in Sands: A

Louisiana State UniversityLSU Digital Commons

LSU Historical Dissertations and Theses Graduate School

1993

Effect of Cementation on Cone Resistance inSands: A Calibration Chamber Study.Anand Jagadeesh PuppalaLouisiana State University and Agricultural & Mechanical College

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion inLSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please [email protected].

Recommended CitationPuppala, Anand Jagadeesh, "Effect of Cementation on Cone Resistance in Sands: A Calibration Chamber Study." (1993). LSUHistorical Dissertations and Theses. 5687.https://digitalcommons.lsu.edu/gradschool_disstheses/5687

https://digitalcommons.lsu.edu?utm_source=digitalcommons.lsu.edu%2Fgradschool_disstheses%2F5687&utm_medium=PDF&utm_campaign=PDFCoverPages

https://digitalcommons.lsu.edu/gradschool_disstheses?utm_source=digitalcommons.lsu.edu%2Fgradschool_disstheses%2F5687&utm_medium=PDF&utm_campaign=PDFCoverPages

https://digitalcommons.lsu.edu/gradschool?utm_source=digitalcommons.lsu.edu%2Fgradschool_disstheses%2F5687&utm_medium=PDF&utm_campaign=PDFCoverPages

https://digitalcommons.lsu.edu/gradschool_disstheses?utm_source=digitalcommons.lsu.edu%2Fgradschool_disstheses%2F5687&utm_medium=PDF&utm_campaign=PDFCoverPages

https://digitalcommons.lsu.edu/gradschool_disstheses/5687?utm_source=digitalcommons.lsu.edu%2Fgradschool_disstheses%2F5687&utm_medium=PDF&utm_campaign=PDFCoverPages

mailto:[email protected]

INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, begimiing at the upper left-hand corner and continuing firom left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

University Microfilms International A Bell & Howell Information Company

300 Nortfi Zeeb Road. Ann Arbor. Ml 48106-1346 USA 313,'761-4700 800/521-0600

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.


Order N u m b er 9419941

Effect of cementation on cone resistance in sands: A calibration chamber study

P u p p a ia , A nand Jagadeesh , Ph .D .

The Louisiana State University and Agricultural and Mechanical Col., 1993

Copyright © 1994 by Puppaia, Anand Jagadeesh. All rights reserved.

300 N. Zeeb Rd.Ann Arbor, Ml 4SI06



EFFECT OF CEMENTATION ON CONE RESISTANCE IN SANDS:

A CALIBRATION CHAMBER STUDY

A Dissertation

Subm itted to the G raduate Faculty of the Louisiana S ta te University and

Agricultural and Mechanical College in partial fulfillment of the

requirements for the degree of Doctor of Philosophy

in

The D epartm ent of Civil Engineering

IbYAnand Jagadeesh Puppaia

B.S., Andhra University, Visakhapatnani. India, 1985 M.S., Indian Insti tu te of Technology, Madras, India, 1987

May 1993


Acknowledgements

I thank Prof. Yalcin B. Acar for his support, guidance and suggestions provided

th roughout th e course of th is research project. As an advisor, he provided th e

im petus and guidance for this study. His boundless energy, perseverance and the

ab ility to cut complex concepts to sim pler ones have provided the m otivation and

direction. His support, patience and particularly the friendship th roughout th is

research are sincerely appreciated.

I also thank Prof. M ehmet T. Tum ay for his suggestions, guidance and friendship

th roughou t th is research. His help in providing the equipm ent for th is research and

also his suggestions for equipm ent design are greatly appreciated. I wish to extend

p a rticu la r thanks to Prof. Roger. K. Seals, Prof. George Z. Voyiadjis, P rof S. S.

Iyengar, P rof W ije W athugala for serving in my dissertation com m ittee. Special

thanks go to Prof. Roger K. Seals for his support during my association w ith him

on th e boiler slag pro ject. Special thanks also go to Prof. Ilan Ju ra n for serving in

my general exam ination com m ittee. I would also hke to th an k Prof. K. Senneset

of University of Trondheim , Trondheim , Norway for his detailed and m eticulous

description of J & S theory.

T he faculty and staff of D epartm ent of Civil Engineering have contribu ted to

th is work by the ir support and in terest in this project. This study is supported by

the N ational Science Foundation under G rant No. MSS-9020368 and D epartm ent

of Civil Engineering. T he au thor is grateful for the research assistan tsh ip provided.

T he calibration cham ber used in th is study was designed and fabricated by Dario de

L im a under the guidance of Prof. M ehm et T . Tumay. The Louisiana T ransporta tion

Research Center has provided funding for this equipm ent under G rant No. 88-1 GT.

T his support is gratefully acknowledged. I would also like to thank Fugro-M cClelland

Engineers, Inc., H ouston, Texas for providing the cone equipm ent for th is study.

I am also thankfu l to my friends, Rainer Echle and G abriela Segarra for their

constan t encouragem ent, help and m ost im portan tly their friendship th roughou t th is

study. I also would hke to thank my brother Susheel and cousin Anil for th e ir help

in my study. I also would like to thank Sivakumar, A1 Lopez, Semih A rslan, A kram


Alshwabekih, G anesh and Loga, Pradeep K urup, R aihan Taha, W ang Junxiong,

Fatim a C hajia and o ther friends for their help in my work. T he value of their

friendship will not be forgotten.

I am also indebted to John and MaceUe Vincent, Bill T ierney for adding a lot to

my life a t LSU. I also hke to thank John, Bill and the staff a t th e m achine shop who

aided in the design and fabrication of the equipm ent. L aboratory support provided

by Louisiana T ransport Research Center is also acknowledged. 1 also would like to

extend my g ra titude for the help provided by Messrs. Paul Griffin, John Oglesby

and Ken Johnston.

I also would like to thank Chris Gascon, Mike Orcino and Court B radford who

assisted me in the experim ental phase of my research. T heir support, constant

encouragem ent, w it and hum or have kept me going. I also would like to th an k all

my other colleagues for their help during the calibration cham ber testing.

Finally, I would like to extend my sincere thanks to my beloved parents, sister,

her husband and my beloved grandm other, and also to other m em bers of my family. I

have felt their love and caring throughout my life and appreciate their understanding

and guidance.

This dissertation is dedicated to m y beloved parents.

To My Mom and Dad

. - Chinni

11]


Contents

Acknowledgements ii

List of Tables viii

List of Figures x

List of Plates ^vii

Abstract xviii

1 INTRODUCTION 11.1 In tro d u c tio n .............................................................................................................. 11.2 O b je c tiv e s ................................................................................................................. 51.3 O rganization of the M anuscript and S u m m a r y ........................................... 6

2 SYNTHESIS OF AVAILABLE INFORMATION 82.1 In tro d u c tio n .............................................................................................................. 82.2 Engineering Behavior of Cem ented S a n d s ...................................................... 9

2.2.1 Definitions and Physio-Chem ical C h a ra c te r is tic s .......................... 92.2.1.1 C em entation in S a n d s ......................................................... 112.2.1.2 C em entation in Collapsing S o i ls ........................................ 112.2.1.3 Cem entation in R o c k s ......................................................... 122.2.1.4 S truc tu re of C em ented G ranular Soils .......................... 12

2.2.2 D isplacem ent R ate Controlled Stress-D eform ation Behavior (S ta ticB e h a v i o r ) ................................................................................................... 13

2.2.3 Unconfined Com pressive S trength , q j ............................................... 212.2.4 H ydraulic C o n d u c tiv ity ........................................................................... 222.2.5 C o m p re s s ib il ity ......................................................................................... 222.2.6 N a tu ra l Versus Artificial C e m e n ta tio n ............................................... 242.2.7 D ynam ic C haracteristics - I n tro d u c tio n ............................................ 27

2.2.7.1 Large S train D ynam ic Stress-D eform ation Behavior(D ynam ic Triaxial T e s t i n g ) .............................................. 28

2.2.7.2 R esonant Colum n Test (Low Strain D ynam ic Behavior) 292.3 Cone P ene tra tion Testing in Sands ................................................................. 31

2.3.1 Cone P e n e tr o m e te r .................................................................................. 332.3.2 C alibration C ham ber T e s t i n g ............................................................. 35

2.3.2.1 C ham ber Size and Boundary Condition Effects . . . 372.3.2.2 C ru sh a b ility .............................................................................. 382.3.2.3 Fabric, Shape and Texture of G r a i n s ............................. 41

iv


2.3.3 Synthesis of Experim ental D a ta O btained in C alibration C ham ber Testing of Uncem ented S a n d s ..................................................... 41

2.4 Cone P enetration Testing in Cem ented S a n d ............................................... 422.5 Cone P enetration Testing Analysis ................................................................ 45

2.5.1 Bearing C apacity T h e o r ie s ....................................................................... 452.5.2 Cavity Expansion T h e o rie s ................................................................... 492.5.3 S train P a th Approach .......................................................................... 532.5.4 N um erical M e th o d s ................................................................................. 552.5.5 D isc u ss io n ................................................................................................... 552.5.6 Friction Resistance (Sleeve F ric tio n ) .................................................. 562.5.7 Em pirical M e th o d s ................................................................................. 572.5.8 S ta te Param eter I n te r p r e ta t io n ......................................................... 63

2.6 S u m m a r y ............................................................................................................... 67

3 METHODOLOGY 683.1 In tro d u c tio n ............................................................................................................ 683.2 Experim ental M o d e l .......................................................................................... 69

3.2.1 E q u ip m e n t................................................................................................... 703.2.1.1 P luviation S e t u p .................................................................... 703.2.1.2 Saturation S e t u p .................................................................... 723.2.1.3 The LSU Calibration Cham ber F a c il ity ............................ 733.2.1.4 Saturation and Vacuum C o n n e c tio n s .............................. 763.2.1.5 Vacuum P u m p ........................................................................ 763.2.1.6 The M iniature Q uasi-Static Cone Penetrom eter . . . 763.2.1.7 The Auxiliary equipm ent ................................................... 773.2.1.8 Control P a n e l ........................................................................... 773.2.1.9 Hydraulic S y s te m .................................................................... 803.2.1.10 D ata Acquisition and M onitoring S y s te m ........................ 813.2.1.11 The Triaxial S y s t e m ............................................................. 833.2.1.12 Unconfined Compression Tests ........................................ 83

3.2.2 Procedure - Uncem ented S p ec im en s.................................................. 843.2.2.1 Triaxial T e s t s ........................................................................... 843.2.2.2 C alibration C ham ber T e s t i n g ............................................ 88

3.2.3 Procedure - Cem ented S p e c im e n s ...................................................... 983.2.3.1 Triaxial T e s t s ........................................................................... 983.2.3.2 Calibration C ham ber T e s t i n g ............................................... 1013.2.3.3 Unconfined Compression T e s t ............................................... 108

3.3 Sum m ary ................................................................................................................... 108

4 ENGINEERING BEHAVIOR OF MONTEREY NO 0/30 SAND 1094.1 In tro d u c tio n ................................................................................................................1094.2 Strength-D eform ation-Pore Pressure R e s p o n s e ..............................................1104.3 S trength P a r a m e te r s .............................................................................................. I l l4.4 Com parison w ith D rained R e s u l t s ...................................................................... 1224.5 C ritical S ta te D ia g r a m ...........................................................................................123


4.6 M odeling P a r a m e te r s ............................................................................................. 1294.6.1 Yield Function ............................................................................................1294.6.2 H ardening F u n c tio n .....................................................................................1304.6.3 Flow F u n c tio n ................................................................................................1314.6.4 P lastic Poten tial F u n c t io n ....................................................................... 1314.6.5 M odel R e s u l ts ................................................................................................132

4.7 Unconfined Com pression T e s ts ............................................................................ 1434.8 S u m m a r y .................................................................................................................. 148

5 CONE PENETRATION TESTING 1505.1 Testing Program .................................................................................................... 151

5.1.1 U ncem ented S pec im ens.............................................................................. 1515.1.2 Cem ented S p e c im e n s ..................................................................................152

5.2 Perform ance A s s e s s m e n t...................................................................................... 1615.2.1 R e p e a ta b i l i ty ............................................................................................... 1615.2.2 Accuracy ...................................................................................................... 168

5.3 Influence of Testing Variables on Cone Test R e s u l t s .................................... 1685.3.1 Uncem ented Specimen R e s u lts ............................................................... 1685.3.2 Cem ented S p e c im e n s ..................................................................................175

5.4 S u m m a r y ...................................................................................................................175

6 FACTORS INFLUENCING TEST RESULTS 1816.1 In tro d u c tio n ...............................................................................................................1816.2 Relative D e n s ity ........................................................................................................ 182

6.2.1 Com parison W ith Results Reported by Villet and M itchell (1981) 1826.2.2 Com parisons W ith Results Reported for O ther S a n d s .....................184

6.3 Factors Influencing Penetration Resistance in C alibration C ham ber . 1876.3.1 C ham ber S i z e ................................................................................................1876.3.2 Com pressibility and Crushability of the Sand Tested ................... 1916.3.3 Boundary C o n d itio n s ..................................................................................2006.3.4 Influence of Soil Particle Size and S h a p e ........................................... 2026.3.5 C e m e n ta tio n ...................................................................................................206

6.4 Sum m ary ................................................................. 211

7 ANALYSIS OF TEST RESULTS: Theoretical and Empirical Methods 212

7.1 In tro d u c tio n ...............................................................................................................2127.2 T heoretical M e th o d ................................................................................................. 213

7.2.1 T ip Resistance - Bearing Capacity Theories .................................... 2137.2.1.1 D & M T h e o ry ......................................................................... 2137.2.1.2 J & S Theory .........................................................................213

7.2.2 T ip Resistance - Cavity Expansion T h e o r ie s ................................... 2177.2.2.1 Cavity Expansion Theory - Procedure 1 ........................... 2227.2.2.2 Cavity Expansion Theory - Procedure 2 ........................... 229

7.2.3 Friction R e s is ta n c e ..................................................................................... 232

VI


7.2.4 Approach for Estim ating Cohesion and Friction Angles . . . . 2.347.3 Empirical M e th o d ......................................................................................................239

7.3.1 Empirical Method Based on a A p p r o a c h .......................................... 2397.3.1.1 Tip R e s i s t a n c e ...........................................................................2397.3.1.2 Friction R e s i s t a n c e ....................................................................2467.3.1.3 Approach 1 ..................................................................................2507.3.1.4 Approach 2 ..................................................................................251

7.3.2 Empirical Method Based on S tate Param eter Approach . . . . 2597.4 Influence of Vertical Confining Pressure on Cementation ........................ 2657.5 D is c u ss io n ....................................................................................................................2657.6 S u m m a r y ....................................................................................................................268

8 SUMMARY AND CONCLUSIONS 2748.1 S u m m a r y ....................................................................................................................2748.2 C o n c lu s io n s ................................................................................................................ 2768.3 Recommendations for Future S t u d i e s ............................................................... 278

References 279

Appendix A 291A .l Literature on Cemented S a n d s ............................................................................. 291A.2 Dynamic Properties [G^nax) .................................................................................300A.3 Cone Test Results on O ther S a n d s ...................................................................... 302

Appendix B 309B .l CU T e s ts ....................................................................................................................... 309B.2 Stress P a t h s ................................................................................................................ 312B.3 Flow Charts and L is t in g s ........................................................................................318

Appendix C 329C.l Cone Test R e s u l t s ......................................................................................................329

Appendix D 350D .l Program Listings for Cavity Expansion M o d e ls ..............................................350

Vita 363

Vll


List of Tables

2.1 Strength Param eters of Cemented Sands (Drained Tests)(Rad andClough, 1 9 8 2 ) .......................................................................................................... 21

2.2 Proposed Classification System for Cemented Granular Soils (Rad and(%ough, 1 9 8 2 ) .......................................................................................................... 22

2.3 UCS of Cemented S a n d s ...................................................................................... 242.4 Young’s Modulus at the 50 and 25 % Failure Stresses for Un cemented

Monterey No. 0 Sand (C.C. 0 % ) ................................................................... 252.5 Young’s Modulus a t the 50 and 25 % Failure Stresses for Cemented

Monterey No. 0 Sand (C.C. 1 % ) ................................................................... 252.6 Young’s Modulus a t the 50 and 25 % Failure Stresses for Cemented

Monterey No. 0 Sand (C.C. 2 % ) ................................................................... 262.7 Strength Param eters of Artificial and Natural Cemented Sands . . . . 272.8 Stiffness Coefficients and n Values of Cemented S a n d s ............................. 312.9 Characteristics of Tested S a n d s ........................................................................ 122.10 Tip and Friction Resistances Reported for Cemented S a n d s ................... 43

3.1 Number of Tests (Calibration Cham ber) ...................................................... 693.2 Number of Undrained Triaxial T e s t s ................................................................. 693.3 Drained Triaxial Tests (Arslan, 1 9 9 3 ) ............................................................. 703.4 Calibration Factors of the C o n e ........................................................................ 773.5 Index Properties of Monterey No. 0/30 S a n d ............................................... 84

4.1 Effective Strength Param eters of Cemented Sands (Undrained Tests) 1224.2 A Comparison of Cohesion Values Obtained in Drained and Un drained

T e s t s ................................................................................................................................1234.3 A Comparison Between Effective Friction Values O btained in Drained

and Undrained T e s t s ................................................................................................1244.4 Modelling P a ra m e te r s ................................................................................................. 1344.5 Modelling P a ra m e te r s ................................................................................................. 134

5.1 Characteristics of Tests (Uncemented) ..............................................................1575.2 Penetration Results (U ncem en ted ) ........................................................................ 1615.3 Characteristics of Tests (C.C. 1 % ) .....................................................................1685.4 Characteristics of Tests (C.C. 2 % ) .....................................................................1705.5 Penetration Results (C.C. 1 % ) ............................................................................ 1715.6 Penetration Results (C.C. 2 % ) ............................................................................171

6.1 The a and n for Various I n v e s t ig a t io n s ..............................................................1926.2 The Q and n for Various C e m e n t a t i o n s ............................................................. 206

Vl l l


7.1 Com parison of Cavity Expansion T h e o r ie s .................................................... 2337.2 T he n for Various C e m e n ta tio n s .......................................................................... 2447.3 T he ay and riy for Various In v e s tig a tio n s ........................................................ 2467.4 T he /? and ni for Various C e m e n ta tio n s ............................................................2507.5 T he Steady S ta te P a r a m e t e r s ............................................................................. 2617.6 T he S trength P a ra m e te rs ........................................................................................261

A .l A Sum m ary of Geotechnical Studies Conducted on Cem ented Sands . 292A .2 Synthesis of D a ta R eported in Studies Investigating Cem ented Sands 295A .3 Test R esults - Bid (1 9 8 7 ) ................................................................. 302A.4 Test Results - Baldi (1 9 8 1 ) ............................................................. 303A.5 Test R esults - V illet and M itchell ( 1 9 8 1 ) ................................... 304A .6 Test Results - H arm an ( 1 9 7 6 ) ....................................................... 305A .7 Test R esults - Fioravante ( 1 9 9 2 ) ................................................... 306A.8 Test R esults - M anassero ( 1 9 9 2 ) ................................................... 307A .9 Test R esults - N u tt and Houlsby (1 9 9 2 )....................................... 307A .10 Test Results - Lhuer (1976) .......................................................... 308

IX


List of Figures

2.1 Various S tructures in Cem ented Soils (Sowers and Sowers, 1979) . . . 142.2 D rained Triaxial Results on an Uncem ented Specim en of M onterey

No. 0 Sand at R elative Density 45-55 % (R ad and Tumay, 1986) . . . 162.3 D rained Triaxial Results on a 1 % Cem ented Specimen of M onterey

No. 0 Sand at R elative Density 45-55 % (Rad and Tumay, 1986) . . . 172.4 D rained Triaxial Results on a 2 % Cem ented Specim en of M onterey

No. 0 Sand at R elative Density 45-55 % (Rad and Tumay, 1986) . . . 182.5 Influence of Cem ent C ontent on Friction Angle of M onterey No. 0

Sand (R ad and Tumay, 1 9 8 6 ) .......................................................................... 192.6 Influence of Cem ent C ontent on Cohesion (R ad and Tumay, 1986) . . 202.7 Hydraulic Conductivity of Cem ented Specimens (El-Tahir and Acar,

1 9 8 3 ) ........................................................................................................................... 232.8 Stress R atio Versus N um ber of Cycles: Cyclic Triaxial Tests on 1 %

Cem ented Specimen of Relative Density 51 % (Rad and Clough, 1982) 302.9 The Variation of M axim um Shear M odulus Versus Confining Stress

for Cem ented Specimens P repared a t Relative Density of 50 % (A carand El-Tahir, 1 9 8 6 ) ............................................................................................... 32

2.10 Schem atic of a Electrical Cone (Juran and Tumay, 1 9 8 9 ) ..................... 342.11 Cham ber Size and Boundary Condition Effects (Parkin and Lunne,

1982) 392.12 Influence of Crushing (Bellotti et al., 1 9 9 1 ) ................................................. 402.13 P enetra tion Profiles in Laboratory Tests (Rad and Tumay, 1986) . . . 442.14 Failure M echanism Assum ed in Durgunoglu and M itchell’s Theory

(Durgunoglu and M itchell, 1973) 472.15 Failure M echanism Assum ed in Janbu and Senneset’s Theory (Janbu

and Senneset, 1 9 7 4 )............................................................................................... 482.16 Com parison Between the M easured and Predicted P aram eters (Baldi

et ah, 1 9 8 1 ) ............................................................................................................. 502.17 Com parison Between the M easured and Predicted P aram eters (Villet

and M itchell, 1 9 8 1 ) ............................................................................................... 512.18 Com parison Between the M easured and Predicted Param eters (Acar,

1987; P uppaia e t ah, 1993) 522.19 Cavity Used in the Cavity E x p a n s io n ............................................................ 542.20 Com parison Between Theoretical and M easured Friction R esistances . 582.21 Schm ertm ann’s M ethod for E stim ating th e Friction Angle .................. 592.22 Evaluation of Constrained M odulus (Robertson and Cam panella, 1984) 602.23 Evaluation of C onstrained M odulus (Jamiolkowski e t ah, 1988)) . . . 61


2.24 Evaluation of Small S train Shear M odulus (Robertson and Cam panella,1984; Baldi e t al., 1981) .................................................................................... 62

2.25 Definition of Steady S tate P aram eter (Been et al., 1986) ...................... 642.26 Norm alized Cone Resistance Versus S ta te P aram eter for M onterey No

0 Sand (Been et al., 1 9 8 6 ) ................................................................................. 652.27 P roperties of Sands Versus S ta te Param eter (Been et al., 1986) . . . . 66

3.1 Schem atic of the P luviation S e tu p ................................................................... 713.2 Schem atic of the C alibration Cham ber (de Lima and Tumay, 1992) . 753.3 Flow C hart for D a ta Acquisition and M onitoring S y s t e m ...................... 823.4 G rain Size D istribution of M onterey Sand .................................................. 853.5 R elative Density Versus D epth in Cem ented S p e c im e n s .............................105

4.1 U ndrained Triaxial T est on U ncem ented M onterey No.0/30 Sasd»(-Dr =4 5 % ) ...........................................................................................................................112

4.2 U ndrained Triaxial Test on U ncem ented M onterey No.0/30 Sand [Dr =65 %) ...........................................................................................................................113

4.3 U ndrained Triaxial Test on U ncem ented M onterey No.0/30 Sand {Dr =8 5 % ) .......................................................................................................................... 114

4.4 U ndrained Triaxial Test on Cem ented M onterey No.0/30 Sand {Dr =45 and C.C. 1 % ) ......................................................................................................115



4.7 Stress P aths From Triaxial Tests on Uncem ented M onterey No.0/30Sand (Dr = 85 %) .................................................................................................. 118

4.8 Stress P a th s From Triaxial Tests on Cem ented M onterey No.0/30Sand {Dr = 85 %; C.C. = 1 % ) ..........................................................................119

4.9 Stress P aths From Triaxial Tests on Cem ented M onterey No.0/30Sand (Dr = 85 %; C.C . = 2 % ) ................................................................... 120

4.10 Stress Paths of Cem ented and Uncem ented S a n d .................................. 1214.11 Steady S ta te Line D iagram s for Cem ented and Uncem ented Sands . . 1264.12 Steady S ta te Line D iagram s for Cem ented and Uncem ented Sands . . 1274.13 Com parisons of SSL ............................................................................................. 1284.14 Assumed C onstitu tive Equations and Related Soil P aram eters (A dopted

from Ju ran and Beech, 1 9 8 6 )............................................................................. 1334.15 Com parisons Between Predicted and Experim ental Drained Triaxial

T e s t s .............................................................................................................................. 1354.16 Com parisons Between Predicted and Experim ental D rained Triaxial



Tests (Volumetric S t r a i n s ) .................................................................................... 138

xi


4.19 Com parisons Between Predicted and Experim ental D rained Triaxial Tests (Volum etric S t r a i n s ) .....................................................................................139

4.20 Influence of G/i7o on Stress R a t i o ....................................................................... 1404.21 Com parisons Between Predicted and Experim ental U ndrained Triaxial

T e s t s .............................................................................................................................. 1414.22 Com parisons Between Predicted and Experim ental U ndrained Triaxial

T e s t s .............................................................................................................................. 1424.23 D ilational Angles of Uncem ented and Cem ented S a n d ...............................1444.24 Influence of Curing Tim e on Unconflned Com pression S trength of 2 %

Cem ented Specimen .............................................................................................. 1454.25 Influence of Curing Tim e on Unconflned Com pression S trength of 1 %

Cem ented Specimen .............................................................................................. 1464.26 d q /d t versus T im e on 1 % Cem ented S p e c im e n s ...........................................147

5.1 Com parison of Ko values w ith Ja k y ’s re la tio n sh ip s .......................................1535.2 Cone P ene tra tion Test Results on a Specimen [Dr — 71.9 %; e-max —

0.85; tmin = 0.56 and 100 kPa) ....................................................................1545.3 Cone P ene tra tion Test Results on a Specim en [Dr = 68.7 %; tmax =

0.85; e-min = 0.56 and 200 kPa) ....................................................................1555.4 Cone P ene tra tion Test Results on a Specimen [Dr = 71.1 %; tmax —

0.85; Cmin = 0.56 and 300 kPa) .......................................................................... 1565.5 Pore Pressures D uring Piezocone P enetra tion of C em ented Specimen

(C.C. 2 % ) ....................................................................................................................1585.6 Com parison of Ko Values w ith Ja k y ’s R elationships (C .C. 1 %) . . . 1595.7 Com parison of Ko Values with Ja k y ’s Relationships (C .C . 2 % ) . . . 1605.8 Cone P ene tra tion Test Results on a Specimen [Dr = 68.4 %; e-max ~

0.85; = 0.56; C.C. 1 % and 100 k P a ) ............................................... 1625.9 Cone P enetra tion Test Results on a Specimen [Dr = 66.4 %; emax =

0.85; emin — 0.56; C.C. 1 % and 200 k P a ) ...............................................1635.10 Cone P enetra tion Test Results on a Specimen [Dr = 70.2 %; emax =

0.85; emin = 0.56; C.C. 1 % and 300 k P a ) ...............................................1645.11 Cone P enetra tion Test Results on a Specimen [Dr = 69.6 %; emax —

0.85; emin = 0.56; C.C. 2 % and 100 kPa) ..................................................... 1655.12 Cone P enetra tion Test Results on a Specimen [Dr — 69.2 %; emax —

0.85; emin = 0.56; C.C. 2 % and 200 kPa) ..................................................... 1665.13 Cone P enetra tion Test Results on a Specimen [Dr = 72.1 %; emax —

0.85; emin = 0.56; C.C. 2 % and 300 kPa) ..................................................... 1675.14 R epeatab ility of the R e s u l t s .................................................................................. 1695.15 T ip R esistance vs Effective Vertical Stress for Various Relative D ensitiesl725.16 Friction Resistance vs Effective Vertical Stress for Various Relative

D e n s itie s ....................................................................................................................... 1735.17 Influence of Location on Test R e s u l t s .................................................................1745.18 Influence of Relative Density on T ip Resistance of Cem ented Specimen

(1 % ) .............................................................................................................................. 176

xii


5.19 Influence of Relative Density on Friction Resistance of Cem ented Specimen (1 %) 177

5.20 Influence of Cem entation on T ip R e s i s t a n c e ................................................ 1785.21 Influence of C em entation on Friction R e s is ta n c e ......................................... 179

6.1 Com parisons w ith V and M ’s R e s u l t s .............................................................. 1836.2 Com parisons w ith B ald i’s C h a r t .........................................................................1856.3 Com parisons w ith Schm ertm ann’s C h a r t .......................................................1866.4 T ip R esistance Versus H orizontal Effective Stress (P resent Study) . . 1906.5 N orm alized T ip R esistance Versus Norm alized Effective O ctahedral

Stresses ....................................................................................................................... 1936.6 N orm alized T ip R esistance Versus Normalized Effective O ctahedral

Stresses ....................................................................................................................... 1946.7 Norm alized T ip R esistance Versus Norm alized Effective O ctahedral

Stresses ....................................................................................................................... 1956.8 N orm alized T ip R esistance Versus Normalized Effective O ctahedral

Stresses ....................................................................................................................... 1966.9 Influence of D iam eter R atio on a for Various D e n s i t i e s ........................... 1976.10 Grainsize D istributions of the Crushed Sand Around the Cone . . . . 1996.11 Influence of Crushing on Tip Resistance .......................................................2016.12 Boundary Condition Influence on CC R esults (Parkin, 1 9 8 8 ) ............. 2036.13 B oundary Condition Influence on CC R esults ............................................ 2046.14 Influence of Boundary Condition on Cone Test R e s u l t s ...........................2056.15 Influence of Cone D iam eter to dgo ratio on n ................................................2076.16 Norm alized Results on Cem ented Specimen (C.C. 1 p e r c e n t ) ...........2086.17 Norm alized Results on Cem ented Specimen (C.C. 2 p e r c e n t ) ...........2096.18 Influence of C em entation on a and n ..............................................................210

7.1 Com parison Between Theoretical and M easured T ip Resistances forC.C. 0 % Specimens (D &: M T h e o ry ) ............................................................... 214

7.2 Com parison Between Theoretical and M easured T ip Resistances forC.C. 1 % Specimens (D & M T h e o ry ) ............................................................... 215

7.3 Com parison Between Theoretical and M easured T ip Resistances forC.C. 2 % Specimens (D & M T h e o ry ) ............................................................... 216

7.4 D ilation Angles Versus Plastification Angles Used in J & S Theory . . 2187.5 Com parison Between Theoretical and M easured T ip Resistances for

C.C. 0 % Specimens (J & S Theory) ............................................................... 2197.6 Com parison Between Theoretical and M easured T ip Resistances for

C.C. 1 % Specimens (J & S Theory) ............................................................... 2207.7 Com parison Between Theoretical and M easured T ip Resistances for

C.C. 2 % Specimens (J & S Theory) ............................................................... 2217.8 R atio of T ip Resistance to L im iting Pressure (C.C. 0 % ) ..........................2247.9 R atio of T ip Resistance to Lim iting Pressure (C.C. 1 % ) ..........................2257.10 R atio of T ip Resistance to Lim iting Pressure (C.C. 2 % ) ..........................2267.11 Com parison of R a t io s ............................................................................................. 228

xiii


7.12 K j Ko Versus Relative D e n s i ty ............................................................................2357.13 Influence of D ilation Angle o n / d / i t ' o ..............................................................2367.14 T he Norm alized T ip Resistance Versus Friction R atio for Various ^

V a lu e s ...........................................................................................................................2387.15 A C hart for E stim ating Cohesion Intercept and Relative Density . . . 2407.16 A C hart for E stim ating Cohesion Intercept and R elative Density . . . 2417.17 R elative Density Versus Friction A n g le s .......................................................... 2427.18 a Versus Relative D ensity for Various D iam eter R a t i o s .............................. 2437.19 a Versus Relative Density for Various Cem ent C o n te n ts ...............................2457.20 Norm alized Plots of D ata to D eterm ine and n„ (P resen t Test Re

sults: C.C. 0 and 2 % ) ........................................................................................... 2477.21 /? and n i from Present Test R e s u l t s .................................................... 2487.22 /3 and n i from Present Test R e s u l t s .................................................... 2497.23 T he ç / Versus Relative Density for C.C. 1 and 2 % ......................................2527.24 C hart for E stim ating Relative Density {qj = 0 kPa) (A pproach 1) . . 2537.25 C hart for Estim ating Relative Density {q/ = 20 kPa) (Approach 1) . 2547.26 C hart for E stim ating Relative Density [qj = 40 kPa) (A pproach 1) . 2557.27 C hart for E stim ating Dr and qj for = 1. (A pproach 2 ) ...................... 2567.28 C hart for E stim ating Dr and qj for = 2. (A pproach 2 ) ...................... 2577.29 C hart for E stim ating Dr and qj for — 3. (A pproach 2 ) ...................... 2587.30 Classification C hart for E stim ating Cem ented D e p o s i t s ...........................2607.31 C hart for E stim ating for Uncem ented Sand ..............................................2627.32 C hart for E stim ating ^ for Cem ented S a n d s ................................................. 2637.33 '0 Versus Cohesion and Friction A n g le s .......................................................... 2647.34 Influence of C em entation (1 %) and Confining Pressure on Tip Resis

tance ..............................................................................................................................2667.35 Influence of Cem entation (2 %) and Confining Pressure on T ip Resis

tance ..............................................................................................................................2677.36 Classification C hart (Schm ertm ann, 1978) .................................................. 2697.37 Classification C hart (Douglas and Olsen, 1 9 8 1 ).......................................... 2707.38 Classification C hart (Tumay, 1 9 8 5 ) ............................................................... 2717.39 Classification C hart (Robertson and C am panella, 1 9 8 5 ) .......................... 272

A .l T he Variation of Gmax Versus Confining Stress for C em ented Specimens300A.2 T he V ariation of Gmax Versus Confining Stress for Cem ented Specim ens301

B .l U ndrained Triaxial Test on Cem ented M onterey No. 0 /30 Sand {Dr =50 % ; C.C. 2 % ) ......................................................................................................309

B.2 U ndrained Triaxial Test on Cem ented M onterey No. 0/30 Sand {Dr =65 %; C.C. 2 % ) ......................................................................................................... 310

B.3 U ndrained Triaxial Test on Cem ented M onterey No. 0 /30 Sand {Dr —85 9%; C.C. 2 % ) ..........................................................................................................311

B.4 Stress P aths of CU Test {Dr = 45 %; C.C. 0 % ) ....... 312B.5 Stress P aths of CU Test {Dr = 45 %; C.C. 1 % ) ....... 313B.6 Stress Paths of CU Test {Dr = 45 %; C.C. 2 % ) ....... 314

xiv


B.7 Stress P a th s of CU Test {Dr — 65 %; C.C. 0 % ) ........................................... 315B.8 Stress P a th s of CU Test {By = 65 %; C.C. 1 % ) ........................................... 316B.9 Stress P a th s of CU Test {By = 65 %; C.C. 2 % ) ...........................................317B.IO Flow C hart of Program 1 ........................................................................................318B . l l Flow C hart of Program 2 ........................................................................................322B.12 Flow C hart of Program 3 ........................................................................................325

C .l Cone P enetra tion Test R esults on a Specim en {By = 48.7 %; Cmax =0.85; Synin. = 0.56 and 100 kPa) ..........................................................................330

C.2 Cone Penetration Test R esults on a Specimen {By — 56.4 %; eâx =0.85; Emin — 0.56 and 200 kPa) ..........................................................................331

C.3 Cone P enetration Test R esults on a Specimen {By = 54.8 %; tmax =0.85; Cynin = 0.56 and 300 kPa) ..........................................................................332

C.4 Cone P enetra tion Test R esults on a Specimen {By = 90.0 %; emm =0.85; Eyniyi = 0.56 and 100 kPa) ..........................................................................333

C.5 Cone P enetration Test R esults on a Specim en {By — 90.0 %; =0.85; = 0.56 and 200 kPa) ..........................................................................334

C.6 Cone P enetration Test R esults on a Specim en {By = 88.0 %; Cjnax =0.85; Cmtn = 0.56 and 300 kPa) ........................................................................... 335

C.7 Cone P enetration Test R esults on a Specim en {By = 86.0 %; Cmax —0.85; Cynin = 0.56 and 100 kPa) ........................................................................... 336

C.8 Cone P enetration Test R esults on a Specim en {By = 84.0 %; Cmax =0.85; Cmin = 0.56 and 100 kPa) ........................................................................... 337

C.9 Cone Penetration Test R esults on a Specimen (By = 86.0 %; C.C. — 1 %;emax — 0.85; = 0.56 and 100 k P a ) ............................................................338

C.IO Cone P enetration Test R esults on a Specimen {By = 84.6 %; C.C. = 1 %;emax — 0.85; em,„ — 0.56 and 200 k P a ) ....................................... 339

C . l l Cone P enetration Test R esults on a Specimen {By = 89.2 %; C.C. = 1 %;eynax = 0.85; Cmin = 0.56 and 300 k P a ) ....................................... 340

C.12 Cone P enetration Test Results on a Specimen {By = 48.8 %; C.C. — 1%]emax — 0.85; em,„ = 0.56 and 100 k P a ) ........................................341

C.13 Cone P enetration Test Results on a Specimen {By = 46.6 %; C.C. = 1 %;eyyiax = 0.85; = 0.56 and 200 k P a ) ............................................................ 342

C.14 Cone P enetration Test Results on a Specimen {By = 53.2 %; C.C. = 1 %;emax = 0.85; tmin ~ 0.56 and 300 k P a ) ....................................... 343

C.15 Cone P enetration Test R esults on a Specimen {By = 88.2 %; C.C. = 2 %;emax = 0.85; Cmin = 0.56 and 100 k P a ) ....................................... 344

C.16 Cone Penetration Test R esults on a Specimen {By = 86.3 %; C.C. = 2 %;emax = 0.85; Emin = 0.56 and 200 k P a ) ....................................... 345

C.17 Cone P enetration Test R esults on a Specimen {By = 84.2 %; C.C. = 2 %;emax = 0.85; = 0.56 and 300 k P a ) ............................................................ 346

C.18 Cone Penetration Test R esults on a Specim en {By — 47.2 %; C.C. = 2 %;emax = 0.85; Emin = 0.56 and 100 k P a ) ......................... 347

C.19 Cone P enetration Test R esults on a Specim en {By = 54.4 %; C.C. — 2%;emax = 0.85; Emin = 0.56 and 200 k P a ) ....................................... 348

XV


C.20 Cone P enetra tion Test Results on a Specimen {Dr — 52.0 %; C.C. = 2 %; emax = 0.85; Cmm = 0.56 and 300 k P a ) ............................................................349

D .l Flow chart for Cavity Expansion Model (Procedure 2 ) ...............................356D.2 R atios ( ^ ) Versus Effective Vertical Stresses (C.C. 0 % ) ........................ 357D.3 R atios ( ^ ) Versus Effective Vertical Stresses (C.C. 1 % ) ........................358D.4 R atios ( ^ ) Versus Effective Vertical Stresses (C.C. 2 % ) ........................359D.5 A pproach 1 in Em pirical M ethod {qj = 10 k P a ) ..................... 360D.6 A pproach 1 in Em pirical M ethod {qj = 30 k P a ) ..................... 361D.7 A pproach 1 in Em pirical M ethod {qj = 50 k P a ) ..................... 362

XVI


List of Plates

1.1 A Vertical Slope in Cem ented Loess Deposits Along Natchez Trail,Natchez, M ississippi ........................................................................................... 2

1.2 Slope Failure Along a W eakly Cem ented Deposit in California due tothe Lom a P rie ta E arthquake (Photograph by Dr. Wayne Clough) . . 4

3.1 Photograph of the Saturation S e tu p ..................................................... 743.2 Photograph of the M iniature Cone (de Lima, 1 9 9 0 ) ....................... 783.3 P hotograph of the Controls on th e Panel B o a r d .............................. 793.4 Applying Vacuum Inside the S p e c im e n ......................................................... 923.5 Specim en A fter Unfolding the Split M o l d s ................................................ 933.6 F inal Assembly of the S p e c im e n ..................................................................... 943.7 Plexiglas M olds for Triaxial S p e c im e n s ........................................................... 1003.8 Scanning Electron M icrographs of Uncem ented and Cem ented (0 and

2 %) s a n d ................................................................................................................... 1023.9 Specimen Undergoing th e Saturation P ro c e s s .................................................1043.10 Scanning E lectron M icrographs of Cem ented Specimen at 10 and 25 cm

D e p th s ...........................................................................................................................1063.11 Scanning Electron M icrographs of Cem ented Specimens at 40 and

50 cm D e p t h s .............................................................................................................107

xvn


Abstract

An understanding of the effect of cementation on geotechnical properties of soil

deposits is gaining increasing atten tion in the profession. When low levels of cemen

tation in sands are neglected, pile capacity and slope stability are underestimated

and liquefaction is overestimated. The recent Loma, P r ie ta earthquake led to fail

ures along the northern Daly City bluffs causing catastrophic failures in residences

scattered on these slopes. These failures were not anticipated, possibly due to (he

confidence felt in constructing on bluffs of cemented deposits. It is essential to devise

schemes to identify cementation in soil investigations and develop methods in eval

uating engineering characteristics of cemented deposits. The objective of this study

is to develop a method to identify cementation in sands and assess the engineering

characteristics of cemented sand deposits using the cone penetration testing scheme.

The scope of the study includes evaluation of the effect of cementation on cone

penetration testing (experimental model) and comparison of these ex])erimental re

sults with theoretical models of penetration mechanism in cemented sands. Existing

models based upon the bearing capacity theories and cavity expansion models are

utilized in theoretical modeling. A constitutive model is developed for strength-

deformation behavior of cemented sands and is used in theoretical modeling.

.Artificially cemented Monterey No. 0/30 is used in the calibration chamber

study. A total of 30 tests are conducted at, three ranges of relative density ( 'h5-nü,

65-75 and above 85 %), three confining pressures (100, 200 and 300 kPa) and three

different cement content (0, I and 2 %). Pluviation method is used for specimen

preparation. Specimens are cured for 7 days, transferred into the flexible wall cal

ibration chamber and then consolidated under R'o conditions. Penetration testing

was conducted with a 1.27c??? diam eter m iniature cone which reduced the chamber

size effects on the results significantly. Separate drained triaxial tests provided the

necessary parameters for strength-deformation modeling of cemented sands.

xviii


The experimental model results coupled with the theoretical model predictions

provide a semi-empirical and empirical schemes for evaluating engineering character

istics of cemented sand deposits. An assessment of the applicability of these models

in prediction of cementation in such deposits is also provided. The results indicate

th a t tip resistance and sleeve friction in cone penetration testing provide a reason

able assessment of cementation. The charts and the analysis method provided can be

used to estim ate the engineering characteristics of such deposits. It is found essential

th a t reliability and accuracy of the proposed methods of analysis be evaluated by

insitu tests in a well-documented, naturally cemented sand deposit.

X I X


Chapter 1

INTRODUCTION

1.1 Introduction

N atura lly cem ented deposits are very common throughout California, Texas,

along the banks of the Mississippi (loess deposits in the Vicksburg area), India,

C anada and th e world. These deposits are often characterized by the ir ability to

w ith stand steep na tu ra l slopes (Clough et ah, 1981). Large deposits of cemented

sands are found along the California coast and are typically 60 degrees or steeper

from th e horizontal and reach heights of 100 m (Clough et ah, 1981). Similar deposits

can be found near Natchez and Vicksburg area in Mississippi. P la te 1.1 shows the

near vertical slopes in a cem ented deposit along the Natchez Trail way near Natchez,

M ississippi. Recent da ta and evidence suggest th a t even the cleanest sand deposits

are na tu ra lly cem ented and therefore engineering judgm ents m ade from specimens

reconstitu ted in the laboratory m ay not be valid (M itchell and Solymar, 1984). This

cem entation is generally provided by agents such as silica, hydrous silicates, hydrous

iron oxides, and carbonates deposited a t the point of contact between sand particles

(Clough et ah, 1981). In some cases, the cem entation is due to welding at the

contacts or tim e dependent streng th gain (Mitchell and Solymar, 1984). This type

of cem entation generally results in low to m oderate cem entation.

An understanding of the effect of th is low to m oderate degrees of cem entation

on the sta tic and dynam ic streng th and deform ation behavior of sands is becom

ing increasingly im portan t in design and analysis in geotechnical engineering. A t the

present s ta te of the a rt, the effect of cem entation on strength-deform ation behavior of

sands is neglected in the design since cem entation often improves the streng th prop

erties. However, recent studies indicate th a t neglecting cem entation, particularly

th e sm aller degree of cem entation bonds results in overestim ation of the liquefaction

resistance, underestim ation of the strength of the soil deposits and also underestim a

tion of th e stab ility of the slopes (Rad and Clough, 1982; Poulos, 1980; Frydm an et

ah , 1980). Slope failures are very common in such cem ented deposits and they lead


É

.' -.'' V ' /

—

Pla ie 1.1: A Vertical Slope in Cemented Loess Deposits Along Natchez Trail,

Natchez, Mississippi


to loss of life and property (Clough e t al., 1981). Recent Lom a P rie ta San Francisco

earthquake led to failures along the cem ented northern daly city bluffs (S itar, 1990)

(P la te 1.2). Liquefaction phenom enon was also observed a t several sites during the

Lom a P rie ta earthquake.

C em entation is also a popular m ethod used for soil stabilization. W hen soils

w ith unsatisfactory engineering properties are encountered, some m ethod of s tab i

lization by using cement, fly ash and other additives is needed prior to any fu tu re

construction on th a t soil. This procedure is frequently used in m any engineering

projects like im provem ent of subgrades and airport runways, stabilizing the slopes

and em bankm ents and also increasing the bearing capacity of th e soil using cem ented

sand columns. T he loads imposed by traffic have to be transferred to soil layers ca

pable of supporting them w ithout shear failure or excessive deform ations. In brief,

cem ent stabilization improves m odulus of elasticity, shear m odulus, coefficient of

subgrade reaction and unconfined compression.

The difficulty in sam pling natu ral or stabilized cem ented deposits prom pts the

need to use insitu testing schemes (Beckwith and Hansen, 1981; Bachus e t al., 1981;

Frydm an et al., 1980). Among several different insitu testing m ethods, cone pene

tra tion testing is gaining wide acceptance and use in the USA and the world due to

its repeatability , economy and capability to provide accurate, repeatab le vertical soil

profiles and pertinen t engineering param eters related to the sounded deposits.

Cone penetration tests in granular deposits are currently used to m easure tip

resistance Çc, sleeve friction fs, and if piezocone is used, the to ta l pore pressure Ut

along the tip a n d /o r the shaft of the penetrom eter. Estim ates of relative density

of the uncem ented granular deposits are m ade using charts obtained in calibration

cham bers. In ternal friction angles are obtained indirectly by correlating the tip

resistance w ith the relative density. C harts are updated for the influence of o ther

variables such as overconsolidation by conducting tests in a calibration cham ber

(Schm ertm ann, 1977; Villet and M itchell, 1981; Baldi et al., 1981; Jamiolokowski,

1985).

The influence of cem entation on penetration resistance is yet to be investigated.

Prelim inary studies investigating the possible effects of low degrees of cem entation in

cone penetration indicated th a t cem entation increases the tip and friction resistances

and decreases the friction ratio (Rad and Tumay, 1986; Akili and Nabil, 1988). These


Plate 1.2: Slope Failure Along a Weakly Cemented Deposit in Caliibrnia due to the

Loma P rie ta Flarthquake (Pliotograph by Dr. Wayne Clough)


studies clearly showed th a t estim ates of engineering param eters of cem ented deposits

using the available m ethods which are developed for clean sands, would be invalid.

Hence, it is proposed to in itia te a testing scheme involving cem ented specimen

testing in calibration cham bers. T he m echanical behavior of cem ented sand deposits

is stud ied by using artificially cem ented sands in the laboratory. Previous studies

ind icate th a t artificially cem ented sand deposits sim ulate the behavior of natu ral

deposits. Hence, artificially cem ented specim ens are used in the tests. Results from

these tests are used to provide a classification scheme for cem ented deposits and also

to provide a m ethodology for estim ating the streng th properties.

1.2 O bjectives

T he proposed study aims to develop the scheme and m ethodology to determ ine

th e streng th param eters of artificially cem ented sands by cone penetration testing

conducted in a calibration chamber.

T he objectives of th is work are:

1. to perform a lite ra tu re review about strength-deform ation behavior of artifi

cially cem ented sands,

2. to perform laboratory triax ial and unconfined compression tests on artificial

cem ented specim ens for reassessing the available d a ta and evaluating the nec

essary strength-deform ation param eters for bo th modeling and calibration pu r

poses,

3. to conduct calibration cham ber tests on artificially cem ented and uncem ented

specimens using a m iniature cone penetrom eter and to evaluate these tes t re

sults in the study of various variables like cham ber size, boundary conditions,

sand compressibility, size and shape of the aggregates and cem entation,

4. to sim ulate the cone penetration w ith existing bearing capacity and cavity

expansion m odels and then develop a m ethodology to evaluate the strength

param eters in light of comparisons of the theoretical predictions w ith experi

m ental results.


1.3 O rganization o f th e M anuscript and Sum m ary

T he following presents a brief sum m ary of the contents of various chapters;

C hap ter 2 covers the lite ra tu re review pertain ing to cem ented sands and their

behavioral studies conducted by various investigators, cone pene tration testing and

the ir applications, calibration chambers and various m ethods of analyses used in cone

pene tra tion testing. Definitions, various factors causing cem entation and cem ented

soil s tru c tu re are presented. S tatic and dynam ic behavior of cem ented sands and

the ir findings are sum m arized. In another section, cone penetration testing history

is briefly reviewed. T his is followed by a discussion on calibration cham bers and

th e ir applications using cone penetrom eter. Studies involving cem ented specim ens

in rigid cham bers are presented along with various studies on uncem ented sands.

Different m ethods used in the analysis of cone penetration testing, the ir advantages

and disadvantages are also discussed.

C hap ter 3 presents the m ethodology for various tests used in the present in

vestigation. Equipm ent and their use are also briefly described. Cem ented and

uncem ented specimen preparation procedures are docum ented. Large scale speci

m en p reparation for calibration chambers is then presented. S atu ra tion procedures

used in these specimens are explained followed by the testing procedures adopted in

th e tests.

C hapter 4 covers the results of undrained, drained triax ial and unconfined com

pression tests. Total and effective stress param eters are evaluated and com pared

w ith drained param eters. Critical s ta te lines for both cem ented and uncem ented

M onterey No. 0/30 sand are presented. Curing period influence on unconfined com

pression streng th is evaluated and discussed. Juran-G uerm azi m odel is updated for

th e effect of cem entation and is used to model the drained and undrained triax ial

te s t results. These m odeling param eters are later used in cavity expansion m odeling.

C hapter 5 sum m arizes the cone penetration testing conducted on both cem ented

and uncem ented specimens in the calibration cham ber. Specim ens are first consol

idated under Ko conditions. Cone penetration tests are then conducted under zero

la tera l stra in boundary conditions (Traditional BC 3). These results are first as

sessed for repeatability , precision and accuracy. Influence of relative density, cement

content and confining pressures on tip and friction resistances are also evaluated.


C hapter 6 presents a sum m ary of various factors affecting the calibration cham

ber testing. Influence of boundary conditions, cham ber size effects, grain size and

shape, com pressibility and crushability of the sands tested and cem entation are in

vestigated. Several cham ber investigations on various sands are used in the above

analysis.

C hapter 7 compares the predictions of theoretical models v/ith experim ental re

sults. T ip and friction resistances are first predicted using two rigid-plastic bearing

capacity m odels and two cavity expansion m odel. The elasto-plastic m odel devel

oped in C hap ter 4 is th en used in an increm ental cavity expansion analysis of the

problem . T he theoretical predictions are com pared w ith the experim ental results.

Based upon th is evaluation, a m ethodology is developed for evaluating the cohesion

in tercept and relative density from cone penetration testing in cem ented sands. E m

pirical correlations are developed for estim ating the relative density and unconfined

compression strength . Two procedures are introduced. T he first procedure is based

on param eters obtained by normalizing the tip resistance and effective stress and the

second procedure is based on the steady sta te line concept. The influence of various

factors such as boundary conditions, cham ber size, sand grain size and shape, com

pressibility of the sands and cem entation on cone test results in cham bers are also

evaluated.

C hapter 8 sum m arizes the findings, conclusions and the shortcom ings of this

study. Recom m endations for fu tu re research are also provided.


Chapter 2

SYNTHESIS OF AVAILABLE INFORMATION

2.1 Introduction

M ost sands in na tu re are lowly cemented. C em entation has a significant in

fluence on their engineering properties (Schm ertm ann, 1991). In norm al practice,

geotechnical engineers do not account for cem entation in sands in design and analy

sis. However, neglecting the cem entation bonds will result in underestim ation of the

s treng th and the liquefaction resistance. The significance of studying the behavior of

cem ented sands under dynam ic loading gains more im portance due to slope failures

in cem ented deposits during the recent earthquakes in the San Francisco Bay area.

T he difficulty in sampling and also non availability of samples of different density

and cem entation levels as needed to conduct comprehensive studies led m any inves

tigators to use artificially cemented specimens. Past studies conducted by Clough et

al. (1981) and Rad and Clough (1982) dem onstrated th a t 1 to 2 % artificially ce

m entation using Portland cement will sim ulate the naturally cem ented sands. These

studies on artificially cemented sands have also been beneficial in evaluating the

feasibility of improving subgrades under highways and runways, stabilizing slopes

in em bankm ents and cuts and also in improving the bearing capacity by adopting

cem ent stabilization.

In the present s ta te of the a rt, there is no insitu m ethod to identify and de

term ine the engineering characteristics of naturally cem ented deposits. T he present

investigation aim s a t the above need by using cone penetration testing. Artificially

cem ented specimens are prepared and tested in calibration cham bers. These results

are used in preparing a prediction scheme to identify natu ra lly cem ented sands and

also to determ ine the engineering characteristics.

In th is chapter, a review of the past studies on sta tic and dynam ic tests con

ducted on artificially and naturally cem ented sands is presented. T he engineering

characteristics of cemented M onterey No. 0/30 and No. 0 sand are discussed. The

presented s ta tic and dynam ic test results are used in the la ter chapters in analysis

8


9

of the results and in preparing semi-em pirical relationships. The history and appli

cations of cone penetrom eter and calibration cham bers are then reviewed. In section

2.4, th e lim ited d a ta available on cone penetration testing of cem ented sands under

low confining stresses are reviewed. T he last section covers th e different m ethods of

analysis used in the in te rp re ta tion of cone penetration results. The advantages and

disadvantages of each m ethod are also discussed.

2.2 E ngineering B ehavior o f C em ented Sands

Sand deposits may undergo a significant loss of strength due to sam ple d istu r

bance, thus behaving sim ilar to sensitive clays (M itchell and Solymar, 1984). Recent

evidences show th a t freshly deposited or densified saturated clean sands m ay exhibit

tim e-dependent stiffening and streng th gain. These phenom ena appear to be due to

th e cem entation a t in terpartic le contact points (Mitchell and Solymar, 1984). It is

necessary to consider the effect of th is low level cem entation when evaluating the

resu lts of laboratory tests on reconstitu ted sam ples, in assessment of ground m od

ification using deep densification, in evaluation and in terpreta tion of relative den

sity m easurem ents, and in estim ation of liquefaction potential. T he m agnitude of

increase in cone penetration resistance a t the m ain dam foundation at Jebba H ydro

electric Development in Nigeria, over a period of several weeks to m onths following

deep densification, clearly suggested the tim e dependent strength gain in the sandy

deposits. Similar observations were noted by D urante and Voronkevich in the hy-

draulically placed em bankm ents (M itchell and Solymar, 1984). The tim e dependent

s treng th gain was sufficient enough to satisfy design criteria, however sam pling or

cone testing im m ediately after construction rendered different conclusions. Hence,

i t is im portan t to understand the effect of cem entation on engineering behavior and

th e possible m echanism s th a t cause the phenomenon in sands. The next few sec

tions cover these aspects and also review the fundam entals of different cem entation

processes encountered in deposits o ther than sands.

2.2.1 Definitions and Physio-Chemical Characteristics

C em ented soils are defined as soils composed of sand or gravel sized particles

or fragm ents of rocks bonded together by a cementing agent to form a larger com


10

posite s truc tu re w ith distinctive geological and geotechnical properties (A l-G hanem ,

1989). This bonding is due to the cem entation by physical or chemical processes or

a com bination of both. T he processes will generally take place over a period of tim e.

T he tim e dependent strength gain in sands a t a test site densified by vibro

com paction and blasting is a ttr ib u ted to different mechanisms (M itchell and Solymar,

1984). The first m echanism discussed is developm ent of excess pore pressures and

their dissipation (M itchell and Solymar, 1984). This m ay be tru e in case of cohesive

clays, bu t in sands, the dissipation takes place ra th e r instantaneously. Hence, it can

not be considered as a process leading to strength gain in sands.

T he second m echanism is th a t explosive gases after b lasting m ay cause the

s treng th increase as a resu lt of an increase in sand compressibility. However, th is

m echanism could not be accounted for strength gain over a tim e period of weeks

to m onths (M itchell and Solymar, 1984). The th ird m echanism is the th ixotropic

s treng th gain. This phenom enon is evident in case of fine grained soils; however, in

sandy soils, th e ex ten t of strength loss on disturbance followed by streng th gain a t

rest is not well established.

M itchell and Solymar (1984) offers the m ost probable cause or m echanism as

form ation of silica acid gel films on particle surfaces and precip itation of silica or

o ther m aterial from solution or suspension as a cem entation species a t particle con

tac t points. They propose th a t th is gel adheres to the surface in a th in layer and

has cem enting properties. The dissolution and precipitation of silica in th e form of

am orphous silica and crystalline quartz may lead to cem entation (M itchell and Soly

m ar, 1984). Equilibrium relationships are difficult to obtain w ith silica. Equilibrium ,

if reached a t all, will require weeks to m onths, i.e., tim e periods consistent with the

observed streng th increase in the field.

T he o ther processes th a t lead to cem entation are the presence of m etallic ions

such as A1 and Fe. M itchell and Solymar (1984) propose form ation of crystalline iron

oxide coatings which m ay cause cem entation. Pressure solution due to high stress a t

grain contact points also leads to preferential solution (M itchell and Solymar, 1984).

T he liberated Si02 supersaturates the pore w ater so th a t some Si02 m ay precip ita te

cLS quartz over growths and causes in terpenetration of grains (M itchell and Solymar,

1984).


11

All the above m echanism s depend on sand particle surface characteristics, th e

am ount and type of cem enting agent, the ground w ater movement and the ex

ten t/a m o u n t of w eathering. The common cem enting agents th a t are na tu ra lly found

are; silica, iron oxides, carbonates in m arine environm ents, clay and silt (R ad, 1984;

Al-Ghanem , 1989).

2.2.1.1 Cementation in Sands

Cem ented sands cover extensive areas in the U.S. (California, Texas, M ississippi,

Arizona, N orth Louisiana) and also m any regions like the M iddle East, South E ast

Asia and Africa. They are form ed in arid and semi-arid environm ents. T he streng th

properties of these soils are different from those generally observed in geotechnical

engineering. This difference is due to th e presence of chemical agents like hydrous

silicates, iron oxides, calcium carbonate, calcium sulphate and sometim es the clay

particles which bond adjacent particles in the soils. This bond which describes the

coherence between the particles is known as cementation and the chemical agents

are known as cementing agents.

T he three reported causes of cem entation in sands are:

1. T he welding betw een the particles a t their contact points due to the in ternal

heat a t the tim e of deposition or due to prolonged pressure a t prom inent points

of contact between grains (Lee, 1975, M itchell and Solymar, 1984),

2. T he presence of cem enting agents like silica or siliceous cem ent, calcium carbon

a te or calcareous cem ent, clay or argillaceous cem ent and iron bearing m inerals

or ferruginous cem ent (Krynine and Judd, 1957, M itchell and Solym ar, 1984).

T he cem entation described by M itchell and Solymar (1984) is of th is kind.

3. In some cases, clay m ay also participate in the bonding. These bonds are weak

in strength and such bonding is generally encountered in loessial soils.

2.2.1.2 Cementation in Collapsing Soils

Collapsing soils are defined as soils which norm ally have some streng th bu t

experience a loss of volume upon loading, w etting or both (A l-C hanem , 1989). These

soils are found in m any parts of the world and can be formed in different environm ents


12

such as loessial, colluvial, alluvial, subaerial and aeolian (A l-G hanem , 1989). These

soils are porous in fabric and they are geologically young. Their s truc tu re consists

of sand grains bonded in loose silty sand. T he fine fraction th a t exists in sm all gaps

betw een adjacent grains of the soils will undergo local compression which u ltim ately

bonds the larger grains.

Clay is another binding agent between sand and silt grains. Several structures

are form ed w ith clay as binding agent and these cem entation bonds can be destroyed

w ith the addition of w ater (Al-Ghanem , 1989). Loessial soils are wind-borne, n a t

urally cem ented collapsing soils. Calcareous m aterials a n d /o r clay is generally the

binding agent. In these soils, th e bonds are weakened by either leaching ou t or

softening of the binders.

2.2.1.3 Cementation in Rocks

W hen the fragm ents of a rock are bonded firmly together w ith a cem enting

agent to form a new rock type, the resulting m aterial is classified as a cem ented rock

(A l-G hanem , 1989). This cem entation can take place either from the infiltration of

w ater carrying chemicals or the dissolution of certain m inerals in the m ass to form

new bonding m aterial (Al-Ghanem , 1989).

Sedim entary rocks are formed due to the consolidation and cem entation of sed

im ents. These are the end products of the weathering process. T he m ost common

cem ents found in these rocks are: silica or siliceous cem ent, calcium carbonate,

argillaceous cement and iron oxides. Limestones are another exam ple of cem ented

rocks. They are composed of calcium and m agnesium carbonates, and are found

in m arine deposits (Al-Ghanem , 1989). Cem ented rocks exhibit higher compressive

streng th when quartz is the cem enting m edium and lower streng th is obtained when

they are cem ented entirely or partially w ith clay.

2.2.1.4 Structure of Cemented Granular Soils

T he cem entation process depends on a num ber of factors including the type and

am ount of cem entation, the degree of packing, the density and characteristics of the

soil particles and the m ethod of deposition. Because of these factors, the cem ented


13

soil s truc tu re also varies. In fact, th e struc tu re , in some cases, explains the chemical

agent and the process in the cem entation th a t m ight have taken place.

Sowers and Sowers (1979) classified the s truc tu re in to two categories: m atrix

structure and skeletal structure. F igure 2.1 shows these structures. The m atrix

s tru c tu re develops when th e volum e of the bulky grains is less than abou t tw ice the

am ount of binder. In o ther words, m ost of the volume is occupied by th e binder and

th ere is little contact betw een the bulky grains. The streng th of these structu res

depends upon the streng th of either the binder or bulky grains, whichever is weaker.

T he skeletal s truc tu re develops when the volume of bulky grains is m ore than

tw ice the volum e of the binder. This s truc tu re can be subdivided in to e ither contact-

bond s tru c tu re or a void-bond structu re . In contact bond structu re , particles are

cem ented a t the contact points. This s truc tu re can be formed in soils w ith large

partic le sizes. The s tru c tu re is relatively rigid and incompressible. This bond is not

stab le and can be lost due to leaching by ground water. T he void-bond s tru c tu re is

due to the contacts am ong individual particles. T he voids are filled w ith binders such

as carbonates, iron oxides and silicates. Here, cem entation develops subsequently

after the s truc tu re forms. This s truc tu re is m ore stable th an contact bond structu re .

2.2.2 Displacement Rate Controlled Stress-Deformation Behavior (Static Behavior)

T he shearing resistance of artificially cem ented sands can be considered to be

composed of two elem ents, one of which is independent of the norm al stress on the

failure plane (cohesion in tercep t), and the o ther increasing w ith th e norm al effective

stress on the failure envelope (friction angle). The increase in cohesion in tercept

w ith the increase in cement content has been shown by several investigators (Rad

and Clough, 1982; A car and E l-tahir, 1986).

T he sta tic triaxial properties of cemented sands are presented in this section.

A to ta l of 43 drained triax ia l tests conducted on cem ented sands are reported by

R ad and Clough (1981). M onterey No. 0 sand, a commercially available washed and

sieved beach sand was used in their investigation. Artificial cem entation of 1 and 2

% was used. T he im portan t variables used in th is study were relative density (D r),

cem ent content (C.C.) and confining pressure (cTc). Three ranges of relative density


14

: Bin<Jer

Matrix Structure

•v.;; Btndar

Void-bond nructuie.

D«na« iv.üi; B inder

Looee B inder

Contaci-bond imicturr.

Skeletal Structure

Figure 2.1: Various S tructures in Cem ented Soils (Sowers and Sowers, 1979)


15

(20 - 30 %, 45 - 55 %, 70 - 85 %), and four ranges of confining pressures (35, 100, 200

and 300 kP a ) and th ree ranges of cement content (0 (uncem ented), 1 and 2 %) were

used. A 14 day curing period was adopted. The drained triaxial tests conducted on

0, 1 and 2 % cement content and 45 - 55 % relative density are shown in Figures

2.2, 2.3 and 2.4. These d a ta are used for obtaining m odulus param eters like Young’s

m odulus and shear m odulus.

T he increase in relative density increases the streng th of the specim en. In unce

m ented cohesionless sands, a rise in relative density results in the increase of friction

angle (Figure 2.5). Similar observations are m ade in cem ented sands. This increase

is due to th e increase in relative density which implies a densely packed soil m ass

w ith m ore contacts between the soil grains. However as depicted in Figure 2.5, the

increase in cem ent content a t a given relative density does not result in any significant

rise in friction angle.

The peak and residual strength param eters are reported in Table 2.1. Figure 2.6

dem onstrates th a t the increase in the relative density and cem entation of the speci

m en results in an increase in the cohesion in tercept. This phenomenon is a ttr ib u ted

to the fact th a t as relative density increases, the num ber of contacts betw een the

particles increase and consequently stronger bonds form (Acar and El-Tahir, 1986;

Riccobono, 1985). A nother im portan t observation is th a t even though the cem ent

conten t induces cohesion intercept at peak strains, this cohesion is alm ost zero a t

residual strains (Table 2.1). T he cem entation bonds will be destroyed at the peak

failure and the residual friction angles at failure are approxim ately the sam e as th a t

of uncem ented sand.

In brief, the m ain conclusions drawn from the sta tic loading response of cem ented

sands are sum m arized as follows.

1. Cem entation increases the peak strength . Cem entation results in a cohesion

in tercept due to the bonding between particles.

2. T he strength is m ainly due to cohesion a t low confining stresses and strains

and due to friction a t higher strains.

3. Failure is of b rittle nature.

4. Cem entation has m inor effect on the friction angle.


16

2000- Rad and Tumay (1986)

e e e e o 1 0 0 kPa &B&&Q 2 0 0 kPa

3 0 0 kPo1 5 0 0

w

M 1000

5 0 0

AXIAL STRAIN (%)

& 8.0z<a: 6.0t—(O 4.0ua: 2.0wz 0.0Z)_J -2 .0o> 4 8 12

AXIAL STRAIN (%)

Figure 2.2: Drained Triaxial Results on an Uncemented Specimen of M onterey No. 0

Sand at Relative Density 45-55 % (Rad and Tumay, 1986)


17


ooo&o w o kPa □oDB-o 200 kPa

300 kPa500

ÜJ

(/} 1000

500

AXIAL STRAIN (% )

4.0

tr2.0( / )

0.0LÜ

- 2.0A XIAL STRAIN (s?)

Figure 2.3: D rained Triaxial Results on a 1 % Cemented Specimen of M onterey No. 0

Sand a t Relative Density 45-55 % (Rad and Tumay, 1986)


18


o o o o o 1 0 0 k P a 0 0 0 0 -0 2 0 0 k P o

3 0 0 k P aS . 1500

{/)(DÜJOHœ 1000

OH

uj 500

A XIAL STRAIN {%)

8.0

I 6.0^ 4 .0

S 2.05 0.0^ - 2.0

A XIAL STRAIN {%)

Figure 2.4: Drained Triaxial Results on a 2 % Cemented Specimen of Monterey No. 0

Sand at Relative Density 45-55 % (Rad and Tumay, 1986)


19

ÜJLUCCOLUO

LU_ JCDZ<

Sh-U(r

B 50

î 2

CEM ENT C 0N T E N T %

Figure 2.5: Influence of Cement Content on Friction Angle of M onterey No. 0 Sand

(Rad and Tumay, 1986)


20

5 0

A A A A A C C. 2%40

3 0

20

1 0 040 8020 60RELATIVE DENSITY (p ercen t)

Figure 2.6: Influence of Cement Content on Cohesion (Rad and Tumay, 1986)


21

Table 2.1: S trength Param eters of Cem ented Sands (Drained Tests)(R ad and Clough,

1982)

Relative Cem ent Cohesion Friction A ttractionDensity C ontent ( ^ ) Angle

% % peak res peak res peak res31 0 0 0 33 33 0 045 0 0 0 35 34 0 077 0 0 0 39 36 0 020 1 4.4 0.0 33.0 32.0 6.8 0.047 1 8.6 0.0 35.5 34.2 12.1 0.071 1 12.4 0.0 37.9 35.4 15.9 0.018 2 10.0 4.4 32.6 32.6 15.6 6.940 2 17.0 9.6 34.8 33.9 24.4 14.380 2 30.0 18.9 38.8 35.8 37.3 24.8

5. C em entation results in an increase in dilation.

6. R esidual friction angles of cem ented and uncem ented sands are sim ilar.

2 .2 .3 U n c o n f in e d C o m p re s s iv e S t r e n g th , qj

Unconfined compression strength , qj is generally used as an index to classify co

hesive soils. Table 2.2 provides the proposed classification of cemented sands based

on unconfined compressive strength values. This classification by Rad and Clough

(1982) provides bo th sim plicity and versatility. R ad and Clough propose th e classi

fication for all cem ented soils irrespective of the cem enting agent.

A car and E l-Tahir (1986) conducted unconfined compression tests on the a rti

ficially cem ented soils. T he tests were conducted on specim ens of varying relative

densities (40-50, 60-75 and above 80 %) and cem ent contents (1 and 2 %). A curing

period of 14 days was adopted. The results of th is study are presented along with

R ad and Clough’s (1982) results in Table 2.3. B oth studies used pluviation for spec

im en preparation . Table 2.3 indicates th a t both results are quite sim ilar im plying

th a t th ere is not a significant difference between M onterey No. 0 and No. 0/30 sands.

These results also dem onstrate the repeatab ility in preparing artificially cem ented

specim ens.


22

Table 2.2: Proposed Clcissificatioii System for Cem ented G ranular Soils (R ad and

Clough, 1982)

CLASSIFICATION

Very weakly cem ented

W eakly cemented

M oderately cemented

Strongly cemented

Very strongly cem ented (rock type)____________

(kPa)< 100

100 - 300

300 - 1000

1000 - 3000

> 3000

D ESC R IPTIO N

cem entation alm ost unapparent to touch breaks down under slight finger pressure; can be scratched w ith the finger tip hardly breaks under finger pressure; can be easily scratched with the finger nail difficult to trim , can be hardly scratched w ith the fingernailvery low streng th soft rock

qj - Unconfined Compressive S trength

According to the classification system , the 1 and 2 % artificially cem ented sands

are categorized as very weakly cemented.

2.2.4 Hydraulic Conductivity

The influence of cement content on the hydraulic conductiv ity is presented (El-

Tahir and Acar, 1983) in Figure 2.7. The increase in cem ent content results in

reduction of perm eability. This reduction is due to the clogging of the pores by the

finer cement particles. As expected, denser specimens have lower perm eabilities than

loose specimens. The question then arises w hether drained or undrained conditions

prevail during cone penetration testing. This definitely will depend upon the ratio

of the ra te of penetration , S, to the perm eability, k, of the m edium , p. It is not well

established above which | values drained conditions will prevail. It is necessary to

conduct experim ents to assess the effects of

2.2.5 Compressibility

An equivalent Young’s m odulus is generally used for cases o ther th an one di

m ensional compression (Schm ertm ann, 1977). The equivalent Young’s m odulus is


23

CD<LÜ2OCLUCL

C o m p a c t i o n to 9 5 % o f M a x i m u m Proc to r D e n s i t y

Compac t ion to 1 0 0 % o f M a x i m u m Proc to r Densi t y

O " " -

c ) 9 0 d o y s o ld s o m p l e

4 6 8 10

C E M E N T C O N T E N T %

Figure 2.7: Hydraulic Conductivity of Cemented Specimens (El-Tahir and Acar,

1983)


24

Ta.ble 2.3: UCS of Cem ented Sands

C em entationC.C.

RelativeDensity

9 / M g / M

% % (kPa) (kPa)0 31 0 00 45 0 00 77 0 01 25 10 71 35 15 NA1 50 19 201 80 28 302 25 22 252 35 33 NA2 50 41 422 80 54 55

“ - A car and El-Tahir (1986)

- R ad (1984)

calculated from drained triax ia l test results. T he Young’s m odulus a t the 50 and 25

% failure stresses are calculated and are shown in the Tables 2.4, 2.5 and 2.6.

2.2.6 Natural Versus Artificial Cementation

T he difficulty in sam pling naturally cemented deposits leads investigators to

use artificially cemented deposits. T he sim ulation of naturally cem ented deposits by

using artificially cemented laboratory specimens is discussed in this section. This

discussion is prim arily focused on the strength and deform ation behavior under con

fined and unconfined loading. The study conducted by Clough et al. (1981) is used

for comparison purposes as th is is the only comprehensive study on the subject.

N aturally cem ented soil samples were obtained from two sites, bo th located

on the San Francisco Peninsula. The first location was situa ted west of Stanford

University and the second was located on the bluffs along the Pacifica Coast a t the

no rthern end of the city, Pacifica, California. SLAC-1 and SLAC-2 were the sands

collected from the first site and PAC-1, PAC-2 were the sands collected from the

second site. The SLAC-2 and PAC-1 are weakly cemented and have a lower density

and fines content than SLAC-1 and PAC-2 sands, which are strongly cem ented. Block

sam pling m ethod was used for obtaining undisturbed samples. D rained triax ial tests


25

Table 2.4: Young’s M odulus a t the 50 and 25 % Failure Stresses for U ncem ented

M onterey No. 0 Sand (C.C. 0 %)

Cem entC ontent

RelativeDensity

ConfiningStress

Young’sM odulus

C.C. D r a E2r> Eho% % kPa MPa MPa0 32 103 35.5 21.30 32 207 45.7 35.50 32 345 53.3 40.00 45 103 26.7 22.90 45 207 80.0 64.00 45 345 160.0 91.40 77 103 64.0 53.30 77 207 266.6 168.40 77 345 400.0 266.7

Table 2.5: Young’s M odulus at the 50 and 25 % Failure Stresses for Cem ented


Cem entContent

RelativeDensity

ConfiningStress

YouMoc

ng’sulus

C.C. D r cr E2, E;,o% % kPa MPa MPa1 25 103 17.7 15.21 25 207 35.6 32.01 25 345 64.0 47.11 50 103 35.5 30.41 50 207 49.2 45.11 50 345 67.1 49.21 80 103 64.0 53.31 80 207 85.3 68.11 80 345 91.4 71.1


26

Table 2.6: Young’s M odulus a t the 50 and 25


Failure Stresses for Cem ented

Cem ent R elative Confining Young’sContent Density Stress M odulus

C.C. D r a 25 Eio% % kPa MPa MPa2 25 35 16.0 15.22 25 103 18.3 16.42 25 207 24.6 22.52 50 35 26.6 22.92 50 103 44.4 40.02 50 301 71.1 60.42 77 35 29.4 27.42 77 103 110.3 88.82 77 207 128.0 91.4

and unconfined compression tests were conducted.

Artificially cem ented sam ples (C.C. 1, 2 and 4 %) were prepared by using Mon

terey No.0/30 sand and they were tested under identical conditions as above. The

results of artificially and natu rally cem ented sands are shown in Table 2.7.

Some observations m ade from the above study arc;

» T he streng th envelopes of the artificially cernentc'd soils closely resem ble those

of the naturally cem ented soils, except th a t the friction angles are somewhat

lower.

e Both artificial and natu ra l weakly cem ented sands show a b rittle failure mode

at low confining stresses with a transition to ductile failure a t higher pressures.

» Volumetric strains increase during shear at a faster ra te and at a sm aller strain

for cem ented sands (natu ral and artificial) than un cem ented sands.

• The residual streng th for a cem ented sand is close to th a t of an uiicem ented

sand, although some degree of residual cohesion intercept was observed for all

the cem ented sands investigated.

0 Like uncem ented sands, density, particle size and shape and grain size d istri

butions all have a significant effect on the behavior of cem ented sands.


27

Table 2.7: S trength Param eters of Artificial and N atural Cem ented Sands

Soil Relative Cohesion Friction UnconfinedType Density Angle S trength

% (!?)peak res peak res

Uncem ented 31 0 0 35 32 0AC2 74 46 5 34 33 180AC4 74 143 20 35 35 545AC4 90 152 25 41 35 670

SLAC-1 NA 365 75 49 36 1930SLAC-2 NA 12 6 40 36 50PAC-1 NA 25 5 40 35 110PAC-2 NA 175 60 37 35 700

R ad and Clough (1982) conclude th a t artificial cem entation sim ulates na tu ra l ce

m en ta tion reasonably well. Cem entation of 1 to 2 % sim ulates very weakly cemented

sands whereas m ore than 4 % cem entation is needed to sim ulate the behavior of

strongly cemented deposits.

A bibliography of several studies conducted on both natu rally and artificially

cem ented sands is presented in Table A .l in Appendix A. T he m ain conclusions of

these investigations are reported in Table A .2.

2 .2 .7 D y n a m ic C h a r a c te r i s t ic s - I n t r o d u c t io n

A brief sum m ary of the dynam ic characteristics is compiled in th is section.

T he prim ary objective of th is review is to quantify the influence of cem entation on

liquefaction resistance and low strain shear m odulus with the aim to provide semi-

em pirical correlations between penetration param eters and dynam ic characteristics.

D ynam ic characteristics of cemented sands are compiled from the studies conducted

by R ad and Clough (1982), Acar and El-Tahir (1986) and Saxena and Reddy (1988).

T he sands used in these studies are M onterey No. 0 and M onterey No. 0/30. Even

though there is a slight difference in grain size distributions, the overall response of

th e two sands to dynam ic loading can be considered as sim ilar (Acar and El-Tahir,

1986). Hence, the results reported are assum ed to be valid for M onterey No. 0/30

sand.


28

Different definitions of liquefaction are provided in the litera tu re . The one pro

posed by Rad and Clough states, th a t for all practical purposes, failure occurs when a

given value of double-am plitude axial strain is reached. They have considered 5 % as

double-am plitude axial strain . This selection is based upon the following argum ents;

1. At th is stra in , it is reasonable enough to assum e th a t all cem entation bonds

would be damaged.

2. In itia l liquefaction for loose to m edium dense uncem ented and cem ented sands

usually occurs when double-am plitude axial strain is equal to 4 to 6 %.

3. A t low relative densities (less than 50 %), double am plitude axial strains higher

th an 5 % usually result in unacceptable reductions in the applied vertical load.

For determ ining the influence of cem entation or cem ent content on liquefaction

resistance, dynam ic triaxial tests were conducted by Rad and Clough (1982) and are

briefly discussed in section 2.2.7.1.

A nother property discussed in the dynam ic behavior is the low strain dynam ic

shear m odulus of the soil. The shear m odulus is defined as the ratio of shear stress

to th e shear strain . T he factors th a t affect shear m odulus are: effective octahedral

norm al stress, void ratio , am bient stress history and vibration history, degree of

sa tu ra tion , octahedral shear stress, grain size characteristics, am plitude of strain ,

frequency of vibration, secondary effects due to tim e of loading and increm ent of

load, soil s truc tu re and tem peratu re (Acar and El-Tahir, 1986).

D uring th e last two decades, a num ber of researchers have suggested empirical

relationships for determ ination of m axim um shear m odulus of soils. The influence

of cem ent content on dynam ic shear m odulus is investigated by Acar and El-Tahir

(1986) and Saxena and Reddy (1988).

2.2.7.1 Large Strain Dynamic Stress-Deformation Behavior (Dynamic Triaxial Testing)

U ndrained cyclic stress-controlled triaxial tests were conducted on uncem ented

and artificially cem ented sand specimens. F igure 2.8 shows the tes t results p lo tted

in term s of the stress ratio versus num ber of cycles necessary to achieve 5 % double


29

am plitude axial strain for average relative density of 51 %. C em entation increases

resistance to liquefaction. This figure also reveals th a t the effect of cem entation on

the liquefaction resistance of sand decreases as the num ber of cycles to liquefaction

increases.

2.2.7.2 Resonant Column Test (Low Strain Dynamic Behavior)

T he m axim um shear m odulus of uncem ented sands can be evaluated by em piri

cal correlations proposed by several investigators (Chung et al., 1984). The m axim um

shear m odulus of sands is given as,

ffm .. = (2.1)

w here S = stiffness coefficient; / ( e ) = a function reflecting the effect of void ratio , e;

(7o = m ean effective confining pressure; Pa = atm ospheric pressure in the sam e un its as

Gmax and (7o; and n = a constant. T he m aximum shear m odulus is often norm alized

for the effect of density w ith,

/(e) = 0.3 +0.7 (2.2)

T he m axim um shear m odulus in the tests conducted by Acar and El-Tahir (1986) is

expressed as.

Similarly, tests were also conducted on artificially cem ented specim ens a t various

rela tive densities and cement contents (1, 2 and 4 %). The results reported in this

section are taken from th e study conducted by Acar and E ltah ir (1986). Solid cylin

drical specim ens of d iam eter of 36 m m and length of 80 m m were tested. P luviation

technique was used for low cem entation and tam ping or com paction was used for

higher cement contents. Confining pressures of 35, 103 and Z ib k P a were chosen to

study th e variability of shear m odulus and dam ping.

T he stiffness coefficient, S , and n values of cem ented specim ens are shown in

Table 2.8. It was found th a t low levels of cem entation results in an increase in stiffness

coefficient, while the exponent, n, is within the variability of values presented for

uncem ented specimens.


30

0.6

o Z CEMENT^ 0.4cd

K

mCDu

4-)U1 0.2

N um ber of Cycles

Figure 2.8: Stress Ratio Versus Number of Cycles: Cyclic Triaxial Tests on 1 %

Cem ented Specimen of Relative Density 51 % (Rad and Clough, 1982)


31

Table 2.8: Stiffness Coefficients and n Values of Cem ented Sands

Relative Cem ent Stiffness ^ m e a n

Density Content coefficient% %25 0 621 0.42

1 867 0.432 1122 0.43

35 0 638 0.441 918 0.432 1184 0.42

50 0 624 0.431 1028 0.422 1318 0.42

75 0 658 0.411 1115 0.422 1387 0.42

Figure 2.9 depicts the variation of m axim um shear m odulus versus confining

stress for cem ented specimens prepared a t relative density of 50 % (results a t o ther

relative densities are presented in Appendix A). The increase in dynam ic shear m od

ulus of artificially cem ented specimens is a ttribu ted to the increase in stiifness coef

ficient.

T he equation expressing the m axim um dynam ic shear m odulus of sands is re

vised for the effect of cem entation as

Gc. = R -S 0.43 (2.4)

0.3 + 0.7e2'

where Gc = m axim um dynam ic shear m odulus including the effect of cem entation;

R = stiffness ratio ( ^ ) ; Sc — stiffness coefficient for uncem ented specimens (631 for

M onterey No. 0/30 sand). R values are reported by Acar and El-Tahir (1986) for 1

and 2 % cem entation.

2.3 C one P enetration T esting in Sands

Several types of insitu testing equipm ent have been gaining wide popularity

in geotechnical investigations for the past two decades. The cone penetration test

(C PT) has become the most widely used insitu technique in the last few years. Even


32

S. 1 0 0 0

o 5 0 0 -in =)

=> o o 2

a:5 100 X in

Monterey No 0 Sond D, = 5 0 % X « 15.23 kN/m^

Cementolion % “ 7y 2 I

3z>1X

2E

5 0 -

-I_____I I I 1.1 % I -1 I I i i . j . i50 100 500 1000

CONFINING PRESSURE, ( kpQ)

Figure 2.9: T he Variation of M aximum Shear Modulus Versus Confining Stress for

Cemented Specimens Prepared at Relative Density of 50 % (Acar and El-Tahir, 1986)


33

in the U.S., where standard penetration test is the conventional testing m ethod, cone

penetration testing has been gaining more and m ore popularity in the last decade.

2.3.1 Cone Penetrometer

T he cone penetrom eter tes t consists of advancing a cylindrical rod w ith a conical

t ip in to the soil and m easuring th e forces required to push th is rod. T here are two

resistances m easured during th e CPT, the tip resistance is the soil resistance to

advance the cone tip and friction resistance is the friction betw een the soil and the

sleeve of the cone. Friction ra tio is defined as the ratio between th e friction resistance

and tip resistance and is expressed in percent. All these p roperties are used to identify

and determ ine the soils and their properties. A typical cone penetrom eter record is

presented in Figure 2.10.

C P T param eters are used for the following assessments:

® Continuous soil stratification,

® Assessment of the undrained shear strength , stress history or over consolida

tion ratio (O C R ), consolidation param eters and conductiv ity characteristics of

cohesive soils,

@ Assessment of relative density, drained strength param eters and com pressibility

characteristics of cohesionless soils,

o Evaluation of liquefaction potential of cohesionless soils,

e D eterm ination of pile foundation capacities,

® Assessment of ground w ater pressures, if piezocone is used,

a Settlem ent calculations of footings in cohesionless soils.

There are quite a few interesting developm ents in the s ta te of the a rt in cone

pene tra tion testing. Different sizes and shapes of cones are in use (Baligh, 1981;

Tumay, e t al., 1981). ASTM Standard D3441 recommends a s tandard cone w ith an

apex angle of 60 degrees, tip area of 10 cm^ and a sleeve area of 150 cm “. Cones

sm aller than the standard cone are generally used in sites where shallow dep ths need


34

: n i

i i s -tu cc s tu a. M-

I!O Ui

il

Figure 2.10: Schematic of a Electrical Cone (Juran and Tumay, 1989)


35

to be explored, in pavem ent and subgrade explorations and in finer classification of

th e s tra ta . Larger cones are generally used when standard cone can not be used to

pene tra te harder s tra ta .

The m ajo r breakthrough in cone penetration testing is the m easurem ent of pore

w ater pressures during penetration (P C P T ). This was first introduced in the early

1970’s by advancing a separate pore pressure probe into th e ground (W issa e t al.,

1975; Torstenson, 1975). Similarly, in 1970-73, the Norwegian In stitu te of Technol

ogy, Trondheim , used a pore pressure probe of the same shape as standard cone, bu t

only pore pressure m easurem ents could be m ade and it was necessary to carry out a

separate C P T in order to correlate the cone resistance and the pore pressure (Janbu

and Senneset, 1974).

In th e m id 70’s, the piezom etric elem ents were incorporated into standard elec

tric cone penetrom eters in which pore pressures were m easured along with cone

penetration resistance in some cases and w ith sleeve friction and cone inclination in

o ther cases (Acar, 1981; Baligh et al., 1981; De R uiter, 1982). Subsequently P C P T

which is also known as piezocone was used to m easure the pore pressures a t the cone

tip and along the shaft (Tum ay et al., 1981). The piezocone was also used to m easure

the dynam ic pore pressures (Smits, 1981).

C urrently new devices are developed and incorporated in th is versatile piece

of equipm ent to m easure shear velocity, conductivity and even a fiber-optic eye for

chemical characterization.

2.3.2 Calibration Chamber Testing

The calibration of geotechnical instrum ents like C PT for insitu tests is achieved

by carrying out laboratory tests on homogeneous and reproducible soil samples,

under accurately controlled states of stress and deform ation. Such calibration is

indispensable in developm ent of insitu testing equipm ent and in the in terp reta tion

of th e param eters obtained from testing (Bellotti et al., 1988).

M ost of the laboratory study in the litera tu re using C P T or P C P T was con

ducted either in large triax ial tests (Canou et ah, 1988) or in calibration chambers

(Holden, 1971 and 1977; Schm ertm ann, 1976; Tumay, 1976; Parkin, 1988; Been et

al., 1987; Baldi et al., 1981). Em pirical correlations in clays are m ore reliable since


36

in situ vane tests can be conducted adjacent to a C P T profile. This m akes it possi

ble to empirically correlate the undrained shear streng th param eter (m easured w ith

vane shear tests) w ith the penetration param eters like tip resistance, friction resis

tance and friction ratio. However in granular soils, it is very difficult to correlate

th e s treng th param eters w ith the penetration param eters. There is no insitu testing

m ethod for m easuring streng th param eters directly in sands. Generally there are two

approaches for in terpreting th e results in such m aterials; either correlating directly

w ith m easured quantities or by conducting tests in large calibration cham bers. T he

la t te r approach has been extensively used by different investigators (Veismanis, 1974;

C hapm an, 1974; Holden, 1977; Bellotti e t ah, 1982; Villet and M itchell, 1981; E id,

1987). T he cham bers designed by several investigators are capable of housing large

dim ensioned soil samples (G hionna and Jamiolkowski, 1991).

T he use of calibration cham ber to calibrate an insitu device has the following

advantages:

1. Tests can be perform ed on uniform and highly reproducible sand specimens

whose properties are well known. Hence, empirical or sem i-em pirical correla

tions can be m ade,

2. I t is possible to m onitor the stress and strain conditions around the sample,

3. Saturation can easily be achieved.

T he m ain four phases of operation in a calibration cham ber are:

1. P reparation of the specimen,

2. Saturation (if required),

3. One dim ensional compression,

4. C onducting the requisite test.

A sam ple of the soil a t a particu lar relative density is prepared in the calibration

cham ber and then consolidated to the desired stress levels. On th is sam ple, the tests

are conducted w ith the insitu testing apparatus and the param eters and fs are

recorded along the vertical profile. Laboratory tests are then carried out on the


37

sam e soil a t the same relative density to determ ine the engineering properties. The

param eters m easured in calibration cham ber can be correlated directly or can first

be correlated w ith relative density.

In general, two types of calibration cham bers are used; cham bers w ith rigid

or flexible walls. A rigid wall calibration cham ber imposes a boundary condition

of zero lateral stra in on the specim en. The flexible wall cham ber allows the lateral

m ovem ent. The design of a flexible wall calibration cham ber perm its an accurate

control and m easurem ent of the vertical and horizontal stresses and strains. Four

types of boundary conditions can be sim ulated (Bellotti et al., 1985; Holden, 1977)

in the tests:

® B C l: constant stresses on the boundaries,

• BC2: zero strains in horizontal and vertical directions,

• BC3: zero lateral strain ,

• BC4: zero vertical strain.

Holden (1971) proposed th a t the field boundary conditions would lie somewhere

betw een constant stress condition and zero lateral strain conditions i.e. B C l and

BC3. This sta tem ent is valid for lower diam eter ratios. For larger diam eter ratios,

bo th boundary conditions should give identical results.

T he factors th a t affect the tes t results in calibration cham ber are described in

the following sections.

2.3.2.1 Chamber Size and Boundary Condition Effects

C ham ber size and boundary conditions affect the cham ber te s t results consider

ably. Hence, an a ttem p t is m ade in th is section to review and discuss them . C ham ber

size implies th a t the size of the cham ber specim en should be such th a t the results

will be representative of those obtained in insitu field testing. T he Çc and / j charts

produced on loose sands were found to have the characteristic shapes th a t reflect

th e density conditions. However the Çc curve in a dense sand instead of reaching a

p lateau , was observed to increase w ith depth (Parkin and Lunne, 1982). This re

flects the density and boundary condition effects. This effect, known as the ‘cham ber


38

size effect’, was pointed out by Parkin (1988) and Park in and Lunne (1982). T ip

resistances are m ainly influenced by the diam eter ra tio ^ where D is the diam eter

of the specim en and d is the diam eter of the penetrom eter tip . It was observed

th a t the cham ber size effect is not very im portan t in loose sands bu t becomes very

im portan t when the relative density of the specim en increases. Particu larly in dense

to very dense sand specimens, the m easured qc value increases as the d iam eter ratio

decreases (Bellotti et ah, 1985), possibly due to the restra in t d ilatancy effects.

I t was observed by Bellotti e t al. (1985) th a t the cham ber size effect is subject

to a complex interaction w ith boundary conditions imposed during the test. Bellotti

e t ah, (1985), Parkin and Lunne (1982) and m any others believe th a t the problem s

of the boundaxy conditions used in the cham ber and the related cham ber size effect

require fu rther intensive experim ental and theoretical investigations. F igure 2.11

depicts the effect of cham ber size and boundary conditions on the C P T for Hokksund

sand (Parkin and Lunne, 1982). Despite the large diam eter ratios, the qc is still a

function of diam eter ratio for dense sands. However, for loose sands, a d iam eter ratio

of 20 seems to be sufficient and beyond th is value, there is no significant increase

in tip resistance. At a given diam eter ratio , the test under boundary condition 3

predicts higher tip resistance th an under boundary condition 1 (Been et al., 1988).

In the case of zero lateral strain condition (BC3), higher stresses will exist at the

cham ber boundaries than in the field a t an equal d istance from cone, hence higher

tip resistance is recorded. In constant stress boundary conditions (B C l) , higher

stresses will develop in the field a t an equal distance from the cone, hence lower tip

resistances are recorded in the cham ber (Been et al., 1988).

2.3.2.2 Crushability

Crushability of the sands used in the cham ber also influences the cone resistance.

Several studies conducted by Baldi (1981) and Bellotti e t al. (1988) showed th a t

extensive grain crushing was observed in the sand during cone testing. T he crushed

aggregates may have different strength properties than the original sand tested and

the influence of this crushing on the tip and specifically the sleeve friction resistance

need to be investigated. Figure 2.12 presents the effect of crushability on cone

resistance (Robertson and Cam panella, 1984; Bellotti e t al., 1991).


39

5 0

NOTE: 'BCKORIZONTAL

S T R E S S CONTROL B3< HORIZONTAL

STRAIN CONTROL

20 8 3 , NC

•B I.N C* 5 0 k N /m

wo

&

UI LOOSED ,= 3 0 %LEG END

+ 8 3 NC

© 81 OC ® 8 3 OC

S 2010 5 0 100

D IA M E T E R RATIO

(C h a m b e r d ia m e te r / C one d ia m e te r )

Figure 2.11: Cham ber Size and Boundary Condition Effects (Parkin and Lunne,

1982)


40

Calibration Chambers“cpt crushing'

qc 25

(MPa) 20

00

-V .A ' '■0 \ « O ',

:

I8g B

Ticino Sand

B pass. 100

0 pass.200

cone diameter=3.57cm

2 4 6 8 10 12pass.(%) ASTM sieve n. 100(0.149mm) pass.(%) ASTM sieve n.200(0.074mm)

Figure 2.12: Influence of Crushing (Bellotti et ah, 1991)


41

2.3.2.3 Fabric, Shape and Texture of Grains

T he o ther factors th a t may influence the test results are fabric, shape and tex tu re

of grains in sands. I t is extrem ely difficult to reproduce the field fabric in the cham ber.

However for sands, Been et al., (1988) believe th a t the fabric will have m inor influence

on the penetration resistance. They advocate th a t the cone m easures the resistances

in a rem oulded zone in which fabric is destroyed. T he size and shape of the sands

used m ay also influence the tip and friction resistances significantly. T he grain size

to the cone size ratio may be one variable affecting the tip resistance and there is no

inform ation on th is aspect.

M ost of the above problems can be alleviated by constructing a large cham ber or

a flexible sm all cham ber and using a small cone for the studies (Eid, 1989; Dario De

lim a, 1990). This m eans a large diam eter ratio can be achieved and hence, cham ber

size and boundary condition effects on results can be m inimized. However, o ther

param eters, like ratio of cone diam eter to grain size and crushability of sand used

m ay also affect the results. In an a ttem p t to evaluate the effect of these param eters,

cone test results on different sands are collected. T he next section covers different

investigations and variables used in each study.

2.3.3 Synthesis of Experimental Data Obtained in Calibration Chamber Testing of Uncemented Sands

Cone test results conducted on several sands in different calibration cham bers

are collected and presented in this section.

T he cone test results are collected from different research projects conducted by

the following investigators:

1. P resent Study - M onterey No. 0/30 Sand, USA

2. Eid (1987) - M onterey No. 0/30 Sand, USA

3. Baldi e t al. (1981) - M edium Coarse Sands, SATAF 1 and SATAF 2, Italy

4. Villet and M itchell (1981) - M onterey Sand Nos. 2, 30 and 60, USA

5. H arm an (1976) - O ttaw a No. 90 (5a) and Hilton M ine Tailings (5b), USA


42

Table 2.9: C haracteristics of Tested Sands

Sand No. Gs Imax(k N Im ^ )

'Jmin(&Ar/m3)

Cu ^max ^min Shape

1 16.90 14.51 1.6 0.803 0.563 SR-R2 16.92 14.56 1.58 0.811 0.566 SR-R3 NA 16.66 14.12 1^85 NA NA SA-A4 NA 1R9 16.9 1.48 NA NA SR-R

5 a. 2 j # NA NA 1.85 0.789 Cf4a6 R5 b. 3.02 NA NA 2.0 1.05 0.620 A

6 2 j # NA NA 1.5 0.977 0.605 SA7 3.02 14.2 17.0 1.3 NA NA NA8 NA NA NA 2.75 NA NA NA

Note: SR - Sub f ounded, R - Rounded, SA - Sub Angular, A - A ngular

6. Fioravante et al. (1992) - Toyoura Sand, USA

7. Lhuer (1976) - Edgar and Reid-Bedford, USA

8. N u tt and Houlsby (1992) - C orbonate Sand, UK

T he physical properties of these sands are presented in Table 2.9. The cone te s t re

sults like tip resistance, friction resistance, friction ratio along with the vertical stress

and relative density are reported in Table A .l to A.8 in A ppendix A. O ther details

like diam eter ratio and the boundary condition in which the test was conducted are

also reported in the sam e table. These results are used to study and analyze the

effect of different variables affecting the calibration cham ber testing. The im portan t

po in t worth m entioning here is all the test results m ay have been influenced by one

or m ore th an one variable. Hence, while analyzing each, d a ta th a t is obtained under

identical conditions are used. For exam ple, da ta obtained from the sands of sim i

lar compressibility, size and shape and specimens tested under identical boundary

conditions are used to investigate the influence of cham ber size effects.

2.4 Cone P enetration T esting in C em ented Sand

Very few studies have been reported in the litera tu re covering cone penetration

testing in cem ented sands (Rad and Tumay, 1986; Akili and Nabili, 1988). These


43

studies aim ed at providing prelim inary assessment of the influence of cem entation

on cone resistance param eters. B oth studies are conducted in rigid cham bers a t the

d iam eter ra tio of 15 and the results are valid for very low confining pressures (less

than 5 kP a).

R ad and Tum ay (1986) presented tip resistance and sleeve friction values m ea

sured in experim ental m odeling of penetration in cem ented sands. Artificially ce

m ented M onterey No. 0 sand was employed in th is study. P luviation m ethod was

used to prepare specim ens in a PVC mold. The diam eter and length of the speci

m ens were 30 and 45 cm respectively. Specimens were cured for 7 days. Testing was

conducted w ith a 2.0 cm d iam eter cone a t 2 ^ speed. The tests were conducted in

th e m iddle of the specim en in order to elim inate the cham ber size effects. Figure 2.13

shows th e penetration profile along the depth . It is observed th a t the tip and friction

resistance increases w ith the cem ent content and relative density. The penetration

values a t 25 cm dep th were reported in order to reduce the rigid bo ttom effects. The

t ip resistance and friction resistance values are presented in Table 2.10.

Table 2.10: T ip and Friction Resistances R eported for Cem ented Sands

Reference Relative Cem ent ResistanceDensity C ontent Tip Friction

% % M N k N

Rad and Tumay 20 1 o j a 15.0(1986) 47 1 1.04 17.0

71 1 24.018 2 1.37 19.040 2 2TW 23.080 2 4TW 27.0

Akili and Nabil 43 0.2 4.2 NA(1988) 90 0.2 8.9 NA

43 1 6.7 NA90 1 10.7 NA

NA - Not Available

Akili and Nabili (1988) also investigated the influence of cem entation on cone

resistance param eters of beach sands and their results are also depicted in Table 2.10.

T he results obtained in both studies are influenced by the rigid boundaries and low

d iam eter ratios (15). This argum ent is valid particularly in the case of dense sands


44

CC - Cem ent Content 1%)

Dr - Relative Density (%)

\ \

CC

Eo (80,2)

X

^ 20UJQ

(71, I)

(18,2)30

(47,1)(40,2)

405.02.0 3.0

TI P R E S I S T A N C E , q ^ . ( M N / m 2 )4.01.0

Figure 2.13; Penetration Profiles in Laboratory Tests (Rad and Tumay, 1986)


45

since dilation of th e sand around the cone will be restric ted (restrained d ilatancy)

by the presence of th e rigid boundaries. This will, in tu rn , influence the cone results.

A flexible double walled calibration cham ber was bu ilt a t LSU (de Lim a and

Tumay, 1991) and is used in the present investigation. A m in iatu re cone is used in

th e testing resulting in a diam eter ra tio of 42 (de Lim a and Tumay, 1991). This

d iam eter ratio reduces the cham ber size and boundary condition eifects significantly.

The results of this study are analyzed along with the above reported test results.

2.5 C one P en etration T esting A nalysis

T he curren t state-of-the-art regarding the analysis and application of cone pen

e tra tion tes t results depend largely on semi-empirical and em pirical correlations.

These correlations are based on the following approaches: bearing capacity theories,

sim ulation of the penetration m echanism by the cavity expansion theories and the

stra in p a th approach. Numerical sim ulations of the problem have been a ttem p ted

by the finite difference and finite elem ent m ethods. Theoretical studies involving in

friction resistance predictions is also presented in section 2.5.6. Existing em pirical

and sem i-em pirical m ethod in sands are explained in the next section. T he final

section covers s ta te param eter in terp reta tion as suggested by Been et al. (1986).

2.5.1 Bearing Capacity Theories

M any investigators have analyzed cone penetration as a bearing capacity prob

lem (Meyerhof, 1963; Durgunoglu and Mitchell, 1973; Janbu and Senneset, 1974;

Baligh, 1975). These theories assum e different failure m echanism s which are then

used to calculate u ltim ate bearing capacities using lim it equilibrium approach. W hen

penetration is trea ted w ith the conventional bearing capacity theories, soil is often

assum ed to behave as a rigid-perfectly plastic m aterial. Therefore, the stra ins and

com pressibility of the m aterial are neglected. These theories assum e different failure

surfaces. The cone penetration is analyzed as an axisym m etric problem and the so

lutions are developed under plane strain conditions using lim it equilibrium approach.

For shapes o ther than the idealized problem s, the solutions are modified by applying

em pirical shape factors.


46

Several studies (Durgunoglu and M itchell, 1973; Janbu and Senneset, 1974;

Baldi et ah , 1981; Villet and M itchell, 1981) have successfully used these bearing

capacity theories in determ ining the m easured resistance param eters. Two theories

which are used in th e present research are described briefly here.

Durgunoglu and M itchell [D & M] (1973) modified the Terzaghi(1943) bearing

capacity equation by considering the effect of sym m etry, effect of foundation shape

and the effect of roughness;

quc = cNcCc + j B NjgC'iq (2.5)

where N^g is the bearing capacity factor for the friction-surcharge term and is

th e corresponding shape factor. The bearing capacity factors depend on the soil

friction angle <j) , base semi-apex angle a , base roughness | and relative dep th of

penetrom eter base The shape factors are calculated using Brinch-Hansen (1961)

param eters. Figure 2.14 presents the bearing capacity factors and the assum ed failure

m echanism .

Janbu and Senneset [J & S](1974) assum ed the failure surface shown in Figure

2.15. T he stress field is also depicted in th is figure. P lane strain conditions are

assum ed. This m ethod assumes th a t the failure surface fans out to different planes

of plastification, /?, depending upon the dilational characteristics of the soil deposit.

This solution is successfully used in estim ating the long term bearing capacity for

point bearing piles in both fine and coarse grained soils (Senneset et al., 1982).

T he general bearing capacity as per this m ethod is expressed as:

Qy + a = Ng (a '„ + a ) - \ -U o - Ny, Aub -f ^ 7 5 A'iy (2.6)

w here is vertical u ltim ate bearing capacity, Ny is the bearing capacity factor for

pore pressure, Uo is in itia l pore pressure, Aub is pore pressure a t foundation base, a

is a ttrac tio n { j ^ ) , or'yo is effective vertical stress, Ng and are bearing capacity

factors and B is width or diam eter of footing or penetrom eter.

This approach uses a ttrac tion a and friction angle (j) in the analysis. T he a t

trac tion represents the m axim um tensile strength intercept. J & S theory considers

th e excess pore pressure effects along the shear surface on the bearing capacity of

th e cone. T he bearing capacity factors are derived from the equilibrium of the given

shear surface geom etry (Senneset et al., 1982).


47

H I h

c

Of = 3 0 ° j /0 = 0 .5

D / B

> 2 0,310

.210

10

25 3 0 35 4 0 45 50

10 '

,310'E D / B:30x

10

25 30 35 4 0 45 5 0FRICTION ANGLE, 4> I D e g r e e s ) FRICTION ANGLE ,

( D e g r e e s )

Figure 2.14: Failure Mechanism Assumed in Durgunoglu and M itchell’s Theory (D ur

gunoglu and Mitchell, 1973)


48

1000

IDEAUZATION

100

O)

CLAY S O SAND

--K

0 0.5 1.0Friction tan^'

Figure 2.15: Failure Mechanism Assumed in Janbu and Senneset’s Theory (Janbu

and Senneset, 1974)


49

In the analysis, a ttraction param eter and the friction angle from triaxial testing

are used for estim ating the cone resistances. The bearing capacity param eters are

taken from charts. A proper plastification angle /? value has to be chosen. The

plastification angle can be either positive or negative and this sign depends on the

compressibility of the m aterial. For loose sands and normally consolidated clays,

positive values of 0 to 30 degrees are adopted whereas for dense and over consolidated

clays, the range is 0 to —40° or in some cases beyond 40. The negative value implies

th a t the soil will dilate during penetration. Therefore, negative plastification angles

are selected for dilating sands.

Villet and Mitchell (1981) and Baldi et al. (1981) dem onstrated th a t D ur

gunoglu and M itchell correlation gave reasonable agreement between the predicted

and m easured friction angles. Their results are presented in Figures 2.16 and 2.17.

Figure 2.18 compares the predicted and m easured values of tip resistance in ce

m ented sands (Acar, 1987; Puppala et al., 1993). The predicted values are obtained

by using Durgunoglu and Mitchell (1973) and Janbu and Senneset (1974) theories.

The m easured values are taken from the study conducted by Rad and Tum ay (1986).

This figure depicts th a t the predictions correlate quite well with measured values.

These theories assume rigid-plastic behavior for the medium. Cemented sands which

are b rittle in natu re exhibit a behavior which is close to the rigid-plastic assum ption.

This may be the reason behind the good correlations obtained between theoreti

cal and measured values. However, it is noted that these results can not yet be

generalized to higher confining stresses.

2.5.2 Cavity Expansion Theories

Analyses by cavity expansion theories are generally used for in terpretation of

pressurem eter tests (Hughes, et al., 1977), studies involving installation of driven

piles (C arter et al., 1986) and in the in terpretation of cone penetrom eter tests

(Greeuw et al., 1988).

The penetration mechanism is idealized as th a t of a spherical or cylindrical

cavity expanding in a semi-infinite medium. The advantage of such an assum ption

lies in uni-dimensional formulation of the problem which facilitates incorporation of

p lasticity models. Ladanyi (1969) used the cavity expansion theories in b rittle rocks


50

C O N E R ESISTA N C E q _ t k g /c m ^ ]

5 0 0 6003 0 0 400200100

MEDIUM DEN SE SA N D

O BCl

• BC3

DENSE S A N D

« BCl

20 * B C 3

EVERY DENSE S A N D

It-aUIQ

0 BCl

o B C 3

3 0

40

50

Figure 2.16: Comparison Between the Measured and Predicted Param eters (Baldi,

1981)


51

MEASURED PREDICTED

. - - oDrl%)

a. 100

50 100 150 200 250 300

VERTICAL S T R E S S ( k P a )

350

Figure 2.17: Comparison Between the M easured and Predicted Param eters (Villet

and Mitchell, 1981)


52

S t

I I Ia

H

IT3£

3

□ J & S T h e o r y ® D& MT h e o r y

2

1

02 30 1

Measured Tip Resistance (MPa)

Figure 2.18; Comparison Between the Measured and Predicted Param eters (Acar,

1987; Puppala et al., 1993)


53

and also applied the approxim ate solutions for lim it pressures in the spherical case to

the bearing capacity problems. Theories proposed by Vesic (1972) and Baligh (1975)

derive the cone resistance solutions in term s of soil compressibihty characteristics.

These theories are based on the work required to expand a cylindrical or spherical

cavity around the cone tip as it is driven into the soil (Eid, 1987).

The lim it pressure in the cavity expansion model is defined as the pressure in

the cavity which corresponds to a continuous deformation w ithout pressure increase.

This lim it pressure can be related to tip resistance in a simple way as suggested by

C arter et a l.(1986). Closed form solutions are presented for cohesionless soils (in the

case of uncemented sands) and also cohesive frictional soils (cemented sands). Mohr-

Coulomb yield criterion is adapted in the analysis. The solution for the pressure-

expansion curve is obtained by considering small strains. This theory is also used in

the present analysis.

The cavity is shown in the Figure 2.19. The plastic domain of the cavity is

extended up to a radius R from the center. Beyond this domain, the soil is in elastic

state. The final solution for lim iting pressure is a function of strength properties and

the dilation angle. Greeuw et a l.(1988) used this approach and dem onstrated th a t

the theoretical predictions correlated fairly well with experim ental tip resistance.

2 .5 .3 S tra in P a th A p p ro a c h

A recent and th ird approach to the penetration in soft cohesive soils was pro

posed by Baligh and Levadoux (1980), Tumay (1985) and Acar and Tumay (1986).

This approach takes into consideration the steady nature of the problem and finds the

strains and displacements around a cone penetrating an inviscid and incompressible

fluid. The strain field obtained from this m ethod would provide a first approxim ation

to strains induced by cone penetration testing and /o r pile driving in soft cohesive

soils. The strain field could then be used as an input to recently improved large

strain soil plasticity models to estim ate the generation of pore pressures, effective

stresses during penetration and dissipation of pore pressures when the driving or

penetration is stopped.


54

Plastic Elastic

Figure 2.19: Cavity Used in the Cavity Expansion


55

2.5.4 Numerical Methods

A nother approach is adopting a large stra in , elasto-plastic form ulation for sim

u lating the penetration m echanism using the finite elem ent m ethod. T he form alism

developed by Kiousis e t al. (1988), is applied to the solution of the cone penetration

problem in a soft cohesive soil. The plasticity m odels used in this specific study were

Von Mises and the cap model by Dimaggio and Sandler (1971). This theory predicts

th e displacem ent-strain , stress and pore pressure fields around the cone.

2.5.5 Discussion

All of these m ethods have their advantages and disadvantages. T he disadvan

tages of each m ethod are:

• In the bearing capacity approach, the effects of factors like soil compressibility,

excess pore w ater pressure, initial s ta te of stress, progressive rup ture , depen

dence of the in ternal friction angle on the m ean effective norm al stress and

the effective stress pa th are neglected (Tumay, 1985). Bearing capacity theo

ries consider the penetration of a rigid cone into compressible soil as a stress

controlled problem even though it is a s tra in controlled problem.

® Cavity expansion theories have been used by m any authors. Sim ulation of

cone penetration using cavity expansion theory is complex because of the large

stra ins around the cone tip and also due to the shape of the cavity which is

neither a cylinder nor sphere.

® T he approach proposed by Baligh and Levadoux (1980) and Acar and Tum ay

(1986) m ay be valid for soft cohesive soils. The m ethod assumes incom press

ibility in predicting the strain paths. Com pressible soils like carbonate sands

can not be in terp reted using this approach.

9 T he disadvantages of finite elem ent solutions are the cost of com putation, dif

ficulty in sim ulation of the problem particu larly near the cone where interface

elem ents and a knowledge of the behavior of such elements is needed. T he

constitu tive models used for the interface elem ents strongly affect the obtained

predictions. Kiousis et al. (1988) notes th a t a viscoplastic analysis will be a


56

m ore appropriate approach th an the classical p lasticity approach in the sim u

lation of the penetration problem by the finite elem ent m ethod. T his m ay be

necessary in cohesive soils since stra in rates in penetration are orders of m ag

n itude higher th an conventional geotechnical problem s. However, th e behavior

of soils a t such large strains and strain rates is not well-defined in the curren t

s ta te of the art.

All the above theories despite the ir lim itations are often used in pene tration

analysis and in estim ating tip resistance.

2.5.6 Friction Resistance (Sleeve Friction)

T he predictive m ethods for estim ating friction resistance assum e th a t sleeve fric

tion is only due to shear resistance. T he following form ula can be used in calculating

friction resistance:

/s = 5s (ct'„ - t-a) (2.7)

in which /s is friction resistance, Ss is |r | ta,n <j)K and |r] is interface friction ratio

defined as

T he stra ins in the vicinity of the sleeve are well beyond the strains corresponding

to peak streng th values (Acar and Tumay, 1986). Therefore, it is more appropria te to

use residual values in estim ating friction resistance. T he roughness coefficient should

be considered as a product of two factors: one corresponding to the m echanical

roughness of p ile/cone surface, and the o ther relating to relative vertical m ovem ent

betw een the pile/cone surface and adjacent soil. The upper lim it for th is factor is

1.0 and the lower lim it is around 0.55 in practice (Janbu , 1976; Acar e t al., 1982).

In the present case, |r | is taken as 0.65 (i.e. for steel to sand).

F igure 2.20 depicts the com parison between the m easured values by R ad and

Tum ay (1986) and predicted friction resistances for different coefficients of earth

pressure. T he above described approach gives very low values com pared to the m ea

sured resistances, if ea rth pressure coefficients a t rest are used. However if passive

ea rth pressure coefficients are used, the friction resistances come close to the m ea

sured values. The im plication is th a t the dilation a t the tip results in expansion

loading on th e shaft to increase the confinement on the sleeve. The ex ten t to which


57

th is m obilization of dilation occurs will depend upon the type of soil, th e density of

the deposit, the fabric and the depth of penetration . This implies th a t the friction

resistance m ainly depends on th e dilational characteristic of sands.

2.5.7 Empirical Methods

Em pirical m ethods are also used in estim ating the streng th and deform ation

properties. Schm ertm ann (1976) proposed an indirect m ethod for estim ating the

friction angle. This m ethod is based on the calibration cham ber results and is shown

in F igure 2.21. Figure 2.21 is used to estim ate the relative density, once the cone

resistance, Qc and vertical effective stress, cr(, are known. T he second figure (2.21) is

used to determ ine the friction angle w ith the known relative density.

O ther em pirical m ethods are used to estim ate the constrained m odulus, elastic

m odulus and shear m odulus (Baldi e t ah, 1981). In case of constrained m odulus,

m ost of these empirical correlations take the following form:

M = aqc (2.8)

T he a value ranges from 3 to 11 (Veismanis, 1974; Parkin e t al., 1982; Acar, 1981).

The m ore recent recom m endation varies between 1.5 and 4 (Lunne, 1991). O ther

results (Robertson and Cam panella, 1984 and Jamiolkowski e t al., 1988) are shown

in Figures 2.22 and 2.23. Relative density need to be determ ined prior to the use of

these figures.

Lunne and Christopherson (1983) recom mended the following sim ple approach

for the estim ation of constrained m odulus in norm ally consolidated sands.

Mo — 4çc for Çc < 10 M P a (2.9)

Mo = (2çc + 20) for 10 M P a < Çc < 50 M P a (2.10)

Mo = 120 M P a for qc > 50 M P a (2.11)

For Young’s m odulus, the following em pirical form ula is generally used.

JS == gc (2.12)

The ,^-value generally lies between 1 to 2. The small strain shear m odulus can be

evaluated either by using Robertson and C am panella’s (1984) chart or by using Baldi

et al. (1981) chart. B oth the charts are shown in the Figure 2.24.


58

IIIg.2-w.2

TJ

£

40

H K = Kp ® K = Ko

30

20

10

010 200 30

Measured Friction Resistance (kPa)

Figure 2.20: Comparison Between Theoretical and M easured Friction Resistances


59

Con* ResistœKO.Qc (MPa)O 20 30 40 50 GO 70

I— I— ILunnt 9f d Schrrwrmana ——

Q 18 20 30 t o SO 40 70 00 90 100

R E L A T IV E D E N S IT Y D ,tP E R CENT)

Figure 2.21: Schm ertm ann’s M ethod for Estim ating the Friction Angle


60

3U3Oo2

oz«X

<4

>-40zou

2000f-e*LD« •> «i.it»snMORMALLT CONSOLI DATED TICINO SAND

I 5 0 0 -

O MEDIUM D E NS E . D, • 4 6 %♦ DENSE . 0 , • 7 0 %A VENT D E N S E , D, ' $ 0 %

4 DOf 1

o: 1 0 0 0

5 0 0 -

2 D o r s

0 . 5 D or

100 2 0 0 3 0 0 4 0 0

C O N E B E A R I N G , . Dors

5 0 0

Figure 2.22: Evaluation of Constrained Modulus (Robertson and Cam panella, 1984)


61

M' ^ 0 P» ( ÿ r * «XP (Ca Dm)

Qc

M Q TANGENT CONSTRAINED MODULUS _ O ^.M EA N EFFECTIVE STRESS

_________I________________ I ____1 2 5 to

OVERCONSOLIDATION RATIO OCR

Figure 2.23; Evaluation of Constrained Modulus (Jamiolkowski et al., 1988)


62

I---------- 1---------------1------------- --------\ Rang» fo r pluvially deposifed

0 -: / T i c in o sand , from re s o n a n tcolumn, q , from calibration

* * ° P 0 RIVER SAND • GIOIA TAURO

SAND W/GRAVEl

ctiambcr

OCR = 1- OCR = 10

in kPa

200 300 500 1000 2000 3000

^c/ffvo3 0 0 0

b on

2 5 0 04 bor»

(^ 2 0 0 0 -

_tDOo 1 5 0 0 -

q:<

2 borft

I borI

10000 5 bor

3Z

5 0 0

O 100 200 3 0 0 4 0 0 5 0 0

CONE BEARING . . b o r s

Figure 2.24: Evaluation of Small Strain Shear Modulus (Robertson and Cam panella,

1984; Baldi e t al., 1981)


63

2.5.8 State Parameter Interpretation

The s ta te param eter is a quan tita tive m easure of the s ta te of a sand th a t com

bines the effect of void ratio and effective stress in a unique way. T he s ta te param eter,

^ is defined as the void ra tio difference between the current void ra tio and steady

s ta te void ratio a t the sam e stress level as depicted in Figure 2.25 (Been et al., 1986).

S teady s ta te line represents a condition of zero dilation during shear and th is s ta te

reflects a com bination of m any physical properties of sands including compressibility,

grain size shape and distribu tion , lim iting void ratio , m ineralogy and friction angle

a t constan t volume, <j)cv

U ndrained triax ia l tests w ith pore pressure m easurem ents are needed to define

th e steady s ta te line for each sand. Once steady s ta te line is known, the steady sta te

pa ram ete r of a tes t can be determ ined by using the tes t density and the stress level

under which the test is conducted. T he cone test d a ta is related to stress level and

density s ta te as s ta te param eter. Hence, they both can be correlated. Been e t al.,

(1986) p lo tted q c ~ P versus P ' for different s ta te param eter values, where P and

P ' are to ta l and effective octahedral stresses. L ater, the same results are adjusted

by applying a cham ber d iam eter correction factor and they are replotted . Linear

contours of equal are noticed. Hence, the use of a norm alized cone tip resistance in

th e form of 2^ 7 is considered appropriate for the standardized calibration cham ber

da ta . F igure 2.26 presents norm alized tip resistance versus ^ for M onterey sand

d a ta as presented by Been et al., 1986. T he well defined correlation is evident and it

suggests th e relationship is good even when the tests are conducted under sa tu ra ted

and dry conditions, norm al and over consolidated sta tes and at different Ko values.

By using th is figure, ip value can be estim ated if cone test results are known.

Once Ip is known, the friction angle and other param eters can be estim ated from

F igure 2.27.

T he difficulties associated in determ ining or m easuring the void ratio accurately

using th e C P T da ta can be overcome by using th is approach. However, fu rther

work is needed to clarify the influence of fabric, insitu stress strain fields, cham ber

boundary conditions and cem entation on the results. This approach is also adopted

in preparing empirical relationships. Hence, several undrained tests are conducted

on cem ented specimens to define steady s ta te lines.


64

4P— TYPICAL STATE POINT

as

8S

SSL

LOGio (P')

Figure 2.25: Definition of Steady S tate Param eter (Been et al., 1986)


65

0

1 0 0 -

20Q\-roCL

300h

Q(- - P: MPa 20 30

400

- 0 20

Numbers represent t{/, [ i.e. 07 is ^ — 0 07

s? This Study- 0 0 5

5001

Figure 2.26: Normalized Cone Resistance Versus S tate Param eter for M onterey No

0 Sand (Been et al., 1986)


I'S

»ao » ••*44 «♦o*oaaso*o*os

• • <«♦ o * ’ ■

9>.«< o

»0 | ^ 4

B« O

• ô

(p

I ®*

at> a"*

.o »® 0 ® ®

e» «

• o

@

»«

8

I

I

:I

(osaiflap) sfssd^ 33NVi@S3M DmjV3{S jO 3TGNV 03WWO

Figure 2.27: P roperties of Sands Versus S ta te P aram eter (Been et al.. 1986)


67

2.6 Sum m ary

C em entation and their causes are explained. Cem entation phenom ena in differ

ent soils are also described. S ta tic and dynam ic properties of artificially cem ented

sands are presented. This is followed by a section in which cone penetration testing

in calibration cham bers is described. Limited d a ta available on cem ented sands are

presented. Various analysis tools used in the in terp re ta tion of cone penetration tes t

results are discussed.


Chapter 3

METHODOLOGY

3.1 Introduction

This study is conducted in th ree phases. The first phase is the experim ental

m odel and it deals w ith perform ing cone penetration tests in a calibration cham ber

on bo th cem ented and uncem ented sands. C em entation levels of 0, 1 and 2 % are

investigated since these represent lower cem entation levels in na tu ra l deposits. Spec

imens are prepared at th ree different relative densities (45-55, 65-75 and above 85

%) and consolidated under th ree different vertical stresses of 100, 200 and 300 kPa.

D ensity levels represent the possible field densities expected and the effective con

fining pressures correspond to depths of 5, 10 and 15 m of unsatu ra ted or dry sands

or 10, 20 and 30 m of sa tu ra ted sands. All the above variables will cover weakly ce

m ented sands and up to depths of 30 m. Several undrained triax ial and unconfined

compression tests are also conducted on bo th cemented and uncem ented specimens.

These results are used in the proposed prediction scheme developed in th e second

phase.

T he second phase is theoretical assessment. Available theoretical m odels are

used to predict the param eters m easured in the experim ental m odel. Using the pre

dictions of both experim ental and theoretical models, a m ethodology is form alized to

identify cem ented sand deposits and evaluate their engineering characteristics. The

last phase is evaluation of the experim ental results in light of the above correlations.

Semi-empirical predictive m ethods are developed in th is section.

T his chapter covers the first phase, the experim ental model. The inform ation

perta in ing to the equipm ent used, specim en preparation m ethod and testing proce

dures are discussed. T he experim ental work is composed of undrained triax ia l tests,

unconfined compression tests and calibration cham ber tests. Table 3.1 gives the

sum m ary of the calibration cham ber tests. Table 3.2 presents the list of undrained

triax ia l tests conducted. Separate drained triaxial tests are conducted by Arslan

(1993). Table 3.3 provides a sum m ary of these tests.

68


69

Table 3.1: N um ber of Tests (C alibration Cham ber)

B.C C.C. SIGC RELATIVE DENSITY TO TAL45-55 65-75 80 and above

B C l 0 1 0 0 1 1 3 51 1 2 1 42 1 1 1 30 2 0 0 1 1 1 31 1 1 1 32 1 1 1 30 300 1 1 1 31 1 1 1 32 1 1 1 3

BC3 0 , 1 1

BC-Boundary Condition; C .C .-Cem ent C ontent

Table 3.2: N um ber of U ndrained Triaxial Tests

C EM EN T C O N TE N T RELATIVE D ENSITY% 45 - 55 65 - 75 80 and above0 3 3 51 3 3 32 3 3 3

Total 9 9 1 1

3.2 E xperim ental M odel

The calibration cham ber is the testing equipm ent used in experim ental m odel

ing of cone penetration . P luviation setup is associated w ith specim en prepara tion

procedure for the calibration cham ber. O ther equipm ent used in calibration cham

ber testing are the auxiliary system including the control panel, hydraulic system

and supporting equipm ent like cranes and the da ta acquisition and m onitoring sys

tem . O ther testing devices used in experim ental testing are the triax ial system and

unconfined compression testing setup.


70

Table 3.3: D rained Triaxial Tests (Arslan, 1993)

CEM EN T C O N TE N T RELA TIV E DENSITY% 45 - 55 65 - 75 80 and above0 3 3 51 3 3 32 3 3 3

Total 9 9 1 1

3.2.1 Equipment

3.2.1.1 Pluviation Setup

Specim en preparation is an im portan t aspect of testing. The difficulties in ob

tain ing undistu rbed natu ra l sand specimens p rom pt the need to use a technique

th a t sim ulates th e depositional process in na tu ra l deposits. T he laboratory m ethod,

known as ‘p luv iation’ or ‘rain ing’ not only provides homogeneous specimens at the

desired relative density bu t also sim ulates a soil fabric presum ably m ost sim ilar to

the one found in natu ra l deposits form ed by sedim entation (Bellotti e t ah , 1991).

F igure 3.1 shows a schem atic of th e pluviation setup used in the present study.

Basically, th is is a three cham ber setup placed one above the other. All the cham bers

are m ade of Poly Vinyl Chloride sections. The top cham ber stores the sand th a t

needs to be pluviated . The m iddle cham ber gives sufficient height of fall for the

sand leaving the top cham ber. T he bo ttom cham ber is the one in which the sand is

deposited and the specim en is formed. T he bottom cham ber is called th e specim en

cham ber. T he bo ttom cham ber is a diam etrically split cham ber held together by a

m eta l fram e. This cham ber is placed on a wooden trolley w ith four wheels on each

side. I t holds the bo ttom p late through four equally spaced, stainless steel Q m m

(1 /4 in.) bolts.

Since the study involves cem ented sands, the bo ttom cham ber needs to be tran s

ferred first to the hum idity room and then to the calibration chamber. This tra n s

ferring requires th a t the top two cham bers be perm anen t fixtures and the specim en

cham ber be the only removable m em ber of the setup. Hence, the top two cham bers

are fixed on a table. T he size of the tab le is 1 . 8 m x 1.5 m x 1.2 m. This tab le also

provides enough room to place the sand in the top cham ber.


71

\ GUIDE WEIGHT

STRING

9.82 m

TOP CHAMipER ALUMINUM SHUTTER

0.64 m 0.75 mEXTENSION

DIFFUSER SIEVE0.82 mTABLE

1.1 m

SAND

SPECIMEN CHAMBER

0 J m0.3 m

Plate 3.1: Schematic of the Pluviation Setup


72

An alum inum shu tter is placed between the top and m iddle cham ber. This

circular shu tter m echanism consists of a set of two plates w ith an identical hole

p a tte rn . Initially, the two plates are placed one on top of each o ther and are bolted

a t th e center. Handles are welded to this shu tte r system to facilitate th e ro ta tion

in alignm ent of th e holes and in itiation of the pluviation. T he holes in th e two

p lates will align when the handles are brought together and will prevent p luviation

in o ther positions. W hen the holes are aligned, the sand will be released from the

top container, p luviating down to the bo ttom cham ber.

T he diffuser consists of a set of two sieves positioned a t 45 degrees to each o ther

and is placed in the specimen cham ber. Four hooks are placed on top of the diffuser

a t equal distances. Strong threads or m etal cables are connected to these hooks. T he

o ther end of the threads are fastened to the top cap which is placed on top of the

sand in the top cham ber. During pluviation, when sand s ta rts pouring in, the top

cap moves downward. This downward movement of the top cap raises th e diffuser

and th e height of fall of sand from the diffuser is m aintained constant th roughout

specim en preparation. This height of fall influences the density of the specim en. It is

necessary to keep this height constant in pluviation in order to obtain a hom ogeneous

specimen.

3.2.1.2 Saturation Setup

I t is essential to allow w ater access into the specim ens both to sa tu ra te the spec

imens and also to allow the pozzalonic reaction in itia te the necessary cem entation.

A system is developed to sa tu ra te th e cem ented specimens. The setup consisted of

a 50 gallon tank and a carbon dioxide cylinder. The tub ing connections, the w ater

and C O 2 tank and the specim en cham ber can be seen in P la te 3.1. C O 2 is used in

specim en preparation since the C O 2 dissolve in w ater m uch more easily th an air.

The system is connected to the tub ing coming from the bottom plate. Pressures of

very low range (less than 10 kPa) are used since pressures higher than the overbur

den pressure of the sam ple m ay d isrupt particle positions in the specim en and m ay

result in inhomogeneities across the specimen. The tank is placed at an elevation


73

of 0 . 6 m above the bo ttom p la te in an a ttem p t to have low w ater pressures in the

tubing. O ther details of the sa tu ra tion process are presented in the section describing

cem ented specimen preparation.

3.2.1.3 The LSU Calibration Chamber Facility

T he Louisiana S ta te University calibration cham ber system (CALCHAS) was

developed by Dr. Dario de L im a as a part of his dissertation (de L im a and Tumay,

1992). This cham ber perm its testing cones of different sizes and shapes under con

trolled boundary conditions. T he cham ber is 1.78 m high and 0.64 m. in diam eter.

The assembly (Figure 3.2) is divided into two sections nam ely the piston cell and

th e cham ber cell unit. The cham ber cell is a double walled ‘flexible’ cylinder m ade

of steel. This un it rests on a bo ttom plate of 0.64 m in d iam eter and 38.1 m m in

thickness. The piston pushes th e bottom plate upwards thereby applying a vertical

stress on the specimen.

The piston cell bottom p la te carries a bearing shaft th a t houses the piston and

allows vertical movement during testing. The annular spaces in the p iston cell as

sembly are filled w ith deaired w ater. During testing, th is deaired w ater is pressurized

and the inner piston cell moves upwards. The required vertical stress is thus gener

a ted as a result of th is upward th ru st. The sam ple cell is a double-wall flexible cell

which can house a sam ple of 0.53 m in diam eter and 0.79 m in height. T he cham ber

walls are m ade of stainless steel and the thickness of the walls is 6.35 m m . The

in ternal diam eter of the outer and inner walls are 0.58 m and 0.56 m , respectively.

T he sam ple top and bo ttom plates are m ade of 6061 T - 6 alum inum and are of

0.53 m in diam eter. The sam ple bo ttom p late rests on the p iston cell un it. The

sam ple top plate is bolted to the cham ber top p late which is 0.64 m in diam eter and

38.1 m m in height. The cham ber top p late , sam ple cell inner and outer walls and

the piston cell ring are kept together via twelve stainless steel rods. These rods are

tigh tened up to 65 N m. to rque to ensure th a t the whole assem bly does not have

any leaks during the testing.

T he annular space between the sam ple and inner wall and betw een the inner and

outer walls are hereafter nam ed as inner and outer cells. D uring testing, these are


74

Pla.tc 3.1: Photograph of the Saturation Setup


75

0.64 m

0.45 m

0.94 m

0.43 m

0.40 m

CONE PENETROMETER

PO RE PRESSURE MEASUREMENT

INNER CELL WATER LINE

OUTER CELL WATER LINE

► CONNECTING RODS

SAMPLE 20 21/32’X3I 1/16

SAM PLE CELL

PISTON CELL

> ■ MEMBRANE

PISTON CELL \\ ATER LINE

PISTON

Figure 3.2: Schematic of the Calibration Cham ber (de Lima and Tumay, 1992)


76

filled w ith deaired w ater via two w ater lines connected to the top plate. H orizontal

stress is applied to the specimen by pressurizing these inner and outer cells.

3.2.1.4 Saturation and Vacuum Connections

The sa tu ra tion and vacuum lines are m ade of PV C tub ing and are 1.25 cm in

d iam eter. T he tub ing can work under pressures as high as 800 kPa. T he spirally

placed sa turation line is fixed to th e bottom plate. This tub ing has several holes

and is w rapped w ith a screening cloth allowing w ater supply into the sam ple while

preventing the sand to clog the holes. One end of the spiral is closed and th e o ther

end goes through the p late w ith a one way valve a t the end. The valve is connected

to th e sa turation setup during the saturation phase.

T he vacuum tub ing is fixed on the top p la te w ith sim ilar arrangem ents as the

sa tu ra tion tubing. T he outer end of this tub ing is connected to the fem ale end of a

quick connector. This end is connected to the vacuum pum p. Once the vacuum is

applied for a sufiicient tim e, the tubing is disconnected from the pum p. T he suction

pressure inside the cham ber is still in tac t due to the quick connection. T he suction

allows placem ent of inner and outer cells over the specim en and avoids any collapse

in th e specimen.

3.2.1.5 Vacuum Pump

The vacuum pum p used in this study is m anufactured by Welch Company. The

pum p is used in two operations: during specim en preparation and during placem ent

of inner and outer cells over the specimen. This pum p can generate suctions of 60

to 100 kPa.

3.2.1.6 The Miniature Quasi-Static Cone Penetrometer

T he m iniature quasi-static cone penetrom eter (M QSC) is a 1.27 cm? cross-

sectional area subtraction type Fugro-M cClelland cone penetrom eter w ith a 6.3 cm

long friction sleeve and an apex angle of 60° (P la te 3.2) (de Lima and Tumay, 1992).

T he MQSC push rod is 9.53 m m in diam eter and 1.82 rn in length. This cone when

used in CALCHAS, a diam eter ratio of 42 is obtained. Existing d a ta ind icate th a t


77

boundary effects are significantly reduced a t this diam eter ratio even in dense sands.

D etails of the calibration factors used for th is cone are given in Table 3.3.

Table 3.4: C alibration Factors of the Cone

Tip A rea Calibration Sleeve A rea C alibration 1 2

(sleeve loaded) (tip loaded)(cm^) (kg/volts) {cm'^) (kg/volts) ( kg/volts)1.27 2 1 0 25 204 188

3.2.1.7 The Auxiliary equipment

Two movable cranes are built as part of the supporting unit for the calibration

cham ber operations. One crane is electrically operated and the other one is m anually

operated . The capacities of these cranes are 2 and 1 tons respectively. This system

is used bo th in weighing and lifting the sam ple in to the cham ber and also in the

placem ent of the inner and outer cells.

3.2.1.8 Control Panel

Controls which regulate operation of the cham ber are grouped on a vertical

wooden panel of 1.22 m x 1.96 m. Copper tubing was used for all control lines to

reduce volume changes and hence the compressibility in the system . T he water

pressure lines are connected to the top p late through the quick connectors. P la te 3.3

shows a photograph of the controls on the panel board.

T he m ain units in the panel of control are: pressure regulators, electro-pneum atic

transducers, pressure transducers, pressure gauges. The back pressure regulators are

used to provide protection against over pressure in dow nstream portion of pneum atic

system . However, in the present system , these act as relief valves and keep the w ater

pressure constant in the inner cell and piston cell during testing. In cases, where ver

tical and horizontal pressures are kept constant during the penetration phase, these

regulators have an im portan t role. Any rise in pressures due to penetration testing

will be relieved by these regulators. The control panel has two of these units which

operate w ithin the range of 14 to 1070 k P a (2 to 150 psf).

T he electro-pneum atic transducer converts an electric to a linear pneum atic

signal. The cham ber has four of these transducers in the panel, two of them for the


78

A - Mi n i.'i I; ui'L-* (!tnn-i l-’f i u ' I ; n inii . 'I :i ,' r ]■'.- P i ‘.■/•.rici'iK'

Plate' 8.2: Pliotogra.[)li ol the-' Miuiaf uvc ( 'one' (do Lima, I'lDO)


TU

CONE P E N E IR G M E T E R

CH UCKING SYSTE M

J AC K

P ORE P R E S S U R E MEASUREMENT

WATER i n G W . L Ü N S

C n H t ’R r S S E D A I RCjaf]

r r i HPRf !

AIR f i l . T U

rii AIR- - 11^0 -

SAMPL E C ELL

P I S T O N CEI L

A' 3 - W A Y BALL V A L V FR- O N - o r r HALL v a i v e

D UU lC K- C nN N KC T OR BP'F A IR C MI LU RACK P R E S S U R E R E G U LA T UR ( R P l . ? . P - I 5 0 P S ] >F ' F A IR C H IL D L / P TR A NSD UCE R (I 1 .3 ' F>SI| F ? , 4 . ;i - lPO P S I )G' MARSH P R O C ES S P R E S S U R E GAUGE ( G , 0 - 100 P S I )S' S E N S Y M t r a n s d u c e r ( S l . 2 . 4 ' 0 - 3 0 P S I ) S 3 . 5 ' 0 100 P S I )V . D L C E D E R 'P L U i T V A L V E

n n U N D A P Y CÜNDI

ciw :ifcw î VI011CAJ

rinu J

Plate 3.3: Photograph of the Controls ou the Panel Board


80

pressure range of 40 to 215 k P a (5.6 to 30 psi) and the o ther two for the range of

21.5 to 860 k P a (3 to 120 psi). A DC signal of 0 to 10 F is generated. Two of

th e transducers are used in the piston cell operation for applying the vertical stress

to the sample. T he other two are used for the pressure com pensation betw een the

sam ple inner and outer cells during Ko consolidation and penetration phase.

T he cham ber also houses five pressure transducers in the range of 0 to 215 k P a

(0 to 30 psi) and 0 to 714 k P a (0 to 100 psi). Two of them are connected to the

w ater line related to piston cell, two others are connected to the w ater line d irected

to inner cell and one in the range of 0 to 215 k P a (0 to 30 psi) is connected to the

ou ter cell water line. Five m arsh process gauges are used in the panel. T hey work in

th e range of 0 to 714 k P a (0 to 100 p si) w ith an accuracy of ± 0 .5 %. For applying

th e pressures (vertical and horizontal pressures), the air w ater interface system is

used. Two PV C cylinders and their caps glued at high pressures are used to apply

th e above pressures. The cylinders are filled with w ater and air in a 90 % to 10 %

proportion w ith an oil interface.

3.2.1.9 Hydraulic System

Hydraulics and the push jack system allow penetration of the cone in a single

stroke. The m axim um stroke is 0.79 m; however, in th is study, the cone is pushed

in two strokes since it was found th a t a single stroke sometimes buckles the cone in

dense specimens. Even when sufficient grip length is provided, cone showed buckling

in the 80 % relative density specimen consolidated under 300 kPa. In view of the

cost, im portance and the non availability of such an equipm ent, tests are perform ed

by pushing the cone in e ither two or th ree strokes.

An analog to digital converter depth decoding system is developed and incorpo

ra ted in this system . The depth decoder is composed of a m etal disk, a light em itting

diode and an optical sensor. Holes are drilled a t equal distances on the circumference

of the disk. As the cone penetrates the specimen, a cable connected from the drill

rod to the shaft of the disk mechanically turns the disk. The distance betw een the

holes on the disk represents a penetration depth of 2 cm. T he light em itting diode

and th e optical sensor are placed on either side of the disk. W hen light em itted by

th e diode passes a hole, the optical sensor senses the light and generates a pulse to


81

th e control un it. This triggers the m ultiplexer to switch on the channels for analog

to digital conversion. This process continues until penetration m otion is stopped.

3.2.1.10 Data Acquisition and Monitoring System

The hardw are used in the d a ta acquisition process consists of Zenith PC m icro

com puter w ith 640 K RAM , 40 MB hard drive, EGA m onitor, a da ta translation

D T 2801 A /D board and a HP7475 p lo tter. The flow chart for da ta acquisition and

m onitoring system is depicted in Figure 3.3.

T he software for specimen consolidation and testing a t different boundary condi

tions is developed in Turbo Pascal version 4.0 environm ent by Borland International,

and the H alo ’8 8 graphics library by M edia Cybernetics (de Lima, 1990). The da ta

tran sla tion board is used for analog to digital conversions(A /D ), digital to analog

conversions (D /A ) and for performing digital I /O transfers. This board has sixteen

12 b it A /D channels and two D /A channels.

Six A /D channels receive the da ta in volts from three transducers (piston cell,

inner and ou ter cell), one LVDT and tip and friction load cells in the cone. Two

D /A channels send the d a ta in volts to two electro-pneum atic transducers. The depth

sensor sends signals to digital I /O transfer. T he calibration of the electro-pneum atic

transducers is done using the commercial software nam ed Labtech Notebook.

The da ta acquisition software includes five com puter program s, one for consoli

dation phase and the rest for the penetration phase. The nam es of these program s

are: CHAMBKO.EXE for consolidation and CHAM BC1.EXE, CHAM BC2.EXE,

CH AM BC3.EXE, CHAM BC4.EXE for the penetration phase. These program s orig

inally w ritten by de Lim a (1990) are modified by the au thor to suit the present

testing. T he details of the modifications associated w ith the boundary conditions

are discussed in the section pertaining to the penetration.

T he d a ta are acquired employing the following procedure. The initial readings

in voltages are taken during the consolidation stage. The d a ta are read and stored in

th e com puter a t every 2 cm depth interval during penetration. The initial readings

are then sub trac ted to calculate the m easured values. Finally, the da ta is sent to a

p lo tter for graphing.


82

a:

3CL.

S8

êl

<

eq

2 “

u

zgCO

e n

en

<Q

Ü\ f -

Figure 3.3: Flow C hart for D ata Acquisition and M onitoring System


83

3.2.1.11 The Triaxial System

T he triax ial system (m anufactured by W ykeham Farrance, UK) a t Louisiana

T ransportation Research C enter (LTRC) geotechnical laboratories is used in the

un drained triax ia l tests. Samples of 7 cm (2.8 in.) in diam eter and 15.2 to 16.5 cm

( 6 to 6.5 in.) in height are used. Cell pressures of 428 k P a (60 psi) can be applied

to th e specimens. Backpressure sa tu ra tion is used. Backpressures of order 428 k P a

(60 psi) can be applied.

This equipm ent is connected to a d a ta logger (A utotech) and a com puter. T he

d a ta logger stores the readings from the load cell (deviatoric stress, channel 7),

s tra in transducer (axial stra in , channel 8 ) and a pore pressure transducer (excess

pore pressure, channel 9). T he frequency of reading the d a ta can be varied. A five

second interval is chosen for the present testing. The tests are conducted under stra in

controlled conditions.

A commercial software CLISP is used to analyze the da ta for each test. This

software takes the inpu t of sample inform ation, sam ple dimensions, details about

back pressures and cell pressure. T he final ou tpu t from the program consists of axial

stra in , axial load, pore w ater pressure, excess pore pressure and deviatoric stress.

D rained test are conducted in a triax ial appara tus designed and fabricated by

Trautw ein Soil Testing, USA. An ELE volum etric transducer is connected to this

setup to record volume changes in the sam ple during shear. Cell and back pressures

are applied through a control panel. D uring the te s t, LVDT, load cell and volume

change transducer readings are recorded.

3.2.1.12 Unconfined Compression Tests

This m achine is m anufactured by ELE, USA. T he unconfined compression setup

allows testing of specimens of different sizes up to 5 cm in diam eter. Testing can be

done a t different speeds varying from 2 to 9 Dial gauge and load cell readings are

taken at constant intervals. The height and d iam eter of the specimen are recorded

and are used in the calculation of strains and unconfined pressures.


84

3.2.2 Procedure - Uncemented Specimens

The uncem ented specimen preparation and testing procedures are described in

the following sections.

3.2.2.1 Triaxial Tests

Laboratory tests consisting of strain controlled consolidated undrained triaxial

tests were conducted using M onterey No. 0 /30 sand. This sand is a commercial beach

washed sand. The grain size d istribution of this sand is shown in Figure 3.4. The

index properties are presented in Table 3.4. The m ain objective of th is testing was

to determ ine the strength-deform ation characteristics, and critical s ta te behavior of

th is sand and assess w hether there were significant differences in the results reported

for th is sand by various investigators.

Table 3.5: Index P roperties of M onterey No. 0/30 Sand

PR O PER TIES RAD (1982) EL-TAHIR (1985)G, 2 T# 2 T#

' ) 'd , m a j : (A A'/y/r ) 16.65 1&6514.04 14.04

^ m a x 0.85 0.85^ m i n 0.56 0.56a 1.43 1.50a NA OT^

£>50 (mm) O J^ 0.43

i) Specimen PreparationThe following steps describe the specim en preparation adopted in the jjresent

study.

1. Dry M onterey sand No. 0/30 needed to achieve the desired relative density was

weighed and placed in equal proportions in six beakers. T he height deposited

during pluviation of the specimen in each beaker would provide an idea ol the

homogeneity.

2. T he pore pressure and back pressure lines in the triaxial system are sa tu ra ted

w ith deaired water. Any visible air bubbles in the w ater inside the lines are

flushed out.


85

1 0 0

80

w Z :

^ 60H%WOK 40W(I4

20Present Study

□QDQD Acar and El-Tahir ( 1 9 8 6 )

10 1 0 00.10.01GRAIN SIZE (m m )

Figure 3.4; Grain Size D istribution of Monterey Sand


86

3. A porous disc is placed over th e bo ttom p late of the triaxial cell. This bo ttom

plate includes a drainage valve which is used in vacuum application.

4. A m em brane is placed over th e disc and th e bottom plate. An 0 -rin g is placed

around the m em brane on the bo ttom plate.

5. -A split mold is placed over th e m em brane. T he m em brane is pulled up and is

placed over the mold.

6 . Sand from each beaker is poured into a funnel which is directly placed over the

mold. T he height of fall, defined here as the d istance between the bottom of the

funnel to the bottom of the soil layer deposited in the mold, is kept constant

throughout the testing in an a tte m p t to assure hom ogeneity throughout the

specimen.

7. W hen all the sand in the beaker is poured, the mold is tapped along the sides

so th a t the sand settles across the top uniformly.

8 . The height of the specimen in the mold is m easured with a vernier caliper and

relative density is calculated. T he height of fall and the am ount of tapping

necessary is varied to achieve different densities.

9. A porous stone is then placed over the to]) of t he sand. The top cap with back

pressure line is placed over th e porous stone.

10. The m em brane is pulled up and the 0 -rings are placed around the top cap.

1 1 . A vacuum of 30 k P a to 40 k P a is applied to the specim en through the drainage

valve in order to supply the necessary confinem ent to hold the specimen to

gether when the mold is removed.

12. The outer cell is placed, bolted , tightened and filled with deaired water and a

small am ount of cell pressure of order 35 kPa is applied.

13. The back pressure valve is opened to allow flow of w ater in the specim en. It is

left open until w ater flows through the drainage valve to the vacuum pum p.


87

14. T he back pressure valve is closed and a cell pressure of 80 k P a is applied. At

th is stage, the drainage valve is closed and vacuum is disconnected.

15. Back pressure of 50 k P a is applied and th is value is always kept lower than the

cell pressure.

16. T he pore pressure m easured was always higher than the applied back pressure.

This excess pressure is due to th e confining pressure applied in the cell.

17. T he drainage valve is opened slightly to release some more air from the speci

m en and then closed.

18. This pore pressure will stabilize back to the applied back pressure after a certain

period of tim e. T he excess pore pressure due to the cell pressure was dissipated

due to opening of the drainage valve. Hence, the pressure inside the specimen

should be same as the applied back pressure. T he cell pressure and back

pressures are raised increm entally. T he above steps are repeated un til either

the desired back pressure is applied or the full sa tu ra tion condition is achieved.

19. Satu ration is m easured by calculating the Skem pton’s B param eter. This pa

ram eter is defined as the ra tio of the increm ent in the back pressure to an

increase in cell pressure ( ^ ) . If the ratio is equal to or greater than 0.90, then

an acceptable sa tu ra tion is reached (Bishop and Henkel, 1962).

20. Sometimes the specim en is left overnight to ensure com plete sa tu ra tion . The

B values of present tests were around 0.93 to 0.97, suggesting th a t fairly high

values of saturation are achieved.

Testing ProcedureThe following steps are used in the triax ial testing.

1. T he back pressure a t the end of sa turation is decreased or cell pressure is

increased in order to reach th e desired effective stress.

2. T he loading fram e assembly is lowered such th a t the assembly ju st touches the

specim en. The fram e is leveled and adjusted w ith set screws in order to achieve

a level top plate.


88

3. T he load ring is zeroed and the LVDT is positioned to m easure the displace

m ents.

4. The triax ial cell is then pushed upwards a t a constant displacem ent rate of

0 0 5 ^ -

5. The displacem ent, load, and pore pressures are au tom atically recorded in the

d a ta logger a t every 5 seconds intervals.

6 . The test is stopped when the load on the ring rem arks constant for a period of

tim e or it decreases and shows no further signs of decrease.

7. The software CLISP is used to in teract with d a ta logger. T he raw data are

reduced through th is software to calculate the axial stra in , deviatoric stress

and excess pore pressures.

8 . These results are stored in Lotus 123 files and the d a ta is used to calculate the

needed to tal and effective stress param eters.

9. These final results are then plotted by using the G rapher software.

10. W hen the test is com pleted, the triaxial cell is drained and cleaned for the next,

test.

Tests are conducted at different relative densities and different confining pressures.

3 .2 .2 .2 C a l ib r a t io n C h a m b e r T e s tin g

i) S p e c im e n P r e p a r a t i o n

Cone penetration testing conducted on uncem ented sands by several investiga

tors reach the sim ilar conclusion th a t testing under dry or full saturation has m inor

or no influence on the cone tip resistance (Schm ertrnann, 1976; Tumay, 1976; Baldi.

1981; Bellotti, e t al., 1985). Hence, uncem ented specim ens are tested under dry

conditions. 'J'his allowed the au thor to reuse the tested sand for a lew more tests,

provided there was no significant crushing around the cone. Grain size d istributions

conducted on the sand collected around this m iniature cone revealed th a t there is

no significant crushing of the sand grains.


89

Specimens are prepared by employing the following procedure:

1. T he shu tte r is ro ta ted to a position where the holes are misaligned and the

p lates prevent pluviation. T he dry sand is then placed inside the top container.

2. T he top cap w ith the guiding weight is placed above the sand.

3. T he m em brane is placed around the bottom plate and 0 -rings are placed

around the specim en to achieve air tight conditions.

4. Vacuum is applied outside the cham ber so th a t the m em brane stretches around

the cham ber walls. This vacuum is m aintained throughout specimen p repara

tion.

5. The specimen cham ber is then rolled underneath the table. The cham ber is

adjusted so th a t it aligns stra igh t below the top two cham bers.

6 . T he diffuser is placed inside the specimen cham ber and all four th reads or m etal

cables from the diffuser are connected to the topi cap.

7. The shu tte r is ro ta ted to align the holes and sand is pluviated.

8 . During pluviation, the level of the sand in the top cham ber goes down and the

top p late moves downward. T he downward movement of the top p late raises

the diffuser by an equivalent d istance keeping the falling height constant.

9. W hen the sand is com pletely rained out of the top cham ber, the specimen

cham ber is carefully rolled out for weighing.

10. The ex tra sand on the top is carefully removed by a sm ooth straight edge.

11. The height of the specim en is m easured.

12. The cham ber is then weighed to calculate an average relative density for the

specimen.

T he guidelines suggested by Rad and Tum ay (1987) are followed in selecting

the pluviation variables. The im portan t variables th a t affect the specim en rela

tive density in pluviation are shu tte r porosity, height of sand fall and diffuser sieve


90

size. Two shu tters of porosities 6 and 15 % and th ree different sieves (ASTM ^ ’s

I in . (12.5 ??z77r), |z n .(9 .5 'm m ) and ^ i n . (6 .3 m m ) ) are used. T he height of lall is

varied between 5 to 15 cm.

For higher relative densities, a sh u tte r porosity of 6.5 % and a sieve opening

of 6 . 3 mm. are selected. For 65 to 75 % relative density, the sh u tte r with th e 6.5 %

porosity and a sieve opening of 9.5 m m are used. For lower relative densities (45 to

55 %), the sh u tte r w ith the 15 % porosity and sieves with openings of 12.5 or

9.5 m m are used.

Several specim ens (10) are prepared using the sam e sieve opening, sh u tte r and

height of fall in order to evaluate the repeatab ility of the specimen preparation. I 'h e

variation in relative density was found to be ± 5 % in the case of 45 to 55 % range. In

o ther cases i.e. when preparing 65 - 75 % or above 85 % relative density specim ens,

the variation is around ± 7 %. This shows th a t reasonably repeatab le specim ens can

be prepared using th is technique.

ii) Testing ProcedureThe cham ber operation is split into 3 m ajor operations.

1. Specimen transfer and placem ent

2 . Consolidation

3. Cone penetration testing

1. Specimen Transfer and PlacementSpecimen transfer and placem ent procedures are an im portan t com ponent ol

the testing procedure. Specimens may collapse when the cham ber molds are sepa

rated . This is due to cohesionless na tu re of this m aterial which does not exhibit any

streng th when there is no confinem ent. T he following steps are taken in an a ttem p t

to minimize d isturbance during specim en transfer and placem ent.

1 . The specim en cham ber w ith the specim en is lifted on to the calibration cham ber

following pluviation and it is placed carefully on the piston.

2. T he top p late is placed over the specim en. 0 -rings are placed around the top

p late covering the m em brane.


91

3. A I in. tubing with fem ale end of the quick connector is connected to the top

p late on the specimen and the o ther end is connected to the m ale end of the

tub ing from the vacuum pum p (P late 3.4).

4. A suction is applied to the specimen for approxim ately one hour and then the

tu b e is disconnected from the pum p by separating the quick connector.

5. T he specim en and the cham ber are then carefully lifted w ithout any tilt and

it is placed on the piston assembly.

6 . T he diam etrically split molds of the specim en cham ber are then removed. T he

specim en stands firmly w ithout any buckling due to the confinement induced by

th e suction pressure (P la te 3.5). The inner and outer cham bers are im m ediately

lowered and placed over the specimen.

7. T he vacuum rem ains the same w ithout any loss, provided there are no holes in

the m em brane.

8 . T he inner cham ber is lowered using the m anually operated crane. Any slight

tilt in the cham ber position will strip the 0 -rings on the top p late causing

the vacuum in the specim en to be released and resulting in the collapse of the

specimen. These problem s are experienced in the first few experim ents.

9. Upon placem ent of the inner and outer cells, th e outer m ost top p late is con

nected to the top p late via twelve bolts of size 12.5 m m . The outer top plate

is later connected to the piston assembly through twelve equally spaced rods

as shown in P la te 3.6. These rods are subjected to a to rque of 65 k N m. This

to rque to each rod will ensure th a t th e whole system is tightly and uniformly

assembled to avoid any leaks during testing. T he leaks in the system can be

detected only when the deaired w ater in ou ter cell is pressurized during con

solidation.

10. The inner and outer cells are filled w ith deaired w ater once the specim en is

placed and all connections are m ade. This is accom plished by directing the

w ater from the container by opening the valves on the control board.


92

Plate 3.4; Applying Va,cnnm Inside the Specimen


93

Plate 3.5: Specimen After Uiilolding l.hc Split Molds


94

Plate 3.6: Final Asscmhly of the Specimen


95

2. ConsolidationT he next step is consolidating the specimen under a given vertical stress. This

is accom plished through the da ta acquisition and reduction software developed by

de L im a (1990). This software handles the following steps:

1. T he inpu t values like inform ation about the soil sam ple and the vertical con

solidation stress is read.

2. T he system acquires d a ta from eight electric transducers via analog to digital

conversion and operates two transducers via digital to analog conversion.

3. I t also displays graphically the real-tim e variation of stresses and vertical dis

placem ent.

T he specimens are consolidated under Ko conditions. T he Ko conditions require

zero la tera l strain . A set of procedure is followed during the process. W hen the

specim en is ready for consolidation, th e program CHAMBKO.EXE is to be executed.

This can be done by typing CHAMBKO and entering it. T he program takes inform a

tion about the soil specim en and inquires about the electro-pneum atic transducers

(5.6 to 30 psi or 3 to 120 psi) and th e pressure transducer ranges (0 to 30 psi or 0

to 1 0 0 psi) to he used.

The following steps are then followed in consolidation:

1. File nam e, details about the specimen, selected ranges of pressure transduc

ers, a Y E S/N O option for applying equilibrium pressure, selection of electro

pneum atic transducer range and the value of consolidation stress are inpu tted .

2. T he equilibrium pressure is the vertical pressure th a t has to be applied to the

inner p iston cell w ater in order to equilibrate the load generated by the weight

of the specim en, specim en bottom p late and m em brane. Any pressure higher

than this equilibrium pressure will cause the piston move upwards in itia ting a

consolidation. For YES option, a default pressure of 3 psi is applied prior to

the consolidation. For the NO option, the user needs to supply this value via

th e control panel.

3. In the NO option, the program reads the equilibrium pressure via the com puter

from 84 or 85 transducer. This can be done by directing the compressed air


96

flow from the air filter through transducer F I while keeping valves B13 and B16

open. W hen the specim en equilibrium pressure is reached, and the specimen

touches the top plate, the user should press the keyboard corresponding to

answer NO. Subsequently, valves B3, 1313 and B16 should be closed, and the

compressed air flow from the air filter should be directed to the E /P transducer

F 1 /F 2 chosen for testing.

4. At this stage, the program takes all zero readings of the transducers, the tip

and friction load cells and the LVDT. Once the valves B lO /B ll and B16/B17

are closed and the valves B 6 and B13 are opened on the control board, the

com puter takes zero readings. Im m ediately after the readings are displayed on

the m onitor, the operator should open valves B lO /B ll and B16/B17 and dose

valves B6 and B13.

5. C om puter flashes on the screen with the graphics featuring vertical deform ation

w ith tim e and vertical stress versus the horizontal stress.

6 . The program will then read the final consolidation vertical stress and it will

generate the necessary input to apply this stress on the specimen in increm ents.

The first increm ent is calculated in the program as voltage. T he program

sends this voltage to the transducer F1 /F 2 where it is converted to pneum atic

pressure. This pressure will m ake the piston move up thereby exerting a vertical

pressure on the specimen.

7. W hen the vertical stress is applied, the specimen will expand horizontally.

This will create a certain pressure in deaired water in the inner cell. The

program reads this inner pressure and then sends a signal in volts ecjuivalent

to com pensate for the inner pressure to the transducer F3/PA.

8 . T he electro-pneum atic transducer, F3 /F 4 will convert the voltage into ]>ressure

and the pressure will be applied to the deaired water in the outer cell. Since

the pressure in the outer and the inner cells are the same, there will be no

lateral movement and horizontal strain in the specimen.


97

9. T he o ther increm ents are also applied identically until th e final vertical stress is

reached. This process is known as Ko consolidation since any lateral movement

is not allowed.

The fundam ental problem s associated with th is phase is emergence of leaks in the

cham ber. These leaks are found due to im proper tigh ten ing of the rods surrounding

the cham ber and also due to the im proper tightening of th e screw bolts on the top

p late . Any small gaps around these bolts will leak the w ater in the inner and the

ou ter cells releasing the horizontal pressure.

3. C o n e P e n e t r a t i o n T e s tin g

T he final phase consists of conducting cone penetration testing a t two or th ree

different locations. Cone penetration testing is conducted under zero lateral stra in

and constan t vertical stress. The program , CHAM BC3 is used to in itia te the process;

however the user has to close valves B 6 and BIT on the panel of control before doing

so. T he following steps are used in th is phase;

1. T he file nam e for zero readings input, i.e. th e file nam e for consolidation phase

is to be entered,

2. The file nam e for d a ta storage is to be given,

3. Laboratory testing inform ation needed should be entered,

4. P rogram provides the option to change the display of the settings for different

sounding depths,

5. Pressure transducer range (one used in consolidation phase) is to be selected,

t). T he electro-pneum atic transducer used in the consolidation phase is to be se

lected.

Upon com pletion of the above steps, the com puter screen flashes with the graphics

which include the tip resistance, friction resistance, vertical stress, inner and outer

horizontal stress versus depth . At th is stage, penetration should be in itiated . The

steps involved in penetration are given below.

1. Prior to executing the program , it is necessary to lift the hydraulic jack with

the help of the crane and place it above the top ])late.


98

2. T he jack is then clam ped to the top p late a t one of the four testing holes on

the top p late w ith four equally spaced bolts. T he jack which is in the collapse

stage is raised to its original position.

3. T he m in iatu re cone is placed such th a t it ju s t touches th e specim en and at th is

position, th e cone is fixed to the clamp on the jack.

4. Once the graphics screen is ready and flashing on the com puter m onitor, the

cone is pushed into the specimen a t a ra te of 2 — .

5. T he readings are recorded a t every 2 cm intervals in the program . T he test can

be term ina ted by pressing any key.

6. W hen the test is completed, the jack is moved to ano ther location and the

above steps are repeated in conducting another test.

7. Upon com pletion of the tests, the cham ber is cleaned and is kept ready for the

next test.

T he cham ber assem bly and specimen preparation takes about 60 to 90 m inutes. T he

application of vacuum and lifting the specim en into the cham ber takes a t least 2 to

2 1 /2 hours. Filling the inner and outer cells w ith deaired w ater takes around 30

and 40 m inutes respectively. Consolidation process requires around 15 m inutes. T he

to ta l tim e for conducting two penetration tests on the specim en is abou t 5 hours.

D ism antling and cleaning the cham ber requires another 2 hours, to ta ling about 12

hours per test.

T he d a ta reduction pertain ing to th is test consist of two phases. T he da ta from

th e te s t are first processed and then p lo tted on a plotter.

3 .2 .3 P r o c e d u r e - C e m e n te d S p e c im e n s

3 .2 .3 .1 T r ia x ia l T e s ts

S p e c im e n P r e p a r a t io n

T he specim en preparation procedure is sim ilar to th a t of uncem ented specimens.

T he only difference is th a t a cem ent/sand m ixture is used in place of only the sand.

T he cem ent used in th is study is ordinary portland cem ent. T he cem ent sand m ixture

is prepared as follows:


99

1. T he specim en m old m ade of plexiglas with m em brane inside is placed on the

bo ttom p late (P late 3.7). These molds are 7.11 cm (2.8 in .) in d iam eter and

15 cm (6 in .) in height.

2. T he bo ttom p late has four equally spaced holes for allowing w ater in to the

specim en. These holes are covered with a filter paper prior to p luviation.

3. 1000 g m of dry sand is weighed and placed in the bowl.

4. For samples with 1 or 2 % (of the dry weight of sand) cementation, 10 or 20

g m of Portland cement is weighed respectively.

5. 4 g m of w ater (0.4 % by the dry weight) is sprayed on the sand and is stirred

w ith a hand mixer. This creates a slight m oisture on the surface of the sand

which allows an even d istribu tion of the cement particles upon mixing.

6. T he cement is added gradually during mixing. The mixing is continued for 60

seconds.

7. T he m ixer is stopped and fu rther mixing is continued w ith hand. This gives a

b e tte r coating of the sand particles by cement.

8. This m ixture known as cement sand is used in specimen preparation.

9. T he specim en is equally d istribu ted in six beakers. The rest of the steps in

the specim en preparation are sim ilar to those described in p reparation of un

cem ented specimens.

10. T he specimen together w ith plexiglas sleeve is weighed to calculate th e relative

density.

11. T he specimen in the plexiglas container is then placed on a bed of sand in a

bucket.

12. W ater is introduced into the bucket so th a t its level gradually rises (approxi

m ately 0.5 mm. per m inute). The sample is completely subm erged underw ater

in approxim ately 5 to 6 hours.


t o o

yp*

Plate 3.7: Plexiglas Molds for Triaxial Specimens

Reproduced witli permission of the copyright owner. Further reproduction prohibited without permission.

101

13. T he bucket containing the sam ple is transferred to the hum idity room for cur

ing.

A curing period of 7 days is selected in this study. This curing period is enough to

form cem ented bonds th a t are required for na tu ra l cem ented deposits while it also

allows p reparation of sufficient num ber of specimens in a short period of tim e. The

scanning electron m icrographs of the cemented and uncem ented sands are shown in

th e P la te 3.8. T he crystallization process th a t m ight have taken place during curing

is depicted in th is plate.

Testing ProcedureT he undrained testing is sim ilar to th a t of uncem ented specimens. The only

difference is th a t higher back pressures are needed during the testing since cemented

sands d ila te m ore than uncem ented sands. Samples were fully sa tu ra ted prior to

each tes t. T he deviatoric stress, axial stra in and the excess pore pressure build up

axe recorded.

3.2.3.2 Calibration Chamber Testing

Specimen PreparationThe following procedure is used in cemented specimen preparation for the cali

b ra tion cham ber study:

1. 13500 gm of dry sand is weighed and placed in a plastic bucket.

2. 135 (1 % cem entation) or 270 (2 %) g m of cement is added to the sand.

3. This dry m ix ture is placed in a m ixer and the container is transferred under

the mixer.

4. P rior to mixing, 54 gm. of water is added and mixed with hand.

5. T he m ixing is sta rted and is continued for about 120 seconds.

6. T he coated cement on mixer blades are wiped off into the sam ple and the

m ixing is continued for another 120 seconds.

7. T he cem ent sand mix is carefully transferred in to the top cham ber in the

p luviation setup.


il)-2

124%200um

15kV Wb 7mm 5 cem02 P 00003

Phite 3.8: S c a n n i n g Kleclron iVlicrograph.s of Uiict'nic'iilcd and ( 'cmrnk 'd (0 and 2

%) sand


103

8. The above steps are repeated until the top cham ber in the pluviation setup is

filled.

9. T he rest of the process is sim ilar to th a t used in uncem ented sands.

10. W hen pluviation is com pleted, small am ount of cement is noticed left in the

top specim en cham ber. This segregated cement is often 3 to 4 % of th e to ta l

cem ent added in each bucket (4 to 6 gm out of 135 gm cem ent). Hence, i t ’s

effect is neglected.

11. The specim en cham ber w ith cem ented sand m ix is weighed to calculate an

average relative density.

12. T he specim en is moved and it is connected to the sa turation tank (P la te 3.9).

13. The carbon dioxide is allowed into the specimen from the bo ttom replacing the

air in the specimen.

14. T he w ater from the tank is allowed to enier into the specimen.

15. W hen the specimen is com pletely subm erged, the specimen is transferred to a

hum idity room for curing.

The specim ens are checked for homogeneity and cem entation. It is assum ed th a t

hom ogeneity is achieved since pluviation studies on uncem ented specim ens prove

th a t reasonably homogeneous, repeatable specimens can be prepared. However, the

m ain concern is w hether the pluviation process preserves the cem entation bonds

form ed at the tim e of specim en preparation. Hence, samples are collected at four

different depths, 10, 25, 40 and 50 cm. The relative densities are calculated and

are p lo tted in Figure 3.5. T he specimen is reasonably homogeneous, while s tandard

deviations are of the order of 2 to 3 %. Scanning electron m icrographs for these

samples suggest th a t the pluviation process did not result in segregation and the

cem ented bonds exist between sand grains (P lates 3.10 and 3.11). Cem ent m aterial

around the sand particles can be seen at all depths. Subsequent to curing, the

specim ens are transferred to the calibration cham ber for the testing.


104

Plaie .4.9; Specimen UndeipjutnK SaUiratioii Procc'.ss


105

4 0Relative Density, Dr ( % )

60 8 0 1 0 0

i 20

aQ)

Q

4 0

Std. Dev.

o æ æ 68.7 2.9B B B B B 8 6 .0 2 .3

6 0

Figure 3.5: Relative Density Versus Depth in Cemented Specimens


1.06

10kU WD:31mm S : c 0 2 0 8 P 0 0 0 0 7

10kV WD:28mm S;c0220 P,00010

:.'.:73V

# # # & ) .......

Plaie O.IO: Scanning Electron Micrographs ol (ieinented Specimen at 10 and 2n cm


1Ü7

i 5kU. N D : 8 n r S = c 0 2 6 0 P = 0 0 0 0 5

gPlaie 0.11: Scanning Mleclroii iVlicrograplis ol (VnK'iiied Six'cimeiis ai 10 and aO cni


108

Testing ProcedureTesting procedure is sim ilar to th a t described in uncem ented sands. Cem en

ta tio n induces some cohesion. However, th is cohesion is not sufficient to provide

th e necessary confinement during transfer and placem ent of inner and ou ter cells.

Hence, application of vacuum is found necessary before rem oving the m old. The

vacuum pressure ranged betw een 60 to 70 kP a . This initial vacuum will induce

a pre-consolidating effect on the specimen; however these specimens are la te r con

solidated to stresses higher than 100 kP a and hence they are considered norm ally

consolidated.

3.2.3.3 Unconfined Compression Test

Specimen preparation in unconfined compression tests is sim ilar to those prac

ticed used in preparing triaxial specimens. The only difference is the size of the

specim en. Plexiglas molds of 7.6 cm d iam eter and 23 cm height are used in these

tests.

It is well known th a t curing period increases the compressive strength . The

effect of curing periods of 3, 7, 14 and 28 days on unconfined compressive streng th

is investigated.

3.3 Sum m ary

Specimen preparation and testing procedures are described. Specimen p repara

tion procedure did not result in excessive segregation of cement and the cem ented

bonds are observed to be intact. The vacuum application inside the specim en prior

to transferring into the chamber is found necessary and this procedure is necessary

even in the case of cemented specimens. Curing period of 7 days and 1 % to 2 %

cem ent contents are probably enough to sim ulate the very weakly cem ented natu ral

deposits.


Chapter 4

ENGINEERING BEHAVIOR OF MONTEREY

NO 0/30 SAND

4.1 Introduction

T he results of consolidated drained, consolidated undrained triax ial compression

tests and unconfined compression tests are presented. Strength-deform ation behavior

of M onterey No. 0/30 sand is evaluated in comparison w ith previously reported

studies. A theoretical model is used to model the behavior of this sand under drained

and undrained testing conditions. These m odeling param eters are later used in the

m odeling of cone penetration testing.

The tests are coded in the following m anner: MSC0CU45, where MS stands

for M onterey Sand No. 0/30; CO stands for cem entation level or cement content

(0, 1 and 2); CU stands for isotropic consolidated undrained triaxial test (CD is

consolidated drained test, UC is unconfined compression test) and 45 represents the

relative density of the tested specimen. Physical characteristics of the sand and

cem ent are presented in C hapter 3.

T he testing program consisted of conducting 30 undrained triax ial tests on un

cem ented and artificially cem ented sands. Cem entation levels of 0, 1 and 2 % and

th ree ranges of relative density (45 to 55, 65 to 75 and above 85 %) are used. The

tests are conducted at confining stresses of 100, 200 and 300 kP a . The objectives of

th is testing program are established as;

1. to define the strength-deform ation param eters and to com pare them w ith the

results reported by previous studies (Rad, 1984) in an a tte m p t to assess re

peatab ility and to evaluate w hether previously reported results can be used in

current evaluations,

2. to provide a basis for correlating penetration param eters w ith the strength

param eters,

109


110

3. to develop critical s ta te lines for th is sand (cem ented and uncem ented) and to

use them in the critical s ta te in terp reta tion of the cone penetration test results,

4. to obtain th e necessary theoretical modeling param eters such as dilational angle

and shear m odulus and to use them in m odeling cone penetration w ith the

cavity expansion theory.

A separate study of th e effect of curing on unconfined compression strength is also

in itia ted . Unconfined compression strength in cem ented sands provides an indi

rect m easure of the cohesion intercept. The cone penetration tests in the calibration

cham ber are conducted w ith 7 days cured specimens. Specimen placem ent and tra n s

fer in calibration cham bers often took several hours (8 to 10 hrs). T he concern then

arises th a t slight variations in curing may affect the strength-deform ation behavior

and hence the tip and friction resistances. The unconfined compression streng th

study provides an evaluation of the relative effect of such changes. Furtherm ore, th is

study renders a qualita tive evaluation of the expected streng th increase.

4.2 S trength-D eform ation-P ore Pressure R esponse

T he triax ial compression test is widely used to determ ine the shear streng th

param eters, the effective angle of internal friction (j)' and the cohesion in tercept, c .

T he shear strength , t as per M ohr-Coulomb failure criterion is expressed as

T = c' -f aj,an(j) (4.1)

where cr|j is the norm al effective stress acting on the failure surface. Three different

triax ial tests, consolidated drained (CD), consolidated undrained (CU) and unconsol

idated undrained (UU) tests are conducted to determ ine shear s treng th param eters.

Consolidated undrained tests are conducted on cem ented specimens in order

to determ ine to ta l and effective strength param eters. Figures 4.1 to 4.6 shows the

stress-strain and excess pore pressure-strain response for uncem ented and cem ented

sand samples. The relative density and confining stresses are depicted in each figure.

I t is necessary to use back pressure both to fully sa tu ra te the specim en and also to

m easure negative pore pressures. All specimens were satu rated using back pressures

ranging from 400 to 600 kP a depending upon the density. Denser specimens (85 %


I l l

relative density) necessitated back pressures of up to 600 kP a . One o ther reason

for th e developm ent of negative pore pressures was cem entation and the increase in

d ilation w ith cem entation can be seen in Figures 4.3 to 4.6. Rest of the figures are

presented in A ppendix B.

4.3 S trength P aram eters

Ladd (1966) suggested different approaches for obtaining streng th param eters

from undrained testing: (1) a t m axim um stress difference, (2) a t m axim um obliquity,

and (3) a t th e point of tangency to the effective stress pa th . T h ird procedure is

generally recom m ended for dilating type m aterials, hence, present tests are analyzed

using this m ethod.

Figures 4.7 to 4.9 show the stress paths in CU tests. The rest is presented in

A ppendix B. These are used in calculating the peak and residual streng th param eters.

T he stress pa ths show th a t the effective stress paths in all tests lie to the right of

the to ta l stress path suggesting the developm ent of negative pore pressures.

The effective stress pa th also shows the initial positive pore pressure distribution

and then the negative pore pressure distribution due to dilation. The effective stress

p a th s after reaching peak, move towards left and this is due to the drop in deviatoric

stress a t the residual strains. The effective stress pa ths of cem ented and un cemented

sands are practically parallel, suggesting cem entation has m inor influence on friction

angle (Figure 4.10). Cem entation leads to cohesive binding between the grains and

cohesion is entirely mobilized at a deviatoric stress lower than peak deviatoric stress.

A t large strains, disintegration of cemented bonds take place which results in lower

residual cohesion values.

Table 4.1 presents effective strength param eters a t each cem entation level. In

all cases, an increase in relative density results in a rise in the friction angle. Similar

observations are also m ade in cemented sands. This is a ttr ib u ted to the higher the

density, higher the num ber of particle contact points between grains hence higher

friction values. C em entation does not result in a significant increase in friction angles

a t the same densities.

Cohesion intercept is zero for uncem ented sands. However, a cohesion intercept

emerges in cem ented sands. This value increases a t higher densities a t the same


112

aOh

COwK

CO

CJ

05O E-h <3HH>W«

2 0 0 0MS-C0-UNDR45

□ □POD 100 kPa 200 kPa

o o o o o 300 kPa1500

■ee1000

500

012 16840

AXIAL STRAIN ( %)

cdOh

CO -1 5 0P5 -3 0 0

-4 5 0-6 0 0

4 8 12 16

AXIAL STRAIN ( %)

Figure 4.1: Undrained Triaxial Test on Uncemented Monterey No.0/30 Sand {D^

45 %)


113

2 0 0 0

03ÛH

1500 -

1000 -

i ninwKHin

Ko<H-H>

Q

500

MS-C0-UNDR65o e e s o 100 kPa t>t>t>t>o 200 kPa o o o o o 300 kPa

AXIAL STRAIN (%)

0-0-0—o-

9-B-B-a—B—Q

,1 I r I I I I I I I I I 1 I I I I I I

12 16

AXIAL STRAIN (%)

Figure 4.2: Undrained Triaxial Test on Uncemented Monterey No.0/30 Sand {D^

65 %)


114

2 0 0 0

1500m

HU1

1000u

□ DDB-o 100 kPa i> >CHH> 200 kPa ooooe» 300 kPo

500>

Q M S -C 0 -U N D R 4 5

168 1240AXIAL STRAIN (%)

AXIAL STRAIN (* )

Figure 4.3: Undrained Triaxial Test on Uncemented Monterey No.0/30 Sand {Dr

80 %)


115

cd PU

U1U1wK

œ

uwKo<d#-4>

P3

2 0 0 0MS-C1-UNDR45

1500

1000

s-E a-eo 100 kPa 200 kPa

oo o o o 300 kPa500

012 1680 4

AXIAL STRAIN ( %)

xn —150? -3 0 0M —450 Ko -6 0 0

AXIAL STRAIN (%)

Figure 4.4: Undrained Triaxial Test on Cemented Monterey No.0/30 Sand {D^ = 4 5

and C.C. 1 %)


116

2 0 0 0MS-C1-UNDR65

cdDh

[/] 1 5 0 0i nHKHin

1000o

□ □□□□ 1 00 kPo 200 kPo

500

12 1680 4

AXIAL STRAIN (*)

CO —150K -3 0 0M —450Pio -6 0 0Oh 8 12 16

AXIAL STRAIN { %)

Figure 4.5: Undrained Triaxial Test on Cemented Monterey No.0/30 Sand {Dr = 65

and C.C. 1 %)


117

2500MS-C1-UNDR85

2000

2 1 0 0 0 œ e-aa 100 kPa 200 kPa 300 kPa

500

168 120 4

AXIAL STRAIN ( %)

CO —150P -3 0 0w -4 5 0 KO -6 0 0CL, 0 4

— &“&e>•g CD D □

I I I I I 1 I I I I I I I r I 1 I I I I I I

8 12 16

AXIAL STRAIN (%)

Figure 4.6: Undrained Triaxial Test on Cemented M onterey No.0/30 Sand [Dr — 80

and C.C. 1 %)


118

2000T.S.P.

M S-C 0-U N D R 85

1500 □ooH3-a 1 00 kPa AÙ.A-AÙ. 200 kPa ooe e e 300 kPo

000 -

500

1500 20001000 < ( '

500

Figure 4.7: Stress Paths From Triaxial Tests on Uncemented M onterey No.0/30 Sand

{Dr = 80 %)


119

2000

T.S.P. E.S.P.

MS-C1-UNDR85

1500 □■o-D-B-o 1 00 kPa a a a a a 200 kPo ooooo 300 kPc

M 1 0 0 0

500

1200(kPa)

600 1800r

Figure 4.8: Stress Paths From Triaxial Tests on Cemented M onterey No.0/30 Sand

(D r = 80 %; C.C. = 1 %)


120

2000

T.S.P. E.S.P.

MS-C2-UNDR85

1500oooGo 100 kPa£r£r£r£r£ 2 0 0 kPO

1000

500

1200p, p' (kPa)

600 1800

Figure 4.9: Stress Paths From Triaxial Tests on Cemented Monterey No.0/30 Sand

(D r = 80 %; C.C. = 2 %)


121

2000

G600O Uncemented, 200 kPo □□□BO Cemented (2 %) , 200 kPa

1500

Effective Stress Paths

1000

500

cohesion intercept development

200015001000500p' (kPa)

Figure 4.10: Stress Paths of Cemented and Uncemented Sand


122

Table 4.1: Effective S trength P aram eters of Cem ented Sands (U ndrained Tests)

Relative Cem ent Cohesion FrictionDensity C ontent ( # ) Angle

% % peak res peak res45 0 0 0 3&7 3&365 0 0 0 40.5 38.785 0 0 0 4 2 4 40.545 1 19.7 6.3 40.5 3R965 1 40.6 19.2 4 2 4 3 8 285 1 5 4 ^ 3&2 4 2 8 3 2 045 2 2&0 6T2 3&9 3 2 365 2 70.5 37.5 382 3 2 985 2 8&3 42.0 3 8 2 3T6

cem entation level. A t high densities, more contacts exist between the particles and

th e probability of having m ore cem ented contacts across a plane increases.

Similar observations are noted in the residual strength param eters. Significantly

lower cohesion intercept values are obtained at larger strains, this is m ainly hypoth

esized to be due to the breaking of the cem ented bonds.

4.4 Com parison w ith D rained R esu lts

The undrained results are com pared with drained results in Table 4.2 (cohesion)

and 4.3 (friction angle). D rained test results are obtained from the on-going study

by Arslan (1993). 7 days cured specimens of M onterey No. 0/30 sand are tested in

th is study.

Effective friction angles in drained testing are lower than those in undrained

testing. T he variation is 5 to 6° in some cases. This variation can be expected when

the q — p' envelopes are used to estim ate c and (j) values. Similar observations are also

noted by B jerrum (1960). D rained and undrained friction angles are com pared from

the results conducted on several types of clays (B jerrum , 1960). T he envelopes from

stress paths which are equivalent to envelopes a t m axim um principal effective stress

ratios (m axim um obliquity) yield different c and <j)' values than those obtained from

drained tests. D rained friction angles are 3° lower than undrained effective friction

angles and in some cases the difference is as high as 7° (B jerrum , 1960). This variation


123

Table 4.2: A Com parison of Cohesion Values O btained in Drained and U ndrained

Tests

RelativeDensity

CementContent

Cohesion(CD) Cohesion(CU)

( ^ )% % peak res peak res

45 0 0 0 0 065 0 0 0 0 085 0 0 0 0 045 1 7.0 0.0 19.7 NA65 1 10.0 0.0 40.6 6.385 1 15.0 5.0 5A5 19.245 2 10.0 5.0 2&0 6.165 2 17.0 7.0 70.5 37.585 2 25.0 13.0 8R3 42.0

NA-Not Available

is possibly due to the differences and the developm ent of pore pressures w ith strains

in undrained testing.

Cohesion values are higher in undrained testing due to the developm ent of neg

a tive pore pressures. These pore pressures induce ex tra confinement on the sam ple

and the final streng th will be more affected by slight differences in densities across

th e specimen. This m ay be the reason for higher cohesion values. T he increase in

curing period also contributes to th is variation. The influence of curing period on

cohesion is separately investigated.

Selection of either drained param eters or undrained param eters depends upon

the conditions prevailing around the cone during penetration . If the hydraulic con

ductiv ity of the soil is high enough to dissipate the excess pore pressures developed,

th en drained triax ial param eters are used in in terp re ta tion .

4,5 C ritical S ta te Diagram

Application of critical sta te soil mechanics to sands has been less successful than

clays due to the difficulty in defining a virgin consolidation line (Been et al., 1991).

However w ith the development of m odern laborato ry techniques such as undrained

triax ia l tests, m easurem ent problems were resolved and thereby critical sta te lines

for sands were generated. There has been a lot of discussion on w hether critical and


124

Table 4.3: A Com parison Between Effective Friction Values O btained in D rained

and U ndrained Tests

R elative Cement Friction(CD) Friction(CU )Density Content

% % peak res peak res45 0 3ff5 3&5 3&7 3 5 j65 0 3&7 81 2 40.5 3 8785 0 40.5 36.0 4&4 4ff545 1 3ff3 3&5 40.5 3ff965 1 37.4 34.4 42.4 38785 1 3 8 J 3&4 4&8 3ffO45 2 37.6 31 0 3&9 3&365 2 3&5 3&4 3&T 3ff985 2 3&4 37.0 3&7 386

N A - N o t Available

steady s ta te lines are identical (Been et al, 1991).

There seems to be a difference between these two lines based on the m ethod of

m easurem ent. Critical s ta te line is derived from drained, strain-rate-controlled test

resu lts on d ila tan t sam ples, whereas steady s ta te line is obtained from undrained

triax ia l tests. Tests conducted on a Erksak sand showed th a t the critical and steady

s ta te lines from drained and undrained tests are identical, implying testing has no

influence on the u ltim ate s ta te (Been et al., 1991).

The s ta te param eter concept developed by Been e t al., (1986) provides a refer

ence critical s ta te from which the sta te param eter (definition of s ta te param eter is

presented in C hapter 2) and other sand properties are derived. The sta te param eter

concept is used in in terp reta tion of cone results. Results from undrained triax ial

tes ts are used in estim ating critical confining pressures and also in developing the

critical s ta te line.

The critical confining pressure is the confining pressure a t which there is no

volume change. In case of undrained testing where the volume change is constant,

critical confining pressure is evaluated at larger or residual strains where there is no

significant change in excess pore pressure. The critical confining pressure is calcu

la ted for each confining pressure a t a particular relative density. The average of the

critical confining pressures obtained in the tests is defined as the critical confining


125

pressure a t th a t density. T he above procedures are repeated for each relative den

sity and cement content. Steady s ta te lines are p lo tted in Figure 4.11. T he figure

shows th a t steady s ta te lines for cem ented specim ens locate close (slightly to the

right) to th e uncem ented critical s ta te line. T he variation between cem ented and

uncem ented values is m inor since they are p lo tted on a logarithm ic scale. In order

to show the difference, the results are rep lo tted on natu ra l scale (Figure 4.12). Ce

m ented specim ens d ilate m ore than uncem ented specim ens im plying developm ent of

higher negative excess pore pressures leading to higher effective confining pressures

and higher critical octahedral stresses. Since there is no significant change in void

ra tio in undrained testing on cem ented specimens, the steady s ta te line (SSL) for

cem ented specimens moves to the right and above the SSL corresponding to the

uncem ented samples. The above SSL are p lo tted com paratively with th e SSL from

tests conducted on Erksak sand (Been et al., 1991) and M onterey No. 0/30 sand

(Been et ah, 1986) in Figure 4.13. Tests on M onterey No. 0/30 sand are conducted

a t lower confining stresses. The steady sta te line is generally approxim ated by a

stra igh t line in e — logp ' space and th is approxim ation is valid for sands w ith sub-

angular to sub-rounded particles (Been et al., 1991). However, the shape of the line

over a wider stress range is different and it is sim ilar to the shape plo tted for Erksak

sand in Figure 4.11. Steady s ta te line curves abrup tly a t a stress level of 1 M P a .

This break in the steady s ta te line is indicative of a change in m echanism of shearing

a t higher levels of stress (Been et al., 1991). It is also hypothesized th a t a t higher

confining stress levels, there appears to be some breakage of grains which result in

the ab rup t change in the slope of the SSL (Been et al., 1991).

Com bining the present da ta on M onterey No. 0/30 sand (conducted a t higher

confining stresses) w ith Been’s results on M onterey sand (conducted at low confining

stresses), a com plete steady s ta te line similar to th a t of Erksak sand is constructed .

SSL of Erksak and M onterey sands are parallel to th is line and then vary due to

different grain sizes and shapes.

This steady sta te line is used to derive steady s ta te param eter, which is subse

quently used in the in terp reta tion of cone penetrom eter results.


126

1.20

Uncemented Cementedf 1% Cemented(2ss1.00

0.90

0.80

0.70

0.60 'A

1000

Figure 4.11: Steady State Line Diagrams for Cemented and Uncem ented Sands


127

1.20

UncementedCementedMssCemented(2%1.00

0.90

0.80

X )

0.70

0.60

0.50 400 1200800I

Figure 4.12: Steady S tate Line Diagrams for Cemented and Uncemented Sands


128

(D

O

"dcr_"D'o>

0.80

0.75

0 .7 0

0.65

GGGGG Uncemented□ B B B B Cernented(l%) iAAAAA Cemented(2*) □ Erksak Sand Monterey No. 0 / 3 0 Sand

0.60

0.55 1000010001 0 010P ’

Figure 4.13: Comparisons of SSL


129

4.6 M odeling Param eters

A sim ple soil m odel proposed by Ju ran and Guermazi (1988) is upda ted and

used to analyze the triax ial test results. This m odel was essentially developed to

sim ulate soil response in cavity expansion tests. This model assum es soil to be

hom ogeneous and isotropic. A strain-hardening elasto-plastic m aterial behavior w ith

a non-associated flow rule is used. T he stra in hardening param eter is chosen to be

th e deviatoric stra in (7 ). T he stress invariants used in form ulations are qm, th e

deviatoric stress, (cr( — a'^) and and the effective octahedral norm al stress {cToct-

(a[ + 2a'^)/3). The différence between stress invariants, Çm and the stress pa th

variables, p, p and q should be noted.

4.6.1 Yield Function

A M ohr-Coulom b type yield criterion is considered;

/(o -ij.7 ) = - ^ (7 ) (4-2)P m

where aij is the stress tensor, qm = (cr( - cr' ) is the deviatoric stress, p'^ = (cr(-f2 (j3 ) / 3

is th e effective octahedral stress in triaxial testing conditions, <j(, (jg are, m ajo r

and m inor effective stresses respectively. The k ( j ) is the stra in hardening function

relating the actual yield surface to the current strain rate . Shear stra in , j i s c i — £3

and is assum ed to be the strain hardening param eter (Ju ran and Beech, 1986).

T he above yield function is generally used for frictional or cohesionless m aterials.

Since, cem ented m aterials exhibit a cohesion in tercept, failure criterion is updated

as:

qm — C,j,c 4- Mifip^ (4-3)

where

= r a

= r a

To m ake analysis sim pler, yield function which intersects q {y axis on q-m — plot)

a t Cÿ C is extended to intersect the p ^ axis. This gives the following form to the

yield function.

•?m — Pum î' ~ (Pl T (4-6)


130

where pi = ^ and p L = Pi + pL- Then

/(o -ij,7 ) = - T ^ - ^ (7 ) (4-7)P u m

D uring modeling, the im aginary ordinate, p[ are deducted from the com puted

values a t each strain level in order to get the correct p|^ values. T he corrected values

represent the actual p values for the above c — (j> yield function.

In the analysis, the following assum ption is made; If the peak cohesion value

is used in the pi values, the corrected p values a t shear strains corresponding to

residual stresses will be lower th an actual values. This, coupled w ith the difficulty

of selecting a different cohesion value a t each shear stra in , 7 , lead to the use of an

average cohesion intercept (average of peak and residual cohesion intercepts) in the

analysis. Even though th is results in some difference in p values, the discrepancy

can be disregarded for all p ractical purposes.

4.6.2 Hardening Function

T he strain hardening function, h (7 ), used in the above equation m ust be specif

ically defined for the case of contracting and dilating m aterial. For loose contracting

sands, a hyperbolic strain hardening function with two m aterial constants is often

used:

For a triax ial test, a = 6 = where uq is the initial consolidation stress, G is

the in itial shear m odulus, and is the friction angle a t constant volume.

For dense dilating sands, it is assum ed th a t the hardening function A(7 ) is

parabolic to hyperbolic and can be w ritten as;

where the constants a, h and c are determ ined from the following conditions;

1 . th e initial tangent m odulus of the ^ (7 ) function is equal to

2 . a t peak stress ratio, ^ (7 ) = M ^, and

3. a t the critical state , h { j) = M^^v,

To


131

these conditions yield:

P:* c;

(4.10)

h = (4.11)

C — (4.12)

where T = 1 + [1 -

4.6.3 Flow Function

A dopting M ohr-Coulom b or the more generalized Drucker-Prager yield criteria

along w ith an associated flow rule leads to overestim ation of dilatancy. Therefore, a

non associated flow rule m ust be used (Ju ran and Beech, 1986). The following flow

ru le which defines stress ratio-dilatancy relationship is derived by Ju ran and Beech

(1986) based on energy considerations:

T] = s in u = — = — {M^cv) — ^ (4.13)(^7" Pm

where

del = plastic volum etric strain increm ent

d Y = plastic deviatoric strain increm ent

T] = dilation rate

u = dilation angle

and jig is a correction m odulus defined as

Ps = Pi w hen-^ < M,^cv; contracting behavior P

Ps = P 2 w hen-^ > dilating behavior (4.14)P

4.6.4 Plastic Potential Function

A non-associated plastic po ten tial function g{qmt Pm) — 0 can be derived, as

sum ing coincidence of principal axes of stress and plastic strain increments.

9{p'm^<}m) = ^ -^{Pm^M<t>cv) (4.15)Pm


132

Figure 4.14 shows schem atically the strain hardening function, h { j) , the associated

volum etric response, and the corresponding stress ratio-dilatancy rate function. M ax

im um dilatancy is equal to the slope of volum etric strain - shear stra in curve a t the

peak of the h{'j) function, whereas the d ilatancy rate a t the critical s ta te is equal to

zero (Ju ran and M ahm oodzadegan, 1989).

4.6.5 Model Results

This soil model requires five param eters: or M P H I or M P H lc v ,

fix and pi2 - D rained results from Arslan (1993) and Rad and Clough (1982) are used

to determ ine these param eters. The following procedure is then adopted in soil

modeling.

1. Three program s, w ritten in Fortran are used for soil modeling. T he flow charts

and source listings of these program s are shown in Appendix B. The first pro

gram reads the drained d a ta and calculates stress ratio , Çm/pLî shear strain ,

7 and dilation rate . From these results, initial values of ^p, <t>cv, g i , and

G/ctq are estim ated. These results are calculated for each test a t each relative

density.

2. These results are then input into a second program which sim ulates drained

triax ial behavior. T he output of these results contain Çm/pL, volum etric strain

and shear strain . These results are then com pared w ith experim ental drained

results. If the sim ulation is not in agreement with the experim ental results,

the program is rerun w ith different param eters, G /oo, pi and p 2 values. The

param eters th a t rendered best sim ulations are tabu la ted in Table 4.4 and 4.5.

Some of the results are shown in Figures 4.15, 4.16 and 4.17. Rest of the

comparison plots are presented in Appendix B. T he peak and residual (constant

volume) (f> values are very close to the experim ental results suggesting th a t best

sim ulations obtained.

Figures 4.18 and 4.19 shows the comparison between volum etric s tra in and

shear strain . Rest of the comparison plots are presented in A ppendix B. Com

parisons show th a t Pi value is 0.55 in the contraction zone and p% can vary from

0.20 to 0.6 in the expansion zone, based upon the relative density. Lower p 2

values yield higher dilational strains which are represented by negative values.


133

q = c C<t> + p’ M(j>

cC*

CONTRACTINGDILATING

peal ^, / d i l a t i n g

E v

CONTRACTING

Figure 4.14: Assumed Constitutive Equations and Related Soil Param eters (Juran

and Beech, 1986)


134

Figure 4.20 shows the influence of varying G/ctq on prediction of drained

strength-deform ation behavior. It is observed th a t the variation is m inor within

the range of changes observed in th is study ( 1 0 0 to 2 0 0 kPa).

Table 4.4: M odelling Param eters

c . c . Dr G /c7o 4>p ^ c v C a v

% %0 45 1 0 0 . 0 36.5 33.0 0 . 0

0 65 108.0 38.7 34.2 0 . 0

0 85 115.0 40.0 35.9 0 . 0

1 45 75.0 36.3 33.5 4.31 65 1 0 0 . 0 37.4 34.4 5.21 85 1 2 0 . 0 39.0 36.4 6 . 2

2 45 1 2 0 . 0 37.6 34.0 13.02 65 140.0 37.5 35.0 17.02 85 170.0 40.0 37.0 25.0

3. T he th ird program sim ulates undrained triax ial behavior. T he param eters th a t

best sim ulated drained behavior are used in this program . The o u tp u t from

th is program consists of Çm/pim excess pore pressure and shear stra in . These

are com pared w ith experim ental results of the present investigation. Figures

4.21, 4.22 show these comparisons. Comparisons show th a t sim ulations from

the model closely approxim ate the experim ental results.

Table 4.5: M odelling Param eters

PiDr (Ty 1 0 0 2 0 0 300 1 0 0 2 0 0 300

% % k P a k P a k P a k P a k P a k P a

0 45 0.55 0.55 0.55 0.35 0.45 0.650 65 0.55 0.55 0.55 0.30 0.40 0.600 85 0.55 0.55 0.55 0.25 0.35 0.551 45 0.55 0.55 0.55 0.25 0.30 0.351 65 0.55 0.55 0.55 0.30 0.40 0.551 85 0.55 0.55 0.55 0 . 2 0 0.40 0.502 45 0.55 0.55 0.55 0.30 0.40 0.502 65 0.55 0.55 0.55 0.25 0.30 0.352 85 0.55 0.55 0.55 0.20 0.25 0.35


135

COMPARISONS (MS-C0-DR85)

2.00

G /S IG O = 1 1 5 .0 ; = 0 .0(Op - 4 0 .5 ; (5cv - 3 6 .0

G8G80 Experimental (100 kPa □QffiQ Experimental (200 kPa

Experimental (300 kPa Theoretical

0 5 10 15 20 25S h ea r S t r a in (p e rc e n t )

Figure 4.15: Com parisons Between Predicted and Experim ental D rained Triaxial

Tests


136


2.00

.7 5

.5 0

.2 5

.00

0 .7 5

G/SIGO = 120.0; C 6.2av "36.40 .5 0

G0GŒ) Experimental (100 kPa BfTIFFI Experimental (200 kPa AAAAA Experimental (300 kPa Theoretical

0.25

0.00255 10 15 20

S h ea r S t ra in (p e rc e n t )


Tests


137

COMPARISONS (M S-C2-D R85)

2.00

1.75

1.50

1.25

1.00

0.75

G/SIGO = 170.0; C 25.0av “37.0^ 0.50

Q G 0 6 O Experimental (100 kPa BTtfildbJ Experimental (200 kPa AAAAA Experimental (300 kPa Theoretical

0.25

0.005 10 15 20 2 50

S hea r S t r a in (p e rc e n t )

Figure 4.17: Com parisons Between Predicted and Experim ental Drained Triaxial

Tests


138

15.00

OMPAEISONS (M S-C0-DR65)

10.00 -

c: 5.00 -

o

0.00

o

0) - 5 . 0 0 -E :

^ - 1 0 . 0 0 -

GGGGOExp, 100 kPa LHH-P IExp, 200 kPa A ù aW W Exp, 300 kPa

0.300.400.60

- 1 5 . 0 030200 10

S h e a r St ra in


Tests (Volumetric Strains)


139

15.00


10.00

(2 5.00 -

D

0.00

O

CD - 5 . 0 0 -E :

^ - 10.00 -

GGGGOExp, 100 kPa l . .- f c r l - 1 - h I Exp, 200 kPa

Exp, 300 kPa0.300.400.55

- 1 5 . 0 030200 10

S h e a r St ra in


Tests (Volum etric Strains)


140

2.00

M 0.50

0.00 □


(Op = 35.3; (Ocv = 34.1Cav - 0 . 0

I I I I I Experim ental G/SIGO = 100.0 G/SIGO = 200.0

I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I t I I I I I I..5 10 15 20

S h ea r S t r a in (p e rc e n t )25

Figure 4.20: Influence of G /ao on Stress R atio


141

1

.00

.7 5

^ 1 . 5 0 h

' ^ 1 . 2 5

4-;co

K

1.00

w(/]0 )

04-)m

.7 5 -i i

.5 0

0.25

COMPARISONS (M S-C0-U N D R80)

-O

G/SIGO = 115.0(Op = 39.1; cpcv = 36.0

G 990O Experimental (100 kPa) I 1 Experimental (200 kPa) Theoretical

0 QQ 0 ........................I l _I_I_I_I I I I_I_I_I_I_I_I_I_I_I_I_I_I---1--L0 5 10 15


Figure 4.21: Comparisons Between Predicted and Experim ental U ndrained Triaxial

Tests


142

2.00COMPARISONS (M S-C1-U N D R85)

1.75

1.50

1.25

1.00

0.75

0.50

G ^ 9 0 Experimental (100 kPa I I I I !■ I Experimental (200 kPa Theoretical

0.25

0 .0 0 Q-


Figure 4.22; Comparisons Between Predicted and Experim ental U ndrained Triaxial

Tests


143

Dilational angles calculated from drained results from program 1 are shown in

F igure 4.23. O ther model param eters, G/<To, (j)p and (f>cv versus relative density and

cem ent content are presented in Table 4.4. These results along w ith dilational angle

are la te r used in estim ating the lim iting pressures by the spherical cavity expansion

theory. This lim iting pressure can be correlated w ith tip resistances.

4 .7 U nconfined C om pression Tests

In an a tte m p t to shorten the tim e of testing, it was decided to test calibra

tion cham ber specimens after 7 days of curing. Therefore, a curing tim e of 7 days

was taken as reference all th roughout the study. T he transfer and placem ent of

th e specim en in the calibration cham ber along w ith cone penetration testing often

took between 10 to 12 hours. T he question then arose w hether th is excess tim e of

curing would affect the streng th significantly. If this change is significant, then any

correlations m ade w ith using the strength-deform ation behavior of the 7 day cured

specim ens would have involved an error.

Unconfined compression tests are conducted on 1 and 2 % cem ented specimens.

Specim ens are prepared a t th ree ranges of relative densities and they are cured a t 3,

7, 14 and 28 days.

The tes t results are shown in Figures 4.24 to 4.25. Figure shows th a t there is

an increase in the unconfined compression strength w ith curing period, however the

s treng th increase beyond 7 day curing period is insignificant. These plots are used to

prepare F igure 4.26 showing dq/dt versus tim e, t for the 1 % cem ented specim en. The

slope, dq/dt decreases significantly w ith tim e, particu larly after 14 days, im plying

curing m ay not affect results significantly beyond 14 days.

The 7-day curing period is chosen for convenience and all comparisons are m ade

a t th e same curing period.

The effect of tem pera tu re on unconfined compression streng th of cement s tab i

lized m aterials was conducted by D um bleton (1962). The ra te of increase of streng th

increased w ith an increase in tem pera tu re , hence all the tests, triax ial, unconfined

compression and calibration cham ber tests are conducted a t room tem peratu re .


144

50

OOOOODr 45 %45 C.C.AAAAA Dr 85 %

40

r—4 35

30

^ 25

5

0

5

00 1 0 0 200 300 400

Effective Vertical S tress (kPa)

Figure 4.23: Dilational Angles of Uncemented and Cemented Sand


145

8 0

C E M E N T CONTENT 2 %

6 0

ao.

4 0coo

■ 0 - 4 5

3

2 0

10 20

C U R I N G T I ME ( d a y s )

3 0

Figure 4.24: Influence of Curing Time on Unconfined Compression Strength of 2 %

Cemented Specimen


146

8 0

CEMENT CONTENT 1%

60

oCL

~ 654 0COo3 45

20

10 20

CURI NG T I M E ( d a y s )

3 0

Figure 4.25: Influence of Curing Time on Unconflned Compression Strength of 1 %

Cem ented Specimen


147

oT3\O

0 _JC

crT3

1 2

C E M E N T CONTENT 1%

8

4

0 3 020CURI NG TIME ( d a y s )

Figure 4.26: d q /d t versus T im e on 1 % Cemented Specimens


148

4.8 Sum m ary

Several triax ial and unconfined compression tests conducted on cemented and

uncem ented M onterey No. 0/30 sand reconfirmed th e following observations.

a Cem entation induces cohesion in tercept in sandy soils.

« C em entation has m inor influence on friction angle.

» C em entation, as like relative density, increases the d ilational behavior.

o Cem entation increases in itial shear m odulus.

« D rained param eters are lower th an the undrained effective stress param eters.

This is due to the procedures used (envelopes drawn in stress paths) in ana

lyzing the tes t data . Similar procedures used by B jerrum (1960) also observed

higher undrained friction angles and cohesion values th an in the drained te s t

ing.

0 S teady s ta te line for M onterey No. 0/30 sand is sim ilar in shape to th a t of

Erksak sand (Been et ah, 1991) and the variation in m agnitudes of e and p'

a re due to the different size and shape of the aggregates and the ir dilational

behavior during shearing.

T he above observations signify the im portance of cem entation. T he curing period

increases unconfined compression strength . However, this increase beyond 7 day

curing period is insignificant. In theoretical calculations in the in terp re ta tion of

cone penetration testing results, the strength param eters are taken exactly a t the

sam e curing period th a t is used for curing cham ber specimen on which cone te s t is

conducted.

A simple elasto-plastic soil model is used to sim ulate drained and undrained

behavior. Yield function in term s of cohesion and friction is developed and a non-

associated flow rule is adapted in the m odeling. D rained test results are first p ro

gram m ed to evaluate the necessary approxim ate param eters. These are then used in

th e second and th ird program s to sim ulate drained and undrained behavior. T he p a

ram eters obtained in drained sim ulations are well m atched w ith experim ental results.

U ndrained sim ulations also closer to experim ental results. These m odel param eters


149

will be used in the in terp reta tion of cone penetration results by spherical cavity

expansion theory.


Chapter 5

CONE PENETRATION TESTINGT he difficulty in undisturbed sam pling of natu ral cohesionless deposits, the cost

and complications of any envisioned field-scale calibration a ttem p ts prom pt investi

gators to use laboratory scale experim ental models. C alibration chambers constitu te

such experim ental models used in calibrating insitu testing devices.

An experim ental model should ideally replicate the geom etrical considerations

and boundary conditions representative of field conditions. Cone penetration in

a homogeneous deposit under field conditions constitu tes an axisym m etric, sem i

infinite, steady-sta te penetration problem . It is a sim ple task to down-scale the

axi-sym m etric na tu re of the problem in to th e laboratory by constructing cylindrical

cham bers and conducting experim ents a t th e center. However, the m ajor concern

arises while down-scaling the semi-infinite boundary to finite boundary. The question

then arises; w hat the ratio of the diam eter of the cham ber to the cone diam eter should

be so th a t the results are not affected by the finite boundary conditions. A system atic

study is not yet available; however the d a ta generated by the increasing num ber of

cham bers of different sizes provide some sta tistical trends.

There are two philosophical approaches to the study of the geotechnical charac

teristics of deposits by insitu testing devices in an a tte m p t to offer em pirical correla

tions or to assess the validity of the theoretical predictions; 1 ) conducting field testing

followed by comprehensive laboratory testing and evaluation of soil characteristics,

com position, environm ental variables, 2 ) conducting experim ental model tests in the

laboratory under controlled compositional and environm ental variables.

B oth philosophies have their advantages and disadvantages. In the case of pen

e tra tio n in cohesive soils, the first approach is generally followed. This is possibly

due to the relative ease in retrieving und istu rbed cohesive samples and also the tim e

constrain ts associated w ith reconstitu ting and consolidating large-scale fine-grained

specim ens in the laboratory. However, calibration cham ber testing on reconstitu ted

specim ens has been favored for coarse grained deposits possibly due to the extrem e

difficulty encountered in sampling such deposits. The disturbance concern is particu-

150


151

larly true in case of uncem ented or lowly cem ented deposits since changes in fabric or

breaking of the cem entation bonds will strongly affect th e results. Consequently, it

seems m ore p ruden t and cost-effective to study cone penetration in cem ented sands

in calibration cham bers under specified com position, fabric and environm ental fac

tors. It is envisioned th a t com plications arising from changes in fabric and gradation

are introduced and evaluated sequentially.

In line w ith the above presented reasoning and philosophy, th e testing reported

in th is study is conducted in a calibration cham ber on reconstitu ted , uncem ented and

artificially cem ented specim ens of M onterey No. 0/30 sand. T his chapter presents

th e results ob tained in these tests together with a com parative discussion of the

influence of various factors like relative density, cement content, testing location on

the results.

5.1 T esting Program

T he tes ts are conducted in a double-walled flexible cylindrical calibration cham

ber on specim ens of 53 cm d iam eter and 79 cm in height. A m in iatu re quasi-static

cone penetrom eter (M QSC) of 1 . 2 cm in diam eter is used, rendering a diam eter

ra tio of 42. T he cone used was a friction cone. The details of the cham ber, cone

and testing procedures are presented in C hapter 3. M onterey sand No. 0/30, a

sand comm only used in laboratory experim ents, is used in tests. O rdinary portland

cement is used as cem enting agent in cemented specim en preparation . P luviation

technique is employed in specim en preparation. Test results are discussed in the

following sections.

5.1.1 Uncemented Specimens

T he testing program for th is study involved a to ta l of 14 tests conducted on

th ree different ranges of relative densities (Table 5.1). Six of them are prepared in

the range of above 85 % relative density, four of them in the range of 65-75 % and

th e rem aining in th e range of 45-55 %. T he first few tests w ith relative density of

above 85 % are conducted to ensure repeatab ility and to investigate the appropriate

testing procedure. Specimen No. 1 2 could not be tested because of leaks in the inner

and the ou ter cells. In another sam ple (No. 14 in Table 5.1), erroneous results were


152

obtained because the electrical wire connections in the cone tip and friction sleeves

were tw isted resulting in the loss of tip da ta while testing.

T he details of the tests during consolidation are shown in Table 5.1. F igure 5.1

presents a com parative plot of the values obtained in the calibration cham ber

testing w ith respect to the ranges of the values th a t should be obtained w ith the

Ja k y ’s relationship (1 — sin4>). T he figure shows th a t Ko values of 0.37 to 0.54 are

obtained for above 85 % relative density; 0.42 to 0.47 for 65 - 75 percent relative

density and 0.35 to 0.62 for 45 - 55 % relative density specimens. Ko values obtained

from the testing are no t always consistent with the Ja k y ’s relationship (1 — sinçi).

Possible reasons are: 1) leaks betw een the inner and outer cells which m ay resu lt in

lower horizontal stresses thereby lov/er K q values, 2 ) leaks between inner and piston

cells result in a horizontal stress value equivalent to the vertical stress (p iston cell

w ater pressure), resulting in higher Ko values.

Figures 5.2 to 5.4 show some of the penetration results in th e tests conducted.

T he rest is presented in Appendix C. T ip and friction resistances are depicted in

these figures. The tip resistance and sleeve friction resistance readings are taken at

around m id-depth (between 30 to 35 cm) in the specimens. In some cases, values

th a t show m inor variation along the dep th of the specim en are taken as the reading.

Table 5.2 sum m arizes the results.

5.1.2 Cemented Specimens

A to ta l of 20 cem ented specimens are prepared and tested. Eleven of the speci

m ens are prepared a t 2 % cem entation and the rest a t 1 % cem entation. Tables 5.3

and 5.4 show the characteristics of the tests and the testing program .

A test w ith 2 % cem entation was used as a tria l test to check w hether vacuum

is needed inside the specimen while transferring the specimen into the cham ber.

Cem entation of 2 % was not enough to provide the sufficient confinement (cohesion)

to avoid collapse of the specimen. Once the outer m olds were separated, specim en

collapsed (No. 11 in Table 5.4). Hence, it was decided to apply vacuum to all

cem ented specimens during tran spo rta tion and before confinement.

Initially, it was planned to carry out the tests w ith a piezocone which m ea

sures excess pore pressures a t the cone tip . It is hypothesized th a t cem entation


153

0

0 .9Ko Relationships

0 .7

0.6

0 .3

OOOOO D,□ □ □ □ □ D, AAAAA D,

8 00.2

4 5

0.00.6 0.8 1.00 .40.20.0

Measured Ko

Figure 5.1: Comparison of Ko values with Jaky’s relationships


to

CN

IL,

154

O l!_l—I I I I I I I I——I I I I I I I I I I I I I I I I I 1_O

rooCN

o-4- oCD

CO

CJ CM

tlO O(N CD

- e g U -

ScQ

j I I I I I I u _l I I I I I I I L_

aÜ

t)CT

_l I I I I I L_o o

CNO oCDOOTj-

oot o

ooCM

OO

OoCM

OM- OCD

( m o ) m d o Q

Figure 5.2: Cone Penetration Test Results on a Specimen (Dr = 71.9 %; e^nax =

0.85; train = 0.56 and 100 kPa)


155

CN

OCOOo o

CN

C J C N

n

ooCD

OO ooo

o

CD o^ CN

^ o

oo o

CNoN- O

CD

( m o ) q i d o Q

Figure 5.3: Cone Penetration Test Results on a Specimen {Dr — 68.7 %; e„

0.85; tmin = 0.56 and 200 kPa)


156

CN

OOoo o

CN

CM

gO

o oN- OCD

OCNO

O ON-OCN

OCD( u i o )

Figure 5.4: Cone Penetration Test Results on a Specimen {Dr = 71.1 %;

0.85; Bmin = 0.56 and 300 kPa)


157

Table 5.1: C haracteristics of Tests (U ncem ented)

Test No. Dr height(m)

CTv[kN Im ^)

^hifho{kN jm ?)

Ko

1 90.0 0.71 1 0 0 54.1/ 51.3 0.542 90.0 0.71 2 0 0 87.1/89.3 0.443 8 & 0 0 . 6 6 300 120.4/123.0 0.404 71.9 0.73 9&4 41.8/ 41.8 0.425 6&7 0.71 204 95.4/95.7 o ^ a6 71.1 0.74 300 121.44/120.3 0.407 4&7 0.75 1 0 0 43.8/44.1 0.448 5&4 0.75 2 0 0 76.9/70.0 &359 s i a 0.73 300 185.6/174.2 0.62

1 0 8 & 0 0.71 1 0 0 37.1/ 35.2 0.371 1 84.0 0.72 1 0 0 3 & 3 /3 4 j 0.351 2 8 & 8 0.69 1 0 0 o.o/o.o (leak)13 6&3 0.71 1 0 0 41.3/41.8 0.4114 8&3 0 . 6 8 - - -

m ay decrease hydraulic conductivity resulting in undrained conditions during cone

penetration . It is envisioned th a t this excess pore pressure could assist in develop

ing a classification scheme. One test conducted at the center of a specim en (C.C.

2 % and Dr 65 %) using a dual piezocone of 10 cm? area dem onstrated th a t ex

cess pore pressures are not developed (Figure 5.5). The pore pressure developed is

im m ediately dissipated during penetration implying th a t the reduction in hydraulic

conductivity may not be enough to cause undrained conditions during cone p ene tra

tion. Consequently, all the tests were conducted with the available m iniature friction

cone.

T he consolidation characteristics of all specimens at cem entation levels of 1 and

2 % are presented in Tables 5.3 and 5.4. Ko values obtained in these consolidations

are com pared with Jaky ’s relationships (Figure 5.6 and 5.7). Specimen num bered

2 (Table 5.3) has a Ko value of 1 due to the leaks between the inner cell and the

p iston cell resulting in same stresses. Similar findings as noted in uncem ented sand

test results are observed.

Cone penetration test results are shown in Table 5.5 and Table 5.6. Cone test

results conducted on specimens of 65 % relative density and for both cem ent contents

of 1 and 2 percents are shown in Figures 5.8 to 5.13. Rest of them are presented in


158

0Pore P ressu re (k g /cm ^ )

10 20 30 40 50

PiezoconeDr 65 C.C. 2 %Confining Stress 65 kPo Penetration Rote 2 c m /s e c

i 20

40

H ydrostatic Line

60

Figure 5.5: Pore Pressures During Piezocone Penetration of Cem ented Specimen

(C.C. 2 %)


159

0

0.9

O 0.8

0 .7

0.6

0.5

0 .4

0 .3

8 0 % 6 5 % 4 5 %

OOOOO D, □ □ □ □ □ D,A A A A A D,

0.2

0.01.00.80.60.40.20.0

M easured Ko

Figure 5.6: Comparison of Values with Jak y ’s Relationships (C.C. 1 %)


160

0

0 . 9

2 0.8

0 . 7

0.6

0 . 3

O O O O O Dr > 8 0 % □ □ □ □ □ Dr = 6 5 5S AAAAA Dr = 4 5 5g

0.2

0.01.00.80.60 . 40.20.0

Measured Ko

Figure 5.7: Comparison of K„ Values with Jaky’s Relationships (C.C. 2 %)


161

Table 5.2: P enetration Results (Uncemented)

Test No. Dr{kNjm'^)

Ko Qc(ksc)

fs(ksc)

f.r.

1 90.0 100 0.54 122 0.34 0.492 90.0 200 0.44 218 0.60 0.503 8&0 300 0.40 350 1.73 0.494 71.9 96 0.42 84 0.205 6 8 J 204 0^& 176 0.79 0.456 71.1 300 0.40 203 1.317 4&7 100 0.44 60 0.20 0.338 56.4 200 0.35 75 0.45 0.609 5 4 ^ 300 0.62 115 0.73 0.6310 8&0 100 0.37 103 0.30 R2811 84.0 100 0.35 125 0.80 0.6412 8&8 100 0.32 - - -13 6&3 100 0.41 - - -14 8&3 - - - - -

Note: 1 ksc — Ikg jcw? — 100 kPa

A ppendix C. These results along with uncem ented cone test results are used to study

th e influence of relative density, cem entation and testing location on cone resistance

param eters.

5.2 Perform ance A ssessm ent

In perform ance assessment of the tests conducted in th is calibration cham ber

using th e MQSC, repeatab ility and accuracy of the results are considered.

5 .2 .1 R e p e a ta b i l i ty

In an a ttem p t to evaluate repeatability and precision, tests on uncem ented speci

m ens w ith similar relative densities are compared. The tip resistance profiles for bo th

specim ens are presented in the Figure 5.14. There is little difference in the tip re

sistances recorded in the two tests; however, there is significant variation in friction

resistance (0.34 to 0.80 kgfcm^) . The test conducted on 84 % relative density spec

im en is affected by the erroneous sleeve calibrations and also due to m ishandling

and not cleaning the sleeve portion. This has happened in the beginning phase of


162

CM

_i I I I I I I I i_ ............................oCD

OOCM

CM

ooo oo

CMOM-

CM o

S RC

lafl o 24 CN

: CN

r o_

- 1)

uCT

.........................................

oCM

oM" OCD

(m o) q'^dgQ

Figure 5.8: Cone Penetration Test Results on a Specimen {D^ = 68.4 %; e„

0.85; Cmin = 0.56; C.C. 1 % and 100A:Pa)


163

oo otûo

CMO

CM

CDCM

OO O

CDOM-

CM

O)CM O

CT

^ O

Q OCM

OCD

( m o ) q -^ d a Q

Figure 5.9: Cone Penetration Test Results on a Specimen {Dr = 66.4 %; e„

0.85; e w = 0.56; C.C. 1 % and 200A:Pa)


164

CM

OO OCMro

C\2

5 p

oo oo o

CM CDO

O

CT

O O•M-

OCD

OCM

(m o) q^^doQ

Figure 5.10: Cone Penetration Test Results on a Specimen {D^ = 70.2 %; e„

0.85; e^,„ = 0.56; C.C. 1 % and 300 kPa)


165

(N

OTf- OCO

o oCN

•ee-CN

(ao

ooCD

OCN

ON-O

l D

•c4oon

cr

ooCN

ON- OCD

( ix io ) q;da(%

Figure 5.11; Cone Penetration Test Results on a Specimen {Dr = 69.6 %;

0.85; eîn = 0.56; C.C. 2 % and 100 kPa)


166

CN

O Ot o

ooCN

CM

m

OCD

O OOo

oCN

OOc r

Oo o o o

(m o)CD

Figure 5.12: Cone Penetration Test Results on a Specimen (D^ = 69.2 %; e„

0.85; tmin = 0.56; C.C. 2 % and 200 kPa)


167

m

CM

n

SO CM

iaX)

;

r oOJ

;IICCI

o

k ”

O C Ô

.. . . . . . . . . . . . . . . . . . . . . . . . I l l 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1O oM-O

CMO

n

cr

oCM

oM- OCD( u i d ) q : ^ d a a

Figure 5.13: Cone Penetration Test Results on a Specimen (Dr = 72.1 %; e„

0.85; e-min = 0.56; C.C. 2 % and 300 kPa)


168

Table 5.3: C haracteristics of Tests (C.C. 1 %)

Test No. Dr height(m)

o-v(tIV /m ")

^ki/ho{ k N jw ? )

Ko

1 86.0 0.71 100 43.5/46.0 0.542’ 8A6 0.71 200 200/200 1.003 8&2 0.70 300 149.4/152.1 0.494 68A 0J2 100 37.3/34.45 0.375 66.4 0.71 200 125.94/116.2 0.636 70.2 0.74 300 148.15/142.8 0.497 4&8 0.71 100 35.7/30.7 0^68 4&6 0.704 200 82.0/93.1 0.419 5&2 0.72 300 171.6/173.2 0.57

Note: * - Leak Between Piston and Inner Cel

th is testing. These m istakes are rectified in the subsequent tests by recalibrating the

cone and cleaning the sleeve portion prior to each test. A nother tes t conducted on a

sim ilar specim en for studying the location influence yielded a value of 0.30 kgjcm?'

which is close to 0.34 kg/cm^. Tests conducted for studying the influence of bound

ary condition (presented in the next chapter) also showed th a t bo th tip and friction

resistances are repeatable.

5.2.2 Accuracy

Accuracy requires testing a m aterial w ith known tip resistance and sleeve friction

values in the present test setup. There is not yet such defined m aterial in calibration

cham ber testing. Furtherm ore, a round robin testing of a specific sand in different

calibration chambers and under selected boundary conditions is not yet available.

However, it is possible to com pare the present test results w ith the previous results

reported on the same sand. Such a comparison is presented in the next chapter.

5.3 Influence o f T esting Variables on C one Test R esu lts

5,3.1 Uncemented Specimen Results

Figures 5.15 and 5.16 shows the tip and friction resistance versus effective ver

tical confining stress a t different relative densities of uncem ented specim ens. Both

tip and friction resistances increase with the effective vertical stress. H igher vertical


169

0.0fs (k g /sq .c m )

1.0 2.0 J.O

Î 20

Dr R a nge 8 6 » 80 p e r c e n t

Dia. R a t io 42 100 kPa

100qc (k g /sq .c m )

200 300 400

6o

a,(DQ

40

Figure 5.14: R epeatability of the Results


170

Table 5.4: C haracteristics of Tests (C.C. 2 %)

Test No. P>r height(m) [kN/rrP)

^hi/ho{kNjrrP)

Ko

1 8&2 0.70 100 54.12/ 51.3 0.542 8&3 0.71 200 87.1/89.3 0.443 8 4 ^ 0.70 300 120.4/123.02 0.404 6&6 0.71 100 41.8/ 41.8 0.425 6&2 0.71 200 95.4/95.7 0.486 72.1 0.72 300 121.44/120.3 0.407 4T2 O j# 100 43.84/44.1 0.448 54.4 0.73 200 76.9/70.0 0.359 52.0 0.71 300 185.6/174.2 0.6210 8&2 0.72 65 65/65 1.0011 8 4 ^ 0.71 - - (collapse)

stress in the specim en cause the specim en to offer m ore resistance to the penetration

of th e cone, resulting in higher tip resistances. The effect of relative density on tip

resistance can also be deduced from the same figure. The higher the relative den

sity, th e closer the sand packing and the contact points betw een the grains which

result in higher tip resistances. Sim ilar observations are noted in the case of friction

resistance.

M ost of the tests are conducted a t location A, the center of the specim ens unless

specified. In an a tte m p t to study the influence of location, an uncem ented specimen

of relative density of 86 % is prepared. T he test was conducted a t location B which is

closer to the edge. T he tes t results are com pared w ith test results on a specim en of 90

percent relative density which are conducted at location A. B oth the specim ens are

consolidated under an effective vertical stress of 100 k P a and tested under boundary

condition 3.

F igure 5.17 com pares the results conducted a t these two locations. It is in terest

ing to note th a t location has m inor influence on the final tip and friction resistances;

however, a t the shallow penetration depths along the specim en, there is a m arked

difference. T he variation is negligible a t higher penetration depths. These m inor

differences are possibly due to both relative density and also coupling of the tip

resistance w ith sleeve friction. T he soil pluviating near the edge of the sieve may

result in a different density than in the middle. Boundary effects may have also led


171

Table 5.5: P enetra tion Results (C .C. 1 %)

Test No. Dr CTv Ko 9c Is f.r.{ k N /w ? ) (ksc) {ksc)

1 86.0 100 0.54 16&0 0.95 0462 &L6 200 1.00 254.0 1.55 0.643 8&2 300 0.49 33&0 1.98 0484 6&4 100 0.37 131.0 0.74 0.565 66.4 200 0.63 191.0 0.86 0.456 70.2 300 0^4 2W&0 1.21 0^47 4R8 100 0.36 68.0 0.20 0498 4&6 200 0.41 974 0.72 0429 5&2 300 0.57 104.0 1.00 0.97

Note: \ k s c — Ikgfcrn^ = 100 Pa

Table 5.6: P ene tra tion Results (C .C. 2 %)

Test No. Dr{kNIm?)

Ko 9c(ksc)

fs(ksc)

f.r.

1 884 100 0.54 202.0 1.10 0.542 864 200 044 2454 1.75 049

3 * 844 300 0.40 350.0 2.11 0.604 694 100 0.42 143.0 0.80 0.555 694 200 048 220.0 1.20 0.546 72.1 300 0.40 307.0 1.31 0437 474 100 0.44 804 0.40 0.508 544 200 0.35 120.0 0.75 0429 524 300 0.62 150.0 1.01 0.6710 864 65 1.0 - - -

Note: 1 ksc - 1 kg/crri^ — 100 kPa

* ' Test Stopped


172

( V

ü

mmCDIL,

■+JUl

ao•I—Ik(D

>

(D>

•rH-t-JO0q-,

t+MM

0qc (kg/cm^)

200 400 6000

0 0 0 0 0 84 < Dr < 90□EXOO 65 < Dr < 75 AAAAA 45 < Dr < 55

D/d 42

2

3

4

Figure 5.15: Tip Resistance vs Effective Vertical Stress for Various Relative Densities


173

OJ

sO

fcJX)

mK0)

4-)CO

cdo

•rH-tJ!h0)

>

O>

• I—I4-3o0)

«44« 4 4W

0.0 1.0 2.0 3 .00

OOOOO 84 < Dr < 90□□□□□ 65 < Dr < 75A A A A A 45 < Dr < 55

D/d 4 2

2

3

4

Figure 5.16: Friction Resistance vs Effective Vertical Stress for Various Relative

Densities


174

f, ( k g / c m )1 2

Center, BC3 Location B, BC3

80

qo (k g /c m ^ )100 200 300 400

Center, BC3 Location B, BC3

Dr 8 6 - 9 0 % D /d = 42CTy 1 k g / c m

60

80

Figure 5.17: Influence of Location on Test Results


175

to th is dilFerence in sleeve friction. At a confining pressure of 100 kPa^ a change of

2 degrees in the in ternal friction angle will result in approxim ately a difference of 25

k P a in sleeve friction (0.25 kgjcm?').

5,3.2 Cemented Specimens

Figures 5.18 and 5.19 present the influence of relative density on tip resistance

and friction resistance of cem ented specimens (1 %). Similar to uncem ented sands, an

increase in relative density results in an increase in both tip and friction resistances.

Figures 5.20 and 5.21 com pare the tip and friction resistances of cem ented and

uncem ented sands. Cem ented sands induce higher tip and friction resistances, a t

trib u te d to the cem entation bonding. These bonds bring abou t a cohesive property

to the sand which significantly increases its strength . T he higher the streng th of

these cem ented sands the higher will be the tip and friction resistances.

T he increase in cem entation results in a drop in friction ra tio (Table 5.5 and

5.6) due to the m ore pronounced increase in tip resistance.

5.4 Sum m ary

A to ta l of th ir ty four (14 uncem ented and 20 cemented specim ens) calibration

cham ber tests were conducted using a m iniature cone penetrom eter. These results

will be used in th e subsequent chapters in preparing em pirical and semi-empirical

correlations.

Piezocone penetration te s t on a specimen of 2 % cem entation showed th a t there

was no excess pore pressure m easurem ents. This implies drained conditions still

prevail in cem ented specim ens used in th is study and the decrease in hydraulic con

ductiv ity is not sufficient to cause undrained conditions during cone penetration .

T ip and friction resistances increase w ith effective vertical and horizontal con

fining stresses and also w ith relative densities. The tip and sleeve friction resistance

are found to be repeatable. Testing location has minor influence on tip resistance.

Increase in cem entation results in an increase in tip and friction resistance and

a decrease in friction ratio. T he increase is due to evolution of cem entation bonds

betw een grains, resulting in cohesion which in tu rn increases the streng th of the

specim ens. This increase in streng th offers m ore resistance to the penetrating cone.


176

qc ( k g / c m ^ )100 200 300 400

mmmam Dr 48.8 % A zàddkA . Dr 68.4 % « » « •« Dr 86.0 %

20

60

cTy - 1 k g /c m D /d - 42; BC3

80

Figure 5.18: Influence of Relative Density on Tip Resistance of Cem ented Specimen

(19%


177

fs (kg/cm^)0 3

0

BBBBB Dr 48.8 % A écA ràdk. Dr 68.4 % 9 9 B & 9 Dr 86.0 %

20

60

CTy - 1 k g / c m D / d - 4 2 ; BC3

8 0

Figure 5.19: Influence of Relative Density on Friction Resistance of Cemented Spec

im en (1 %)


178

qc (k g /cm ^ )0 200 300 4001 0 0

0

c.C. 2 %

20

60 Dr - 85 %a, - 1 kg/cm ^ D /d - 42; BC3

80

Figure 5.20: Influence of Cem entation on Tip Resistance


179

fs ( k g /c m ^ )2 30

0

20

60 Dr - 8 5 %

Qv - 1 kg/cm ^ D /d - 42; BC3

80

Figure 5.21: Influence of Cem entation on Friction Resistance


180

hence higher tip and friction values are observed. Increase in tip resistance is m ore

pronounced th an friction resistance, resulting in lower friction ratios.


Chapter 6

FACTORS INFLUENCING TEST RESULTS

6.1 In troduction

A t the present s ta te of the a rt and despite significant achievements in calibration

cham ber research, there are num ber of concerns associated w ith engineering in te r

p re ta tion of insitu tests and the use of cham bers in developing an understanding of

insitu testing devices (Ghionna and Jamiolkowski, 1991). Generally, cham ber tests

are conducted on specimens of freshly reconstitu ted specimens whose fabric is dif

ferent from those of the natu ra l soil deposits. The structu re of a na tu ra l deposit is

highly developed as a result of cem entation, drained creep and diagenesis (G hionna

and Jamiolokowski, 1991). Hence in terpretations developed based only on tests con

ducted on clean sands m ay not be valid for certain natural deposits, specifically when

there is cem entation.

A nother concern arises when the various factors th a t influence cham ber test

results are considered; soil characteristics like grain size, com pressibility and relative

density and also cham ber param eters such as diam eter ratio , boundary conditions

and the size of the cone significantly affect the test results. Several studies have been

a ttem p ted to understand the above problem s w ithout reaching definitive conclusions

(Eid, 1987; G hionna and Jamiolkowski, 1991).

P resent study was conducted on both cemented and uncem ented sand speci

mens. A diam eter ratio of 42 was achieved by using the m iniature cone (M QSC).

These results are used in providing an assessment of the effect of different variables.

Uncem ented sand results are first compared w ith various relative density-cone resis

tance curves. A chart for estim ating the relative density for M onterey No. 0/30 sand

is also presented. Various factors affecting cham ber results are evaluated. D ata from

seven different calibration cham ber investigations along w ith present test results are

used in th is study.

181


182

6.2 R elative D en sity

T he tes t results of th is study are com pared w ith various relative density charts

proposed by different investigators. T he findings and observations from these com

parisons are discussed in the following sections.

6.2.1 Comparison With Results Reported by Villet and Mitchell (1981)

In an a tte m p t to investigate the variability in results from one calibration

cham ber to another, a prelim inary study is conducted. Villet and M itchell (1981),

Schm ertm ann (1978) and Baldi (1981) have reported extensive studies w ith the cal

ibration cham ber. Com parisons are m ade w ith V & M ’s results since they have used

M onterey No. 0 sand. T he results of this study a t relative densities of 45 - 55, 65 -

75 and above 80 % are p lo tted on the V &: M ’s relative density prediction chart for

com parison purposes (Figure 6.1).

V &: M tests are conducted in a pressure cham ber of 80 cm (32 in.) in height

and 76 cm (30 in. ) in d iam eter. A cone penetrom eter of 10 cm^ tip area (3.56 cm

d iam eter) and a base apex angle of 60 degrees is used. T heir d iam eter ratio is 20.

Specimens are consolidated under Ko conditions. A fixed Ko value of 0.5 was used

for their tests. Horizontal stress of m agnitude equal to half of th e vertical stress was

m anually applied to the specim en, thereby rendering a Ko value of 0.5.

T he results obtained in th is study are system atically lower than those reported

by V & M. T he au thor offers th e following reasons for th is difference:

1. Tests by V &: M are conducted under constant stress boundary conditions. The

Ko values used by Villet and Mitchell were around 0.5, substantially larger than

th e Ko values obtained in th is study. Note th a t horizontal stresses (equivalent

of Ko conditions) are m anually applied in V &: M study, unlike the ones obtained

in the present study, which are m easured for zero lateral s tra in conditions.

2. V & M used a conventional cone of 3.56 cm in diam eter while this study used

a 1.27 cm diam eter cone. I t is not well established, how this factor will indi

vidually affect the results. Studies conducted by Eid (1987) show th a t there

is a scale effect in the cone tes t results based on the size of the cone used.

Therefore, the results from other chambers cannot be used as a yardstick in


183

>bmxn0)

-+->C/2

03O

-rH-P

CD>

CDi>•iH

4-)ÜCD

P h4-1W

0Tip R es is tance , (MPa)

20 40 600 1— I— I— I— I— I— I— I— I— I— I— I— I— I— I— I— r

100 -

200 -

300 -

P resent StudyDr 5 0 % Dr 7 0 %

mmsm Dr 90 %

Dr 70 90Villet & Mitchell (1981)

400

Figure 6.1: Comparisons with V and M’s Results


184

evaluating the variations in the present study due to the differences in testing

and cham ber param eters.

3. T he d iam eter ra tio in Villet and M itchell is 20, significantly smaller th an the

d iam eter ratio of 42 used in th is study. In dense sands, increases substan tia lly

w ith th e decrease in diam eter ratio. Therefore, the results reported by V & M

m ay have been significantly influenced by the boundaries.

6 .2 .2 C o m p a r is o n s W i th R e s u l ts R e p o r te d fo r O th e r S a n d s

Com prehensive calibration cham ber studies with a sand are also reported by

Schm ertm ann (1978) and Baldi (1981). The sands used in these studies are different

and are classified as low to m edium compressible. Schm ertm ann’s results are sta ted

to be valid for the Fugro cone in norm ally consolidated, sa tu ra ted , recent, unce

m ented , fine SP sands. The sand is defined to be subrounded to angular. B aldi’s

resu lts are valid for T icino sand (Italy). This sand is sub-angular to angular in shape.

T he Ko values of B aldi’s tests range from 0.37 to 0.46. Dry pluviation m ethod was

used in bo th studies for specimen preparation. Both studies were conducted at a

dia.meter ra tio of 34.

T he comparisons are shown in Figures 6.2 and 6.3. The uncem ented tes t results

ob tained in th is study are p lotted on B aldi’s and Schm ertm ann’s results from the

relative density-tip resistance-vertical stress curves for comparison purposes. It is

no ted th a t th e results reported by Schm ertm ann deviate more a t higher confining

stresses (greater th an 150/cFa). The boundary conditions, the physical character

istics of the sand in Schm ertm ann’s results are significantly different from those of

th is study. In addition, the lower diam eter ratio (34 versus 42 used in th is study)

also influences these results. The comparisons w ith B aldi’s chart yield the following

conclusions. In case of loose specimens, the results seem to be closer, however there

is some variation at higher densities. It is interesting to note th a t in the case of lower

densities, th e results are not as affected as those a t higher densities by the diam eter

ra tio , boundary conditions and probably the grain size and shapes. D iam eter ratio

of B ald i’s test results are lower than the present test results (34 versus 42) and also

these tests are conducted at constant stress boundary conditions (B C l). These vari

ations, according to previous investigations (Been et ah, 1986) will influence cone


185

0Tip R esis tance , (MPa)

20 40 60cd

>b

mtn0)u

-4 - >

cda

-r-t-PSh(D

(U>

“pH4-)Ü0

CWP hM

0

P resent Studyà Â Æ à A Dr 5 0 55 ©@©©© Dr 7 0 55

Dr 90 %100

200

300

Dr 5056

Baldi (1981)

400

Figure 6.2: Comparisons with Baldi’s Chart


186

Tip R es is tance , (MPa)6020 400

cdpH

P resent Study

>b Dr 90M 100 CO 0 U

H->CO

U 200• pH4->

0^ 300O0

w

90Dr 50% 70Schm ertm ann (1978)

400

Figure 6.3: Comparisons with Schm ertm ann’s C hart


187

test results a t higher densities. A nother factor th a t m ight have affected the results is

the shape of the grain. Ticino sand of B aldi’s study is m ore angular th an M onterey

No. 0/30 sand which is subrounded to subangular in shape.

The results obtained indicate th a t it is quite difficult to com pare the penetration

results obtained in one cham ber w ith one type of sand w ith ano ther due to the

effect of boundary conditions, diam eter ratio , differences in grain shape, size and soil

com position and specim en preparation. The d a ta obtained in this study seems to

be w ith in th e variability of the results reported by the two previous studies (Baldi,

1981; Schm ertm ann, 1978).

I t is clear th a t tip resistance is influenced by several factors. Sands of different

grain size and shape behave differently when tested under identical conditions. Ad

ditional testing variables like cham ber size and boundary conditions will influence

the results. A fu rther a ttem p t is m ade to collect the penetration d a ta on the sands

and study the influence of the above m entioned factors.

6.3 Factors Influencing P en etration R esistan ce in Calibra

tion C ham ber

T he cone results reported by different investigators are compiled in Appendix

B. These results are used to conduct an assessment of the complex problem s facing

calibration cham ber testing. In an a tte m p t to bring a form alism to the effect of

different calibration cham ber test results, available d a ta are compiled in groups. In

analysis of the effect of one variable, care was given to select the d a ta which were

obtained under sim ilar conditions and com positional variables. For exam ple, d a ta

obtained from the sands of sim ilar size and shape, com pressibility and tested under

sam e boundary condition are used to analyze the influence of cham ber size effects.

6 .3 .1 C h a m b e r S ize

Since the early eighties, the influence of the cham ber dimensions and configura

tion on penetration results are regarded as the fundam ental variables affecting the


188

results (G hionna and Jamiolkowski, 1991). Holden (1971) dem onstrated th a t the

cham ber size significantly affects the results in dense sands.

T he effect of cham ber size on the m easured results is la ter addressed by other

investigators (Park in and Lunne, 1982; Jamiolkowski e t.a l., 1985; Bellotti et. al.,

1985; E id, 1987; Schnaid and Houlsby, 1990; M ayne, 1991). Parkin (1988) explained

th a t a diam eter ratio of 50 will be needed when dense sands are tested . He also

found ou t th a t for loose sands, even a t low diam eter ratios (around 21), tip resistance

results are independent of boundary conditions. A lthough, a system atic study is not

yet available, higher diam eter ratio will be required testing dense sands in calibration

cham bers. This will be specifically tru e a t lower confining pressures when the sand

could d ila te during shearing.

M ayne (1991) observed th a t the cone da ta tested under flexible wall calibration

cham bers are less th an they would be in an infinite m edium. M ayne (1991) suggests

th a t yielding which occurs while testing under flexible wall calibration cham bers may

be th e reason for th is decrease (M ayne, 1991). T he author notes th a t cem entation

m ay also be another factor. M ayne (1991) further suggests th a t tip resistances

m easured in cham bers need to be corrected for the cham ber effects. M ayne (1991)

proposes a correction factor th a t takes care of such cham ber effects. This factor was

derived based upon exam ination of six available tes t d a ta on sands and is purely of

em pirical nature.

/Is . _ ^Ç c ,c o r r — Ç c ,m e a s I j (^-f)

T he correction factor, defined as the ra tio of corrected to m easured tip resistances

ranges from 0.4 to 0.9 for diam eter ratios of 15 to 40 and for relative densities of

20 to 100 %. For denser specimens (above 80 %), the correction factors are 0.4 to

0.6 which imphes th a t the m easured resistance need to be alm ost doubled to get the

corrected value.

T he accuracy of these results is debatable and the validity of the above equation

raises several following concerns. T he results used in the above study are not nor

m alized w ith respect to horizontal and vertical stresses a t which they are obtained.

Instead , they are reported as values for an average effective vertical stress of 150 k P a

and a Ko value of 0.43, thereby neglecting the influence of these stresses on the tip

resistances.


189

This type of analysis clearly ignores the influence of effective vertical and hor

izontal stresses on the results. Since, tip resistance increases with both horizontal

and vertical effective stresses, the tip resistance of each tes t needs to be normalized

w ith both vertical and horizontal stresses. These norm alized values have to be used

to study the influence of cham ber size effects. T he influence of horizontal stress on

(jc is sim ilar to vertical stress (F igure 6.4). Hence, norm alization has to be done

by tak ing into account th e horizontal effective stress. Influence of horizontal and

vertical effective stresses is incorporated by norm alizing w ith respect to octahedral

stress. Poet- T ip resistance is norm alized as the ratio of {(jc — Poct) to Œaim where (p,

is the m easured cone resistance; Poct is to tal octahedral stress ) and cr„(„, is

atm ospheric pressure. Effective octahedral stress, p' ^ is norm alized by taking the

ratio of to Oatm- T hen, those normalized values can be correlated in the following

form,

= a , (6.2)^ a i m a i m )

where Og and rio are the param eters th a t include tlie influence of relative density,

grain size, shape, fabric and com position (soil param eters), d iam eter ratio and bound

ary conditions (cham ber param eters). The subscript 'o ’ denotes the octahedral stress

norm alization. The following procedure is used in evaluating the effect of different

param eters on Og and 7i„.

1. Eight different results reported for sands, including the present test data are

used. Physical properties of these sands and the cham ber param eters in which

they are tested are presented previously in C hapter 2. The tip resistance values

are norm alized w ith respect to«TfilTrj '

2. Similarly, norm alized effective octahedral stress, ( 4 ^ ^ I values are calculated■' ' y J

in all th e tests,

3. Logarithm of these norm alized values are com puted,

4. T he logarithm ic values are correlated w ith each o ther,

5. A best fit straight line is passed through each set of relative density range,


190

0.0cTh (k g /cm ^ )

0.5 1.0 1.5 2.040040

▲▲▲▲A 84 < Dr < 90 %

300

200

m(D

1 0 0

1501 0 0 200500

CVi

Ü

bC24

oc r

H orizontal Effective S tress (kPa)

Figure 6.4: Tip Resistance Versus Horizontal Effective Stress (Present Study)


191

6. The y-axis in tercept, log Ao and the slope of the line, «o, are calculated, and

the values a (or Oq) and n (or iio) thus obtained are tabu la ted in Table 5.3.

Figures 6.5, 6.6, 6.7 and 6.8 show such plots for various investigations. ITom

each figure, a and n are estim ated . These are reported along w ith o ther details of

these studies in Table 6.1. T he m ean and standard deviation of the n values are

calculated. The m ean value varies betw een 0.55 to 0.92 with a s tandard deviation

value of 0.03 to 0.22. M ost of these investigations have low standard deviation values

which implies th a t n may be regarded constant a t all densities for a particu lar type

of sand. T he n value seems to be a property of the soil a t a s ta te a t which there is

no volume change (critical s ta te ). This m ay be the reason for n being independent

of relative density. After calculating the average and standard deviation values of n,

lines are passed through the sam e da ta points in these figures a t slopes equivalent to

nmean ± 1 Standard deviation. T he new a values (denoted as O j and a -2 in Table 6.1)

are determ ined and the average of these values are taken as the corrected q value.

These corrected values are determ ined for each density in all investigations and they

are reported in the sam e table.

Influence of cham ber size is studied by plo tting the mean o values on the y-

axis and diam eter ratio on the x-axis (F igure 6.9). For each set of relative density,

a best fit curve is obtained. This figure signifies the im portance of cham ber size

dim ensions on cham ber test results. The influence of cham ber size (d iam eter ratio)

on a is not as evident a t lower densities as in the case of higher densities, ddic

d iam eter ratio influences the results on denser specimens and influence decreases

w ith a decrease in relative density. Hence, it is recom m ended to use higher diam eter

ratios for calibration cham ber tests on dense sands. A d iam eter ratio of above 40 is

needed to reduce the cham ber size effects on dense specimens (above 80 %). For the

specim ens of above 90 % relative density, d iam eter ratio of above 50 is recom mended.

6.3.2 Compressibility and Crushability of the Sand Tested

Bellotti et a l.(1991) shows th a t significant crushing takes place in cone penetra

tion tests perform ed in calibration cham bers which in tu rn affect the C P T results.

This problem will be more evident, if the sand used in the cham ber has low com-


CD■DOQ .C

gQ .

■DCD

C/ÎC/)

8■D

CD

3.3 "CD

CD■DOQ .C

aO3"Oo

CDQ .

■DCD

C/)C/)

Table 6.1: The a and n for Various Investigations

Study(BC)

Sand dso (mm)

Dr D.R. a nk )

Ol(nm+cr)

Û2(nm-cr)

a{a i+ a 2 )/2

EID (1987) 0.45 25 17 98 0.90 97.9 94.0 95.9(B C l) SR-SA 25 42 70.8 0.79 0.75 71.8 6&2 70.5

60 34 141 0.48 (0.16) 128.7 137.9 133.360 42 229 0.72 219.7 235.5 227.660 64 302 0.87 29&9 320.4 309.7

Baldi (1981) 0.39 45 34^ 88.7 0.57 8R7 8&7 88.7(B C l) SA-A 75 34.2 166.3 0.57 0.55 166.3 166.3 166.3

90 34.2 253.5 0.52 (0.03) 250.2 253.5 251.8Fior(1991) 0.16 45 60 70.5 0.54 92.8 71.3 82.1

(B C l) SA 70 33.4 190.5 1183 0.78 195.0 183.5 189.380 60 242.0 0.98 (0.22) 2H&0 235.9 238.9

Pupp(1993) 0.45 45 42 61.7 0.70 59.1 61.5 60.3(BC3) SR-SA 65 42 12&8 O j# 0.84 127.5 131.3 129.4

87 42 186.2 0.95 (0.12) 186.4 180.6 18&5VM(1981) 0.45 30 20 70.8 0.95 0.92 70.4 73.5 71.9

(B C l) SR-SA 60 20 121.3 QjW (0.04) 119.7 121.6 120.6Lhuer(1976) 30 34.2 37.1 0.79 37.1 36.7 36.8

(B C l) 65 34.2 125.8 0.67 0.72 126.7 125.7 126.280 34.2 1&T8 0.71 (0.06) 15&3 153.7 155.0

Harm(1976) 30 34.2 40.6 0^8 40.7 39.6 40.2(B C l) SA 65 34.2 13&4 0.80 0.76 139.5 134.6 137.1

80 34.2 200.9 0.68 (0.07) 211.6 201.6 206.6N utt(1991) 0.24 30 34.2 39 0.66

Note: a - S tandard Deviation

- Average n; SR - Subrounded; SA - Subangular; A - Angulartoto

193

3.00

B ald i (1 981)

2.50

B

2.00

1.50QGCDD45 & 34

A A A A A 90 & 34

1.001.00-0 .5 0 0.00 0.50

l0 ê (P ocl/âlm)

3.00

T his S tu d y

2.50

5

2.00

CTW

1.50 45 & 42 65 & 42 87 & 42

1.00-0 .5 0 -0 .3 0 - 0 .1 0 0.10 0.30 0.50

^®ê(P ocl/oâlm)

Figure 6.5: Normalized Tip Resistance Versus Normalized Effective Octahedral

Stresses


194

3.00

Lhuer (1976)

2.50

a■fl

fti2.00

1.50 -c œ œ 3 0 & 3 4 .2 n r r i T i 6 5 & 34.2 a a a a a 8 0 & 3 4 .2

1 .00,-0 .5 0 -0 .3 0 - 0.10 0.10 0.30 0.50

10§(P ool/âtm)

3.00

Harman (1976)

2.00

bO 1.00 Dr & D.R.

0 0 0 = 0 3 0 & 3 4 .2 □ m m 6 5 & 3 4 .2 A v w \ Rn & 34.2

0.000.60 - 0.20 0.20 0.60

10g(Pocl/ciaün)

Figure 6.6: Normalized Tip Resistance Versus Normalized Effective Octahedral

Stresses


195

3 .00

N utt, 1991 (30* Dr)

2.50

1b

2.00

bC1.50

1.00,0.30 0.50-0 .5 0 0.10-0 .3 0

3.00

Villet & Mitchell (1981)

2.50

ob

2.00

Or U D.R.OC000 30 & 20 m i l 1 60 & 20

1.50

1.00,-0 .5 0 0.30 0.10 0.30 0.50

Figure 6.7: Normalized T ip Resistance Versus Normalized Effective Octahedral

Stresses


196

3.00

FioravBnle (1991)

2.50

Ib

aI

saso

2.00

Or & D.R. n rrm 45 & 60 iT n n 70 & 33.4 a a a a a so & 33.4

1.50

1.00-0 .5 0 - 0.10 0.10

3.00

Eid (1907)

2.50

Ib

2.00

Or & D.R.1.50 r r r m so & 64

r r n n 6o & 42a a a a a 60 & 34

25 & 42

1.00,-0 .3 0-0 .5 0 0.10 0.30

F ig u r e 6 .8 ; N o r m a liz e d T ip R e s is ta n c e

S tr e s se s

V e n u s N o n n a U a d E f b c ü v e O c t a l e d n J

Reproduced with permissionof the copyright owner. Further reproduction prohibited without permission.

197

cd

A

2 0 0

1 00

D.R. 60 D.R. 50

42

060 80 1002 0 40

Relative D ensity (p ercen t)

Figure 6.9: Influence of D iam eter R atio on a for Various Densities


198

pressibility. T he crushing of the sand around the cone m ay have m ore pronounced

effect on friction resistance as this value will be m easured after crushing takes place.

Several investigators noted the im portance of particle crushing which results in an

increase in com pressibihty (Bellotti et ah, 1991). The o ther im portan t effect is a

m ore pronounced curvature of the failure envelope up to an u ltim ate value of friction

angle which would be unaffected by fu rther crushing. T he combined effect of all

these factors is a reduction in tip and friction resistance.

B ellotti e t al. (1991) tested four different sands and evaluated their crushability

indexes from the grain size distributions of the sands tested at different pressures

in one-dim ensional compression tests. C rushability indexes are proposed in various

form s by various investigators (Bellotti e t ah, 1991). These indexes express the

am ount of crushing of the aggregates th a t took place during cone penetration . This is

done by calculating the ratio of the am ount of the tested sand passing in a particu lar

standard sieve to th a t of untested sand. Crushing m easurem ents in C P T exhibited

a well defined correlation between crushing am ount and cone resistance (Bellotti et

ah , 1991). These correlations are independent of consolidation stress, relative density

and overconsolidation ratio (Bellotti et al., 1991).

T he sand used in the present study can be regarded incom pressible even though

it contains some m ica (less than 1 %). The present results are com pared w ith the

results obtained in Ticino sand to assess the influence of compressibility. Since the

crushing of sand particles is observed in dense specimens (above 84 %), only dense

specimens are considered for this analysis.

T he sand around a radial distance of 3.5 cm around the cone was collected

subsequent to penetration and a grain size d istribution analysis was conducted. A

com parison of the grain size d istributions is presented in Figure 6.10. T he sand

tested under 200 a and 300 a shows some variation in gradation. The results

also dem onstrate th a t there is substan tia l crushing of the m aterial retained on ASTM

sieves No. 30, 40 and 50.

T he au tho r also observed crushed sand powder all along the penetration profile.

The sand collected from under the cone was passed through an ASTM No. 200

sieve. A coefficient nam ed as crushability under cone (CUC) is used to m easure the

crushability. CUC is defined as the ra tio of the weight of sand passed through a No.

200 sieve of a penetration tested sand per 1000 ^m to th a t of an untested sand.


199

100

60

40

20

Dr, Untested

0.01 0.1 1 10 Grain Size (mm)

100

Figure 6.10: Grainsize D istributions of the Crushed Sand A round the Cone


2 0 0

A CUC value of less than 5 implies sand is m oderately compressible, between 5

and 10 is known as of m edium com pressibility and above 10 it is term ed as highly

compressible. CUC depends upon the influence zone from which th e penetrated sand

is collected. In the present study, sand is collected from a zone of th ree tim es the

penetrom eter d iam eter around the tested location. The CUC values obtained from

th e sand collected in tests under vertical consolidation stress of 200 and 300 k P a are

2.8 and 3.8, respectively. T he low CUC values suggest th a t the sand has m oderate

com pressibility; therefore the crushing around the cone will decay significantly as

one moves away from the cone.

The tip resistances m easured under these tests were p lotted in the F igure 6.11

along with the d a ta of Quiou sand, a subrounded to subangular sand (A lm eida et

al., 1991). These results, although few, suggest th a t crushing does not significantly

affect th e tip resistance around the cone in M onterey No. 0/30 sand. However, there

is significant crushing taking place in the case of Quiou sand. A possible reason for

th is difference is the size of th e cone (3.57 cm cone was used in the study on Quiou

sand).

6.3.3 Boundary Conditions

The four trad itional boundary conditions used in cham ber testing were described

earlier. Park in and Lunne (1982) s ta te th a t the penetration resistance in the field

will lie betw een a CC tes t w ith a zero la tera l stra in boundary (BC3) and CC test

w ith a constant lateral stress boundary (B C l). The zero strain boundary condition

overestim ates Qc, as higher stresses will exist a t cham ber boundary than in the field

a t an equivalent distance from th e cone. Conversely, the constant stress condition

underestim ates the Qc as higher stresses m ay develop in the field a t an equal distance

from the cone (Been et ah , 1988).

Figure 6.12 shows th e zero stra in and constant stress da ta converging when the

cham ber d iam eter is g reater th an 50 tim es the cone diam eter. Boundary conditions

have little influence on the results in loose sands. T he present tes t results which are

conducted under boundary condition 3 are used to discuss the boundary condition

effects. T he results from two o ther studies, V illet and M itchell (1981) and Eid (1981)

are also used. These tests were conducted under boundary condition 1. T he change


201

40

( £ 3 0

0)ü

2 2 0 03030)

K

aH 10

1—m —rn —i—i—i—i—|—i—i—i—i—i—i—i—i—i—;—i—i—i—i—i—i—i—i—r

Cone Dia 1.27 cm , This Study Cone Dia 3.57 cm , Quiou Sand

Q I I I I I I I I I I I I I I I I I I I I I I I I I I I ' I

0 5 10 15P e r c e n t P a ss in g ASTM No. 200

Figure 6.11: Influence of Crushing on T ip Resistance


2 0 2

in a w ith relative density is presented for these two studies in Figure 6.13. B C l

induces higher a. values than BC3; tip resistances being higher in BC3. Higher

octahedral stresses are observed in BC3 than in B C l. Zero lateral stra in boundary

induces higher lateral stresses resulting in higher octahedral stresses thereby reducing

a values. These investigations have different diam eter ratios and their influence on

the above results are not accounted for in the above discussion.

The influence of boundary condition on test results is investigated by conduct

ing cone tests on two identical cem ented specimens subjected to different boundary

conditions; namely the constant stress (B C l) and zero lateral s tra in (BC3) condi

tions. Specimens are prepared a t 45 to 55 % relative density range at a consolidation

pressure of 300 fcPa. T ip and friction resistance results are shown in Figure 6.14.

T he two tests produce sim ilar results, verifying the indifference in tip resistance to

boundary conditions a t higher d iam eter ratios (42) in the present tests. The larger

d iam eter ratio implies the cone test results are least affected by the boundary condi

tions. This implies th a t the testing rendered results sim ilar to th a t of field conditions

of semi-infinite medium.

6.3.4 Influence of Soil Particle Size and Shape

It is an established fact th a t sands from different locations will not behave

identically when tested under sim ilar conditions. This is a ttr ib u ted to the size and

shape of the particles which lead to changes in dilational characteristics. I t was

suggested by G hionna and Jamiolkowski (1991) th a t significant studies have to be

done to understand the effect of these param eters on penetration . An a tte m p t is

m ade in th is section to evaluate the various test results conducted on different types

of sands. Figure 6.15 was prepared by plo tting n on the y-axis and the ratio of cone

diam eter to dso of the aggregate on the x-axis. The n value varies between 0.60 to

0.95 w ithout exhibiting any trend.

The n value seems to be m ore influenced by the size of the aggregates and the

diam eter of the cone. An increase in the ratio of the diam eter of cone to d^Q results

in an increase of n. It can also be concluded th a t the n value is not influenced

by the relative density and the stresses th a t the specim en is subjected to. The

effect of grain shape on n can not be deduced since sands used in th is analysis are


203

3 0

ËSjg«A

S .»coü

4 0 6 02 00Oiom©î®f Refm

Figure 6.12; Boundary Condition Influence on CC Results (Parkin, 1985)


204

400

Q0COO D /d = 42, M onterey No. 0 /3 0 □BBBQD/d = 40, T icino Sand

o e ©0OBCl, This Study GEBBG BC3. Baldi (1981)300

1 0 0

80 1 0 02 0 40 600Relative D ensity (p e r c e n t )

Figure 6.13: Boundary Condition Influence on CC Results


205

q= (k g /cm ^ )100 200 300 4 0 0

BC 3

Cone Dia 1.27 Dia. Ratio 42

f , (kg/cm ® )

BCBC

20

0 .4 0

60

Cone Dia 1.27 Dia. Ratio 42

Figure 6.14: Influence of Boundary Condition on Cone Test Results


206

Table 6.2: T he a and n for Various Cem entations

C.C.(BC)

Dr D.R. a n Tljjik )

Û2 a( o n + o ^ /2

C.C. 1 45 42 8&2 0.32 85.2 8&3 84.3(BC3) 65 42 166.0 0.48 0.42 16&8 170.6 16R7

85 42 208.9 0.47 (0.09) 20&8 211.9 210.9(1C . 2 45 42 102.3 0.48 99.3 102.3 100.8(BC3) 65 42 195.0 0.67 0.57 195.2 199.5 19T4

85 42 251.2 0.58 (0.09) 24R4 254.7 251.5

Note: cr - S tandard Deviation; C.C. - Cem ent C ontent

subrounded-subangular to angular in shape.

6.3.5 Cementation

Few studies in the lite ra tu re reported testing cemented sands (Rad and Tumay,

1986; Akili and Nabil, 1988). The available studies were conducted in rigid chambers

a t low confining pressures (less than b k P a ) . In order to study the influence of

cem entation, tests have been conducted on artificially cem ented specimens at various

confining pressures. These results have been reviewed in C hapter 5.

T he influence of cem entation is investigated by determ ining a and n values in

the tes t results obtained in cem ented specimens (Table 6.2). Figures 6.16 and 6.17

present the norm alized results a t bo th cem entation levels. The a and n are calculated

from these figures and are shown in Table 6.2.

These results are p lo tted com paratively w ith those of uncem ented results in Fig

ures 6.18. T he a value clearly increases w ith cem entation a t each relative density.

This increase is due to the increase in tip resistance w ith cem entation. However,

cem entation influence on n is not clear except th a t n decreases w ith cem entation

a t each relative density. In brief, cem entation reduces the relationship betw een nor

m alized tip and octahedral stresses into non-linear form. The following non-linear

relationship is obtained between cement content and n (Figure 6.18).

= 0.84 - 0.90 In ( l + 0.56 (C.C.)°-^^) (6.3)

This relationship dem onstrates th a t results will be strongly affected by slight

increases in cem entation levels. T he au thor notes th a t although there is an increase


207

0.00

1.00

0.90

0.80

0.70

0.60

C 0.50

0.40

0.30 P

0.20

0.10 P

A - 34.2(D.R.) B - 42.0

Best Fit for rim values C - 60.0

1 I 1 I I I I t 1 I 1 I I 1 I 1 1 1 1 l - J - I I I l - J 1 - 1 - I I I L _ I _ _ I _ L I I 1.

20 120 220 320 420C one D ia m e te r / d 50

Figure 6.15: Influence of Cone D iam eter to dso ratio on n


208

cdb

uoA

ucr'bOo

3.00

Cem ent Content 1

2.50

2.00

Dr & D.R.85 & 42

AAiftAA 65 & 42 45 & 42

1.50

1.000.30-0 .3 0 0.50-0 .5 0

Figure 6.16: Normalized Results on Cemented Specimen (C.C. 1 percent)


209

ab

uoAIucr

"feOo

3.00

Cem ent Content 2 %

2.50

2.00

1.5085 & 42 65 & 42 45 & 42

1.00-0 .5 0 -0 .3 0 - 0.10 0.10 0.30 0.50

o c t / c ^ a t m )

Figure 6.17: Normalized Results on Cemented Specimen (C.C. 2 percent)


210

300

200

100

40 60 80Relative Density (percent)

10020

1.0

0.8

0.4

0.2

0.00 32

Cement Content (percent)

Figure 6.18: Influence of Cem entation on a and n


211

in n w ith an increase in relative density a t 0 %, a decrease is observed at 1 and

2 %. W hen th e standard deviation in n in all th e studies presented in Table 6.1 is

considered, it is concluded th a t n m ay best be considered to be constant for one type

of soil and testing equipm ent. Therefore m ean values are used as n values in the

analysis.

6.4 Sum m ary

The two im portan t param eters th a t affect cone test results are cham ber size

and boundary conditions. From comparisons w ith previously reported results, it is

concluded th a t the test results obtained in th is study are not significantly influenced

by th e cham ber size and boundary condition effects. D iam eter ratio of 42 is found to

be sufficient to reduce the cham ber size effects, particu larly a t lower densities (less

th a n 50 %). B oundary conditions have m inor influence when the tests are conducted

a t th is diam eter ratio. C rushability tests indicate th a t crushing under the cone is

insignificant and can be disregarded. The tip and friction resistances reported are

norm alized and the effect of different factors are evaluated.


Chapter 7

ANALYSIS OF TEST RESULTS:

Theoretical and Empirical Methods

7.1 In trodu ction

T he curren t sta te-o f-the-art in the analysis of cone pene tra tion testing depends

largely on em pirical or theoretical correlations developed th rough an evaluation of

theo re tical and experim ental m odels. T he theoretical models of pene tra tion are

based upon the following approaches: bearing capacity theories which assum e rigid

p lastic m ateria l behavior, sim ulation of the penetration m echanism by th e cavity

expansion theory which allows easier incorporation of p lastic ity m odels b u t simplify

th e geom etry of the problem , and the stra in path m ethod which assumes th a t the

s tra in held can be estim ated independent of the stress held. Some of th e above

theories have been successfully used in past studies to in te rp re t cone pene tra tion

te s t results, yet im provem ents are still needed.

In th is chap ter, two bearing capacity theories and two cavity expansion theo

ries are used to p red ict the m easured cone tip resistance and these predictions are

com pared w ith experim ental results. Friction resistance is also evaluated. Based

upon these prediction schemes, a m ethodology is developed to estim ate th e streng th

param eters.

A nother m ethodology based on em pirical correlations is also presented. This

approach is derived based on an assessm ent of the experim ental and also theoretical

tip and sleeve friction values. P redictions from this m ethodology are com pared w ith

experim ental results. The second em pirical m ethod which is based on steady s ta te

line is also presented in another section. T he final section presents a discussion on

various classification charts and their use in the identification of cem ented sands.

212


213

7.2 T h eoretica l M eth od

7.2.1 Tip Resistance - Bearing Capacity Theories

Several investigators have analyzed cone penetration as a bearing capacity prob

lem (M eyerhof, 1961; Durgunoglu and M itchell, 1973; Baligh, 1975; Janbu and Sen-

neset, 1974). These theories assum e different failure m echanism s which are then used

to calcu late u ltim ate bearing capacities using lim it equilibrium approach. W hen

pene tra tion is trea ted w ith th e conventional bearing capacity theories, soil is often

assum ed to behave as a rigid-perfectly plastic m ateria l. Therefore, the elastic strains

and com pressibility of the m ateria l are neglected. Bearing capacity theories which

include com pressibility have also been proposed (Vesic, 1973).

7.2.1.1 D & M Theory

T ip resistance is calculated by determ ining Nc and N^q p resented in E quation

2.5 from the charts provided by Durgunoglu and M itchell (1973) (F igure 2.14) and

the streng th properties reported in Tables 4.2 and 4.3 (C hap ter 4). Peak values of

friction angle and cohesion in tercept are taken in calculating the tip resistance.

Figures 7.1, 7.2 and 7.3 presents th e m easured tip resistances w ith predicted tip

resistances of cem entations 0, 1 and 2 % respectively. The results of a previous study

conducted a t m uch lower confining stresses than this study are also presented (Figure

2.18). T he theoretical predictions correlate excellently w ith m easured resistances.

T he D & M theory assumes rigid-plastic behavior for the m edium . T he stress-strain

behavior of cem ented sands can be considered close to rigid-plastic, particu larly a t

peak s treng th . This m ay be th e reason behind the good correlations.

7.2.1.2 J & S Theory

T he bearing capacity given in equation 2.5 is modified for drained conditions as

follows:

G = A''g-(cr'„ + a) + | - 7 - R - (7. 1)

T he contribu tion of the Aly to the overall resistance is relatively insignificant. T here

fore, th is te rm is neglected to obtain:

Ç t — Ç v — ■ { * ^ v o T q) ~ Û (7.2)


214

50

40

100 k P a 2 0 0 k P a 3 0 0 k P a

© A

55 55 75 % 90 %

10 500 20 30 40Measured Tip Resistance (MPa)

Figure 7.1: Com parison Between Theoretical and M easured Tip Resistances for C.C.

0 % Specimens (D & M Theory)


215

50

cdeu

4 0euüücd

.2 3 0meu

K

eu2 0

(U4-1oTdeuu,

Ph

10

0

- C.C. 1 D . R . 4 2 ; BC3

- c

-H -------- > /

-

B y /

a /A y /

- c B y /

- @ A y AA - 1 0 0 k P a

- B / / B - 2 0 0 k P a- @ /A C - 3 0 0 k P a_ / A- A /

4 5 — 55 %- yfe A A ]Q 6 5 - 7 5 %

A 1 1 1 1 1 1 1 1 1 1 1 1 1

= " " " D r

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

8 4 -

1 1 1 1 1 1

90 %

1 1 1 1 1 t 1 1

] 10 2 0 30 40 5Measured Tip Resistance (MPa)

Figure 7.2: Com parison Between Theoretical and M easured T ip Resistances for C.C.

1 % Specim ens (D & M Theory)


216

50

40

■ 3 0

100 k P a 200 k P a 300 k P a

557590

504010 20 300Measured Tip Resistance (MPa)

Figure 7.3: Com parison Between T heoretical and M easured Tip Resistances for C.C.

2 % Specimens (D & M Theory)


217

T he streng th param eters ‘a ’ and are taken from the Tables 4.1 and 4.2. Ng value

is a function of angle of plastification, j3. /3 values ranging from -5 to -35 degrees

are generally chosen for dilating sands (Janbu and Senneset, 1974). T he value of

th is angle m ainly depends upon relative density and confining pressure. Currently,

selection of the /9 value is m ostly based on experience. In order to form alize th is

selection, the following procedure is adopted. Based on the good predictions by D &:

M theory, it is assum ed th a t J & S theory will also provide good predictions. T he (3

value th a t will give the m easured tip resistance is first determ ined. T he calculated /?

values range from -5 to -25 degrees for low relative densities (45 to 55 %); -15 to -33

for m edium relative densities (between 65 to 75 %) and -20 to -34 for higher relative

densities (above 85 %). T hen these /3 values are correlated to the d ila tion angle, v

(F igure 7.4). T he following best fit line is obtained from the figure.

/3 = 1 .431 / - 3.88 (7.3)

I t should be noted th a t the dilation angle depends on th e relative density and the

confining pressure. F igure 4.23 can be used for estim ating the d ilation angle.

Using th is equation, tip resistances are predicted and com pared w ith m easured

t ip resistances in Figures 7.5, 7.6 and 7.7. There is very good agreem ent betw een

th e m easured and calculated resistances.

Figure 7.4 can be used in determ ining the plastification angle if d ilation angle

is known. D ilation behavior of o ther sands will be different for each sand, hence th is

figure can not be used for sands o ther th an M onterey No. 0/30 sand.

7 .2 .2 T ip R e s is ta n c e - C a v i ty E x p a n s io n T h e o r ie s

Cavity expansion theories have been applied to practical problem s such as in te r

p re ta tion of pressurem eter tests (Hughes e t ah, 1977; Ju ran and M ahm oodzadegan,

1989) and cone penetration tests (Greeuw et al., 1988). Num erical solutions are re

quired when large stra in deform ations are considered in th e analysis. T he analysis

can be simplified by considering sm all s tra in deform ations, and several closed form

solutions were presented for such analysis (C arter e t ah, 1986; Vesic, 1972; Ladanyi,

1963).

T he present work a ttem p ts to study two procedures in cavity expansion theories

and the ir applications w ith regards to sim ulating cone pene tra tion m echanism . The


218

U1•~ 3

C DI—(fcîfiA

c

O•I—I4 - )aÜ

4 - ,• f-4- p

cdr—(O .

40

30

20

0

_ C.C. _ C.C.

C.C.

OOOOO C.C. 0 □ □ □ □ □ C.C. 1 AAAAA C.C. 2

0

0 10 20 3 0 40D ila t ion Angle

Figure 7.4: D ilation Angles Versus Plastification Angles Used in J & S Theory


219

50

40

100 k P a 2 0 0 k P a 3 0 0 k P a

5 5 % 7 5 % 9 0 %

1 0 30 40 500 20Measured Tip Resistance (MPa)

Figure 7.5: Com parison Between T heoretical and M easured Tip Resistances for C.C.

0 % Specimens (J & S Theory)


2 2 0

5 0

4 0

3 0

1 00 k P a 2 0 0 k P a 3 0 0 k P a

55 % 75 % 90 %

503 0 4 010 200


Figure 7.6; Com parison Between Theoretical and M easured T ip Resistances for C .C .



2 2 1

50

CODh

40(UÜ

CO4-)

30COCU

K

P4

T3(D

4-JÜ

" rHT)CUk

CI

20

1 0

0

C.C. 2 D.R. 4 2 ; BC3

- c yB y /

: Ba y /

-y &

- A y-

r c /^ A - 100 k P a

- n J B - 2 0 0 k P a_ A y c - 3 0 0 k P a- 9 y /- y / Dr 4 5 - 5 5 %- / AÆiAAA Dr 65 — 75 %

y ^ EBBBB Dr 8 4 - 90 %

A \ 1 1 1 1 1 1 1 I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11 1 1 1 1 1 1 11 0 2 0 3 0 4 0 5 0


Figure 7.7; Com parison Between T heoretical and M easured T ip R esistances for C.C.



222

first procedure is presented in the closed form solutions given by C arter e t al., 1986.

T his solution is for sm all strains and assumes a constant dilation angle. T he second

procedure investigated by the au tho r is th e m odeling used in cylindrical cavity ex

pansion suggested by Ju ra n and M ahm oodzadegan, 1989. The au thor developed a

spherical cavity expansion solution using th is procedure and updated the form ula

tion for th e effect of cem entation. Predictions using both the constant d ila tion angle

solution (C arte r et ah , 1986) and the newly developed spherical cavity expansion

solution are presented.

7.2.2.1 C avity Expansion Theory - Procedure 1

T he first theory is proposed by C arte r e t ah (1986). Closed form solutions

are presented for the sm all strain expansion of cylindrical and spherical cavities in

an ideal cohesive frictional soil (C arter, e t ah, 1986). These solutions p red ic t the

lim iting pressure necessary in expansion of the cavity. The lim iting pressure, p i is

th e pressure in the cavity which corresponds to a continuous deform ation w ithout

pressure increase. M ohr-Coulom b yield function and an elasto-perfectly plastic soil

behavior are used in deriving the following expression for lim iting pressure (C arter

e t ah , 1986):

2G N - 1Po + c cot (f) N + k

where

^ p z J -_ c c o C ^ y [ p l + cco tiafi + ccot cj) I \ a R + c cot

(7.4)

T = ( l - + i ) ( ^ l + ^ j ( 7 . 5 )

Z = + P-6)

(7.9)

7 = (7.10)

- = M I ("')


223

N = 1 + ^ (7.12)1 — S lT K p

k { l - j x ) - k fx jM + N ) + [(Â: - 2)fz + 1] M N[ { k - l ) f i + l ] M N ^

(7.14)

and k = 1 (for cylindrical cavity) and k — 2 (for spherical cavity); <j> is the friction

angle; i/ is the d ilation angle; fi is the poisson ratio; Po is the in itia l octahedral stress;

G is th e shear m odulus and pi. is the lim iting pressure.

A n assum ption is m ade w ith regards to isotropic pressure, po,

Po = - ( 1 + 2Ko)(Ty. (7.15)

T h e following steps are used in the analysis:

1. M odeling param eters, G, </>, dilation angles are earlier determ ined for various

cem ent contents and relative densities from the drained triaxial tests. These

param eters are obtained by m odeling drained triax ial tests using the Ju ran-

G uerm azi m odel. I t is to be noted th a t the dilation angles are com puted by

tak ing th e average of the increm entally com puted dilation angles over the di

lation region of the volum etric strain - axial strain curve. W hen negligible

dilation angles are calculated (w ithin the zone where volum etric stra in rem ains

constan t), th is averaging is term inated . This procedure is used since the di

lation angle used in the cavity expansion m odel is defined as the slope of the

expansion portion of th e curve.

2. An ite rative program is w ritten in Fortran to estim ate the lim iting pressure.

T he listing of th is program is presented in A ppendix D. M odeling param eters

ob tained above are the input param eters to th is program .

3. Once the lim iting pressure is com puted, the ratio of th is lim iting pressure to

m easured tip resistance is calculated. These ratios are then p lo tted versus

relative density for cement contents 0, 1 and 2 % in Figures 7.8, 7.9 and 7.10.

— = (1.2 — 0.2 C .C .) Spherical Cavity (7.16)

— = / ( D r , C .C .) > 3.5 Cylindrical Cavity (7.17)PL


224

15.0

10.0 -

uA

5.0

C.C. 0 %OOOOO Spherical Cavity □□□□□ Cylindrical Cavity

A - 100 KPa B - 200 KPa C - 300 KPa

□(ÿi

Q Q I I I I I I I I I I I I I I r I I I I I I I I I I I I f I I I I I I I I I I I I I I I I I I I I I I

0 20 40 60 80 100

R ela t iv e D en sity , Dr (%)

Figure 7.8: R atio of T ip R esistance to L im iting Pressure (C.C. 0 %)


225

1-3A

uc r

10.0

8.0 -

6.0 -

4 .0 -

2.0 -

ex:. 1OOOOO Spherical Cavity □□□□□ Cylindrical Cavity

A - 100 KPa B - 200 KPa C - 300 KPa

P

0.00

I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I

20 40 60 80 100

R ela t ive D en sity , Dr { %)

Figure 7.9: R atio of T ip Resistance to Lim iting P ressure (C.C. 1 %)


226

P h

10.0

6.0 -

4.0 -

C.C. 2 %OOOOO Spherical Cavity □□□□□ Cylindrical Cavity

8.0 - A - 100 KPa- B - 200 KPa- C - 300 KPa

BC

□

0.0 ~ l I I I I I I I I I I I I I I I I I I I I I I I I I I I ! I I I I I . . . . . . . . . . . . . . . . . . . . . I I I I I I I I I I

0 20 40 60 80 100

R elative D ensity, Dr (%)

Figure 7.10: R atio of T ip R esistance to Limiting Pressure (C.C. 2 %)


227

T he above ra tio indirectly explains the variation betw een the ac tual and the

idealized expansions under the cone. The idealized expansion can be spherical or

cylindrical in shape. W hen these ratios are closer to 1 , th e idealized expansion can

be assum ed to be close to real expansion. Ratios from both cavity sim ulations clearly

show th a t spherical cavity sim ulations using C arte r e t al. (1986) form ulation yield

lim iting pressures closer to tip resistances.

However, there is a wider variation in the ratios obtained from cylindrical cavity

expansion. A t present, the au tho r does not have any explanations for th is, except

th a t the cavity under the cone is possibly far from being cylindrical in shape. Ratios

from spherical cavity expansion from this study are com pared w ith th e sam e from

ano ther study conducted on uniform fine quartz sand, O osterschelde sand (Greeuw

e t al., 1988) (F igure 7.11). A trend sim ilar to the decrease in ratio w ith increase

in relative density is observed in bo th results. However, the ratio is higher in the

resu lts ob tained in Oosterschelde sand. This m ay due to the size of the cone used in

th e respective studies. A 36 m m diam eter cone is used in the tests on O osterschelde

sand whereas a 12.7 m m diam eter cone is used in the present study. This ra tio is

strongly affected by the d ilation angle. Furtherm ore, in the study by Greeuw et al.

(1988), th e following equation (7.17) is used for determ ining dilation angles instead of

m easuring th is angle using the volum etric stra in d a ta obtained from drained triax ial

tests.

sin <?!> - sinsm r/ = ------— — r— — (7.18)

1 — sm ffi ■ sin (pcv

This equation underpredicts the dilation angle and also it does not account for

th e influence of confining stress. It gives the same dilation angle for all confining

stresses. Hence in th a t study, the calculated lim iting pressure is lower.

T he m ajo r draw back of the analysis by Greeuw, et al., 1988 lies in the model

assum ed for the soil (Ju ran and M ahm oodzadegan, 1989). T he soil (sand) is assum ed

as linearly elastic - perfectly p lastic and the plastic flow is defined by constan t ra te of

d ilation. It is known th a t dense sands undergo contraction and strain hardening prior

to peak principal stress ratio and then followed by a post strain softening behavior

(Ju ran and M ahm oodzadegan, 1989). The dilation is m axim um at the peak and is

m inim um or close to zero a t the critical s ta te or constant volume. T he closed form

solution presented by C arte r e t al. (1986) is derived based on a constan t dilation an-


228

A

0.04DOOOO Ooosterschelde Sand (36 mm die. cone) -□□□□□ Present Study (12.7 mm die. cone)

8.0 -

6.0 -

4.0 --G- -O

g 0 L_i I I I I I I I I I I I I I I I I I I I I I I I I I I I I

30 50 70 90

R ela t ive D ensity , Dr (%)

Figure 7.11: Com parison of Ratios


229

gle a t th e peak and it does not account for the stra in softening behavior. It is known

th a t post peak stra in softening can influence the response of insitu soil significantly,

hence the above solutions appear to be restrictive for proper in terp re ta tions (Ju ran

and M ahm oodzadegan, 1986).

7.2.2.2 C avity Expansion Theory - Procedure 2

T he analytical procedure proposed by th e au tho r is based on the soil m odel

proposed by Ju ran and Beech (1986). The soil is assum ed as homogeneous, isotropic,

and s tra in hardening elasto-plastic m aterial w ith a non-associated flow rule. A m odel

is described in C hap ter 4, and the m odeling param eters are presented in Tables 4.4

an d 4.5. T h is m odel uses octahedral stress variables (q and p) where as th e present

soil m odel is based on deviatoric stress {i = ^ - '-^ ) and average effective stress

(s = However, soil m odeling param eters do not change when the m odeling

p aram eters , q — p or t — s' are used. For convenience, the yield function, hardening

function and plastic po ten tia l functions are hereby presented in term s of t and s .

A M ohr-Coulom b type yield criterion is considered:

= (7-19)

w here a{j is th e stress tensor. T he ^ (7 ) is the strain hardening function rela ting the

ac tual yield surface to th e curren t strain ra te . The strain hardening function used

in the above equation m ust be specifically defined for the case of contracting and

d ila ting m ateria l. For dense d ilating sands, it is assum ed th a t the hardening function

^ ( 7 ) is parabolic to hyperbolic and can be w ritten as:

(7.20,

where

. = (7.21)G sin (pcv

6 == sinifplT (7.22)G

c = sin (j)cv (7.23)

and r = 1 + [1 — (sin<?ict; sinç^p)]^.

T he difference betw een the above a, b and c values and those of octahedral stress

m odeling should be noted.


230

T he following flow rule which defines stress ratio-dilatancy relationship is derived

based on energy considerations (Ju ran and Beech, 1986):

7; = sin 1/ = 3 - ^ - — s in é c v 7 (7.24)d y //j s

where

de^ = p lastic volum etric s tra in increm ent

= p lastic deviatoric s tra in increm ent

T] — dilation ra te

1/ = dilation angle

and lis is a correction m odulus defined as

— /ii when 4 - < sinçi>ci;; contracting behaviors

fis = /f2 w h en — > sin ç!>cü; dilating behavior

Using th is soil m odel, the following procedure is adopted to ob tain a spherical

cavity solution. Spherical sym m etry, radial equilibrium and com patib ility conditions

are used in th e analysis.

T he strains in the expansion are given by:

dz z z . ^Ér — —3—; Eg — —; = — (7.Z.5)dr r r

where Cr, Cg and = radial, circum ferential and spherical strains; x = the radial

displacem ent on the face of the cavity; r = radius of the cavity.

N is defined as the ra tio of volum etric stra in to the deviatoric strain:

iV = slnV. = - = i ^ i ^ ^ (7.26)7 €r - e e

From the above expression, the following relationship is obtained:

Cr 2 + sin Ip(7.27)

eg 1 — sin Ip

if e lastic stra ins are neglected. E quation 7.27 can be expressed in the following form:

d-y = - 3 " (7.28)1 — sin 77


231

By differentiating E quation 7.26 w ith respect to 7 , the following relation between

th e dilation angle, i/, and N is obtained:

d Nsin [/ = TV + 7 (7.29)

(7 7

T h e s tra in com patib ility condition is given by:

Cr — ee + r — (7.30)0 7

T h e rad ial equilibrium equation is given by:

q. 2 = 0 (7.31)or p

Com bining th e above equations 7.28, 7.29 and 7.30 along w ith th e radial equilibrium

equation 7.31, the following equation for increm ental lateral confining pressure is

ob tained:

If th e sand exhib its cohesion in tercep t, the following addition has to be m ade to the

increm ental pressure:

Actcc = ------- -— -ccos(f) (7.33)1 + eg eg

T h e following increm ental procedure is used to com pute the lim iting pressure.

1 . A sm all displacem ent, x is assum ed. S trains, dilation angle, v, and the devia

toric s tra in , d'y, are determ ined.

2. Using the dy, the s tra in hardening function is calculated.

3. T he nex t step is estim ating the increm ent in lateral confining pressure by using

equation 7.31. This is added to the in itial lateral confining pressure to obtain

the curren t lateral pressure.

4. T he above steps are repeated until th e increm ental lateral pressure com puted

is relatively insignificant. T he late ra l pressure a t the end of th is final step is

taken as the lim iting pressure.


232

A Fortran program is w ritten for the above m odel (A ppendix D) and lim iting

pressures are estim ated for different densities and cem ent contents. The lim iting

pressures are com puted and tab u la ted along w ith tip resistance in Table 7.1. T he

ratios of tip resistance to lim iting pressures are p lo tted versus vertical confining

pressure and these figures are presented in A ppendix D.

This procedure estim ates lower lim iting pressures th a n th e procedure given by

C arte r et ah, 1986. T he first procedure assum ed a constan t dilation angle. T he

valid ity of th is assum ption is dubious since the dilation angle a t higher stra ins is

zero. As a resu lt, th e procedure proposed by C arte r e t ah, 1986 clearly overestim ates

th e lim iting pressure.

T he ra tio of tip resistance to lim iting pressure range from 2.2 to 3.5 a t densities

of 45 %, 5.5 to 7.5 a t densities of 65 % and around 8 for densities a t 85 %. This ra tio

increases w ith increase in density. A ratio called the ‘form fac to r’ is given betw een

tip resistance and lim iting pressure by Vesic (1977).

9c = N f ■ PL (7-34)

Qc - tan^ f j • ( 1 4- sin (f)cv) ■ exp 7T — <Pcv) tan d> (7.35)

T h e actual shape under the cone is different from the idealized spherical shape. T he

N f which takes care of the shape difference as a function of friction angle a t constan t

volume. T he factors are calculated for the present test resu lts and are included in

th e sam e table. T he fact th a t th e sam e form factor proposed by Vesic (1977) is of the

sim ilar order w ith the ^ values suggests th a t the assum ption on the cavity shape

m ay be the reason behind high ^ ratios.

Cavity expansion theories and th e bearing capacity theories seem to provide

good predictions of the m easured tip resistance.

7.2.3 Friction Resistance

The theoretical prediction of friction resistance assum es th a t sleeve friction is

only due to shear resistance. T he following form ula can be used to calculate the

friction resistance:

/. = 5. ■ K. + «) (7.36)

in which


233

Table 7.1: Com parison of Cavity Expansion Theories

D r 45 % D r 65 % Dr 85 %O-v 1 0 0 2 0 0 300 1 0 0 2 0 0 300 1 0 0 2 0 0 300

( ^ )c . c o %

PLQc

2 . 2

6 . 0

2.97.5

3.511.5

2.48.4

3.117.6

3.720.3

2 . 8

1 2 . 2

3.42 L 8

4.135.0

PiC arte r (1986)

2.71.46

2 . 6

1 . 2 1

3.31.32

3.51.07

5.71.30

5.5L25

4.31.08

6.41.14

8.51.43

N f .0 - 4. 3 L.0 - 5.0 5.0 - 6.5C.C 1 %

PL<lc

2 . 2

6 . 8

2 . 8

9.73.510.4

2.413.1

3.119.1

3.72 4 ^

2.71 & 8

3.22&4

3.93 1 8

3s.PL

C arte r (1986)3.11.06

3.51 . 0 2

3.0Oj&

5.51.05

6 . 2

1 . 0 0

6 . 6

1.046 . 6

0.817.9

0 . 8 6

8 . 6

0.90N f c .0 - 4.0 L.0 - 5.0 5.0 - 6.5

C.C 2 %PL9c

2.58 . 0

3.21 2 . 0

3.915.0

2 . 6

14.03.4

2 & 0

4.130.7

2.92 & 2

3.72&5

4.435.0

3s.PL

C arter (1986)3.2

0.953.8

1.043.8

1 . 0 1

5.41 . 0 0

6.50.89

7.51 . 0 1

7.0&83

8 . 0

&878 . 0

0.74

N f 3 .0 - 4. 3 1 . 0 - 5 . 3 5.0 - 6.5Units: - kPa; and - M Pa


234

fs = shear or friction resistance

Ss = |r | • tan<l) • KI I < o n 5

I' I — tan0

T he ea rth pressure coefficient K is generally taken as the /{o in the pile friction

capacity calculations below a critical dep th (Poulos, 1976). F igure 7.12 shows the

ra tio s of K /K o values for various densities a t each cement content. These figures

depict th a t K j Ko values are closer to 1 for lower densities (around 45 %) and increase

w ith higher densities. In soils w ith high dilation characteristics such as cem ented

sands, and a t higher densities, using Kg in the equation will p red ict lower friction

resistance. T he restra ined d ilatancy will increase the confinement on the interface

resu lting in higher K values.

F igure 7.13 presents the influence of dilation angle on K j Ko- Since d ilation angle

depends on the rela tive density and confining stress, sim ilar observations are noted.

R esults from Baldi (1981) are also p lo tted in this figure. These results were ob tained

on a norm ally consolidated Ticino sand. This variation between both results is due

to th e difference betw een dilation characteristics of the two sands.

7.2.4 Approach for Estim ating Cohesion and Friction Angles

A chart for estim ating relative densities and cohesion was proposed by P u p p a la

et al., 1993. This chart was based on tests conducted by R ad and Tum ay (1986)

using cem ented sands. These results were obtained in a rigid PV C cham ber and

a t very low confining stresses (less th an b k P a ) . Furtherm ore, the influence of rigid

boundary conditions, cham ber size effects and high stresses were not evaluated. T he

present study gives a way to incorporate the effect of all these factors.

P resen t results were conducted in a flexible double-walled calibration cham bers

and were reported for th ree confining stresses, relative densities and cement contents.

Different bearing capacity theories were used to predict these m easured values. These

theories pred icted tip resistance reasonably well. Hence, the prediction of the bearing

capacity theories are used in the following semi-empirical approach developed.

Cone penetration tes t results obtained in calibration cham bers dem onstrate th a t

tip resistance increases nonlinearly w ith the increase in vertical effective stress bo th


235

O%

8

oooœc.c. 0□□□□□ C.C. 1 AAAAA C.C. 2

C.C.

6

4

2

Ko L in e

0

40 60 80 1 0 0

Relative Density

Figure 7.12; K fK o Versus Relative Density


236

O

8

OOOOO C .C . 0 %- - □ □ □ □ □ C .C . 1 % A A A A A C .C . 2 %

B a l d i ( 1 9 8 1 )6

4

2,AQ

0

0 1 0 20 3 0 4 0D ila t io n a l Angle

Figure 7.13: Influence of D ilation Angle on K j K o


237

due to changes in dilational characteristics and peak friction angle w ith the increase

in confinem ent and also due to grain crushing a t higher confinem ent. Therefore, the

em pirical and sem i-em pirical correlations of tip resistance employ a non dim ensional

tip resistance ( - ^ ) /( where (Jatm is atm ospheric pressure and n varies betw een

0.55 to 0.84. Sim ilar norm alization was found necessary in relating cone penetration

resistance to low stra in dynam ic properties by o ther investigators (Rix and Stokoe,

1991). This implies a value is a function of the in itial s ta te of the soil.

T ip resistances pred icted by D &: M theory showed good agreem ent w ith m ea

sured values up to a confining stress of 350 k P a . Hence, non-linearity m ay be ne

glected a t vertical stress less th an 350 k P a and for relative densities less th an 80 %

(Villet and M itchell, 1981). Sim ilar findings were observed in o ther studies where

bearing capacity theories were used (Baldi, e t al., 1981). Possible reason for th is

linear relationship betw een Qc and was due to th e non-linearity of the s treng th

envelop (B aldi, et al., 1991); i.e. <p' value decreases as confinement increases. B ear

ing capacity theory predictions of th e tip resistances showed quite good agreem ent

in th is study. This implies, a linear relationship can be assum ed between the tip

resistance and vertical effective stress when bearing capacity form ulations are used.

Hence, norm alized tip resistance, ^ elim inates the influence of the vertical effective

stress.

The following m ethodology is used in preparing the semi-em pirical approach.

Cem ent content and relative density are expressed in term s of cohesion and friction

angle. T ip, friction resistances and friction ratios are calculated for different cohesion

and friction angles. Using these values, a chart is prepared.

Figure 7.14 com pares the norm alized tip resistance, ^ w ith respect to friction

ra tio for bo th bearing capacity theories. These theories provide sim ilar predictions

of norm alized cohesion in tercept, from a knowledge of friction ratio and tip resis

tance.

Figure 7.15 and 7.16 provide charts which can be used to estim ate the norm alized

cohesion in tercep t and the relative density when the friction ra tio and norm alized

tip resistance are known. T he first chart is prepared for ranging from 0 to 1 .

T he second figure is for values of 0 to 5. P resent testing d a ta is depicted in this

figure for com parisons. This d a ta shows th a t reasonable estim ates of ^ and Dr

are possible by th is chart. Once th e range of relative density is known, the friction


238

10000

ucr

LUOZ<HcnenLUorCL

PûLUN

<

a:oz

1000

100

D aw j a s

85

7060

45 30

o

1 0 I I ' ' I I I I I I I ' ' ' I ' ' I I ' I I I I ' ' ' I I I 1 ' I Ll ' I0 . 2 0 .4 0 .6 0 .8

FRI CTION RATIO (%)

1 I I1.0

Figure 7.14; The Norm alized T ip Resistance Versus Friction R atio for Various ^

Values


239

angles can be estim ated by using F igure 7.17. A lthough the above procedures provide

good estim ates of the streng th param eters, insitu d a ta is still necessary to verify and

im prove th e prediction scheme.

7.3 Em pirical M eth od

Two em pirical m ethods are presented in th is section. T he first m ethod is based

on the octahedral stress norm alization which is described in C hap ter 6 . T his ap

proach is la te r modified such th a t it can be used in th e in te rp re ta tion of insitu tests.

T he second m ethod is based on the s ta te param eter concept. This m ethod requires

a classification chart for identifying cem ented deposits.

7.3.1 Empirical M ethod Based on a Approach

E xisting relative density charts are developed from cham ber tes t results on a

p a rticu la r type of sand. These charts are generally valid for th a t particu la r type of

sand. An a tte m p t is m ade here to norm alize various cone results and use them in

form ulating a new approach for estim ating the relative density. This m ethod is also

ex tended to estim ate cem entation levels.

7.3.1.1 T ip Resistance

Figure 6.9 (C hap ter 6 ) which depicts the d iam eter ratio influence on a for various

densities can be used for estim ating relative density. The a values for different

relative densities a t a certain d iam eter ratio are taken and are rep lo tted (F igure

7.18). The best fit regression lines equations for each d iam eter ra tio are presented

as follows;

Poet(âtm

= (3.6Dr - 54.0) ( , D.R . = 60V âtm j

(7.37)

= (3 .2D , - 43.0) f , D .R . = 50 (7.38)V âtm jatm

- = (3.0D , - 39.0) f , D .R . = 42 (7.39)\âtm J

W hile the above equations generalize th e change in tip resistance w ith confinem ent,

n values are considered specific to th e study. For uncem ented, clean, norm ally con-


240

uO ’

ÜJOZ<h-C/îc75ÜJorCL

PoÜJN

<S(TOZ

0000

8 5

7 0

6 0

1000

4 53 0

100

SYMBOL (C /o ^ O ^ Dp (% )

^ 0 . 4o 0 . 2 5 6 8 - 7 5o 0 . 1 5 4 5

> 8 0

0.2 0 .4 0.6 0.8

FRI CTION RATIO (%)

Figure 7.15: A C hart for E stim ating Cohesion Intercept and Relative Density


241

1 0 0 0

ucr

LUOz:<Hcn côLUorCL

QLUN

croz

100 -

8 0

6 0

4 5

3 0

0 . 5

0 . 3 8 6

0.20.1

6 9

4 - 8

1 0 0.2

-1 I I I I I I I I 1 I I I I I I I I I I I I I I I I I I I 1 I I I I I I I I I0 .4 0 .6 0.8

FRICTION RATIO (%)

1.0

Figure 7.16: A C hart for Estim ating Cohesion In tercept and Relative Density


242

4 0 .0

P 3 8 .0

< 3 4 .0

3 2 .0 m

OOOOO Peak □□□□□ Residue

3 0 .01 0 06 0 8 00 20 4 0

R elative D ensity , D,

Figure 7.17: Relative Density Versus Friction Angles


243

400

■■BHa Dr 25 %- □□□□□ Dr 40 % ■ AAAAA Dr 70 %- OBBas Dr 80 %300

10 0

5 45 6525DIAMETER RATIO

(S

<

300

D.R. 60

D.R. 42

200

10 0

02 0 40 60 80

Relative Density (percent)1 0 0

Figure 7.18: a Versus Relative Density for Various D iam eter R atios


244

Table 7.2: The n for Various C em entations

Partic le Shape Range SuggestedSub angular to angular

Sub angular Sub rounded to sub angular

0.550.76-0.780.80-0.92

solidated sands, the au tho r finds th a t n values obtained are affected by the particle

shape (Table 7.2).

T he above equations, 7.37, 7.38 and 7.39 are used to produce th e following

equation which depicts th e influence of cham ber d iam eter ratio . This equation is

valid for cham ber diam eter ra tio values up to 60.

qc Poet ^ + 1.75)Dr - (0.84 D .R . + 2.6)] ( (7.40)O' atm \ O a tm ,

All th e above equations are derived bcused on the results for clean, uncem ented sands.

T he effect of cem entation which is valid up to a cham ber d iam eter ra tio of 42 is

included below.

T he a values of 1 and 2 % cem ent contents are com paratively p lo tted along w ith

the uncem ented results (Figure 7.19). The following equations are ob tained for the

cem ent contents a t 1 and 2 %.

Çc — Poet

âtm

Qc Poet

^ atm

(3.08 Dr 1 %

(3.64 D , - 4 7 . 4 ) 1 ' ^ ^ , C .C .\ Oatm /

2 %

(7.41)

(7.42)

Combing equations 7.39, 7.41 and 7.42 yield

qc — Poet

âtm

where n value is:

= [(2.9 + 0.3 C .C .) Dr + (4.0 C .C . + 38.33)]

n = 0.84 - 0.9 In ( l + 0.56(C.C.)°'^^) (7.44)

This equation includes the effect of both cem entation and relative density. T he

only disadvantage in th is approach is the use of octahedral stress variation. In

field tests, it is difficult to estim ate th e lateral stresses, thereby octahedral stresses.

An a tte m p t is m ade to m ake th is approach applicable to field tests by replacing


245

A

3 0 0

C.C.

200

1 0 0

020 40 60 80 1 0 0

Relative Density (percent)

Figure 7.19: a Versus R elative Density for Various Cem ent C ontents


246

Table 7.3: T he and for Various Investigations

C.C Dr log ay fly ^mean(Std. Dev.)

Oiy

0 % 456585

1.721.922.05

0.981 . 0 0

0.980.99

(0 .0 1 )

5383

1 1 2

1 % 456585

1.801.992 jW

0.850.890.92

& 8 8

(0.03)

6397158

2 % 456585

I j #2.092.31

0.710.750.83

0.76(0.06)

73123204

octahed ra l stresses w ith vertical stresses. Vertical stress norm alization is perform ed

on p resen t results and the variables in these are defined as ay and n„. The ay and

Uy values are estim ated (Figure 7.20) and are presented in Table 7.3. T ip and sleeve

friction values used in th is norm alization are taken from experim ental investigations.

Some of the values which differ significantly from theoretically com puted values are

rep laced w ith theoretically com puted values. This is done to improve the em pirical

correlations.

T h e following em pirical equation is obtained for the relationship between nor

m alized tip resistance and vertical effective stress.

= [(1.71 + 0.99 C .C .)D , - (41.9 C .C . + 34.3)] 'âtm ^ a tm

where

(7.45)

Uy = 0.99 - 0 .1 (C .C .)

This linear equation is valid up to 2 % cem entation level.

(7.46)

7.3.1.2 Friction Resistance

T h e friction resistance in te s t results can be expressed in the following form .

\ "1/ .âtm , âtm J

(7.47)


247

Ib\b1c?

'bfio

Ib

\bIcrtjflo

3.00

T h is S tu d y (C.C. 0 ss)

2.50

2.00

1.50

C a X O 4 5 & 4 2 I ' l l I I I 6 5 & 4 2 a a a a a 8 0 & 4 2

1.00- 0.00 0.20 0.40 0.60- 0.20

log(^v'/cratin)

3.00

T h is S tu d y (C.C. 2 ss

2.50

2.00

1.50

cmx) 4 5 & 4 2 L 1 . 1 I I I 6 5 & 4 2 A A A A A 8 0 & 4 2

- 0.00 0.20 0.40 0.60

1 1 ) g ( O ' V / O ’ g )

Figure 7.20: Norm alized P lots of D a ta to D eterm ine and riy (P resen t Test Results:

C.C. 0 and 2 %)


248

1.00

Cernent C o n ten t 0 %

0 .5 0

0.00

Dr (%)

OOOOO 85- 0 . 5 0

AAAAA 45

- 1.00- 0 . 3 0 0 .700 .5 0

Figure 7.21: /? and n i from Present Test R esults


249

.00

C em ent C on ten t I %

0.50

Sb 0.00

tsflo

Dr (ss) 00000 85

-0 .5 0

’ ■ -0.30 - 0.10 0.10 0.30 0.50 0.70

log(c^vAatm)1.00

C em ent C on ten t & ss

0.50

S0.00b

bfiO

Dr (ss)00000 85 0000 145 a a a a a 65

-0 .5 0

- 1 .00-0 .3 0 - 0 .10 0.10 0.30 0.50 0.70

Figure 7.22: ^ and n-i from Present Test R esults


250

Table 7.4: T he /? and n \ for Various C em entations

C.C. Dr ni ^l,m0 45

6585

0.230.400.42

1.141.091.25

1.16(0.08)

0.220.390.45

1 456585

0.600.740.97

0.460.440.63

0.51(0.10)

0.580.870.91

2 456585

0.410.830.91

0.850.460.60

0.64(0.19)

0.500.741.07

In order to determ ine ^ and n i , friction resistance results of th is study are norm alized

and p lo tted (F igure 7.21 and 7.22). T he /3 and ri\ are estim ated in each figure and

tab u la te d (Table 7.4).

A fter perform ing sim ilar analysis as in the case of tip resistance, friction resis

tan ce can be expressed in th e following form:

f s = [(0.002 C.C. -t- 0.006) Dr + (0.27 C.C. - 0.02)]O ' . .

"J(7.48)

âim \^ a tm ,

w here n \ = 1.16 — 1.37In (1 -f 0.56(C.C.)°-^^).

T ip and friction resistances are defined empirically in the equations 7.45 and

7.48. These equations are used in the following two procedures proposed in deter

m ining geotechnical characteristics of deposits from cone penetration tests. These

procedures are presented in the following subsections as A pproach 1 and 2. These

em pirical procedures are only proposals and they have to be validated in the field.

7.3.1.3 Approach 1

T his approach can be used provided unconfined compression streng th , qj is

known. Hence, th is approach requires block sam pling of the specim ens from the

tes tin g location or some o ther m ethod of estim ating the unconfined compression

s tren g th or th e cohesion in tercept of the deposit.

T he expression for cem ent content as a function of qj and Dr is necessary in

th is procedure. T he unconfined compression tes t results conducted on cem ented

specim ens of 1 and 2 % a t a curing period of 7 days are considered, qj is correlated


251

w ith rela tive density for bo th cem ent contents using th e following equations (F igure

7.23).

qj = 0.44 D r + 5.6 C .C . = 1% (7.49)

qj = 0.33 D r + 19.-5 C .C . = 2% (7.50)

These equations are combined to produce the following equation for qj which is a

function of relative density and cem ent content,

qj = (0.165 C.C . + 0.09) Dr -f (9.7 C .C . - 1.37). (7.51)

From th is, cem ent content expression is derived,

(gy + 1 .3 7 -0 .0 9 D r ) 'C .C .= (7.52)

(0.165 D r+ 9.7)

T he above equation 7.52 along w ith equation 7.45 are used in preparing the

following charts for estim ation of relative densities. C em ented deposits are defined

according to their unconfined compression streng th (R ad, 1984). Low cem ented

deposits are defined to have a qj of 100 kP a . However, cem ent contents of 1 and

2 % which are used to sim ulate these low cem ented deposits exhib ited unconfined

com pression strengths of up to 50 k P a (R ad, 1984). T he above equation is used along

w ith tip resistance expression in preparing charts for various qj values. These charts

(Figures 7.24, 7.25 and 7.26) are used for estim ating relative densities for qj values

of 0, 20 and 40 kP a respectively. Rem aining plots for o ther unconfined compression

streng ths are presented in A ppendix D.

T his m ethod can not be extended to higher unconfined compression streng ths

unless fu rther cem ented specim ens (higher C .C .) are tested . A qj value of 5 0 k P a

represents a cem ent content of above 3 % and th e equations derived for n , qc and

C.C . will not be valid a t these cem ent contents. F urther research is necessary to

estab lish the relationships a t higher cem ent contents.

7.3.1.4 Approach 2

In the event the block sam pling is not available. Approach 2 is proposed. B oth

t ip and friction resistance expressions along w ith cem ent content equation 7.52 are

used in preparing the Figures 7.27, 7.28 and 7.29 to estim ate bo th qj and Dr- Each


252

8 0

OOOOO C.C. A A A A A C.C.

6 0

4 0z qd

d

20

1 0 020 4 0 6 0 8 00Relative D en s ity (%)

Figure 7.23: T he q/ Versus Relative Density for C.C . 1 and 2 %


253

ob

- > b

( Qc CTy ) / ( CTqI ppi )

2 0 0 4 0 0T—I—I—I—I—I—I—I—I—I—I—I—I—I I I I—I—r

-

600

SYMBOL Dr(%) D.R.A 4 9 4 2B 6 9 4 2C 8 9 4 2D 2 9 2 0E 6 3 2 0F 2 5 4 2G 61 4 2H 61 3 4

qf = 0

A, B, C - Present StudyD. E - vniet a Mitchell ( 1981 )F, G . H- Eld ( 1987)

Figure 7.24: C hart for E stim ating R elative Density {qj — 0 kPa) (A pproach 1)


254

0( Qc ( ^ a t m )

2 0 0 4 0 0 6 0 0

ab

* > b

S Y M B O L Dr(7c A 4 9B 4 7C 5 3

2020

= 20 kPo

2

3

Figure 7.25: C hart for E stim ating Relative Density {qj = 20 kPa) (A pproach 1)


255

0

( Qc CTy ^ t m )

2 0 0 4 0 0 6 0 0

>b

SYMBOL Dr(%) A 8 6B 6 9C 7 2

^f,m4 24 04 0

2

= 4 0 KPa0/

3

Figure 7.26: C hart for E stim ating Relative Density {qj = 40 kPa) (A pproach 1)


256

5 0 0

4 0 0

E

3 0 0

' 200 ü

CT4 0 " ° 8 0

2 5 6 0

5 0100

30 3 0 8 6

0 3

Figure 7.27: C hart for E stim ating Dr and qj for = 1. (A pproach 2)


5 0 0

4 0 0

3 0 0

bIo

cr

200

100

257

Dr C.G.

X 0 9 0 0Y 3 0 8 5 1Z 4 5 8 6 2

01 I I I I 1 I 1 I 1 1 I 1 I I I 1 I I I 1 I 1 I r I 1 1 1

2



258

5 0 0

4 0 0 -

b 3 0 0 \

b' 200 o cr

100 -

Dr C.C.

X 0 8 8 0Y 3 0 8 9 1Z 4 5 8 4 2

I ' 1 I I I I I I i I I ' I 1 I 1 I I t 1 I I ' -1_1_I ' I0 1 2 3

^ ^ a t m



259

figure is valid for a certain vertical effective stress. For stresses o ther th an those

given, charts are to be in terpolated .

T he above approaches not only identify th e cem ented deposits b u t also give the

s treng th properties. However, a t present these m ethodologies are applicable only

for subrounded to subangular m aterials like M onterey sands. G eneralization of th is

approach could not be done due to the lack of cem ented cone tes t results on o ther

sands. Furtherm ore, the validity of bo th approaches are to be investigated for field

te s t results.

7.3.2 Empirical Method Based on State Parameter Approach

In th is approach, th e first step is to identify w hether the sand deposits a re ce

m ented or uncem ented. Hence, a classification chart is developed by p lo tting friction

ra tio on the x-axis and norm alized tip resistance on the y-axis (F igure 7.30). T he

friction ra tio is calculated by tak ing the ra tio of friction resistance to tip re

sistance This is expressed in %. T here is a m arked zone in the figure which

represents th e possible cem ented deposits. Any cone results of the sand th a t lie in

th is zone can be expected to have cem entation.

A nother em pirical approach generally used in the estim ation of relative densities

is based on th e concept of s ta te param eter. T he definitions of the s ta te param eters

are presented earlier in C hapter 4. S ta te param eters are calculated by sub trac ting

th e void ra tio a t the steady s ta te from the natu ra l void ratio . These are calculated

for bo th cem ented and uncem ented test results. Void ratio a t steady s ta te is sam e

for all confining pressures up to 300 k P a used in the triax ia l test. This is because

th e SSL a t these confining pressures is still a line parallel to the x-axis. T here is no

variation betw een s ta te param eters of sands a t different cem ent contents since critical

s ta te lines a t different cem entation levels lie close to each o ther (F igure 4.11). Table

7.5 presents the steady s ta te param eters.

Figure 7.31 (uncem ented) and 7.32 (cem ented) present qc-cr on x-axis and a' on

y-axis. Each curve in the figure represents a particu lar s ta te param eter, t}’. Vertical

stresses axe used in th is approach since it is difficult to es tim ate the lateral stresses

in the field. Once s ta te param eter is known, streng th param eters can be ob tained

from the ip-<p and (/^-cohesion correlations (Figure 7.33). Table 7.6 also presents the


260

10000

LUOZ :<k-^ 1000 tf)LUa:

:c.c.%=2

û_I -

QLUN

100_ J<

oroz

320

F R I C T I O N R A T I O ( % )

Figure 7.30: Classification C hart for E stim ating Cem ented Deposits


261

Table 7.5: T he Steady S ta te P aram eters

D r e ^ s s l Ip

90 100 0.59 0.77 -0.1890 200 0.59 0.77 -0.1888 300 0.59 0.76 -0.1772 100 0.63 0.77 -0.1469 200 0.64 0.77 -0.1371 300 0.63 0.76 -0.1349 100 0.69 0.77 -0.0856 200 0.67 0.77 -0.1055 300 0.67 0.76 -0.10

Table 7.6: T he S treng th P aram eters

C.C. D r ^peak ^Tes Cpeak Cres0 -0.18 85 39.0 35.0 0 0

-0.14 65 36.5 315 0 0-0.09 45 35.0 34.0 0 0

1 -0.18 85 38.0 36.0 14.0 0.0-0.14 65 36.5 3&5 11.5 0.0-0.09 45 35.0 35.0 9.0 0.0

2 -0.18 85 39.0 3R0 30.0 19.0-0.14 65 37.5 3&5 25.0 15.5-0.09 45 36.0 35.0 20.0 12.0

Note: R esults are ex trapo la ted for the above relative densities from Rad (1984)

stren g th param eters for various “ip values of M onterey No. 0 sand. Since figures 7.31

and 7.32 are presented for various cem ent contents, th e first s tep is to identify the

am ount of cem entation. F igure 7.30 can be used for th a t purpose.

Insitu d a ta are still necessary to verify and improve th is scheme. Above ap

proaches are em pirical and b e tte r correlations can be obtained by increasing the

num ber of tes t results and including field correlations.

Selection of the m ethod (em pirical or semi-em pirical) will no t affect th e in ter

p re ta tions since both m ethods provide sim ilar predictions. However, the em pirical

scheme can be im proved by including m ore cem ented specim en results.


»262

qc-o-v (kg/cm ^)

C\J

O

bfi

>b

400300200000

U n c e m e n te d S an d s

-0 .1 7 5-0 .1 3 5-0 .0 9 0

2

3

Figure 7.31: C hart for E stim ating ^ for Uncem ented Sand


263

qc-cr, ( k g /c m ^ )

aobO

40 0300200100

C em ented (C.C. 2 % )

Ip - —0.175 = —0.135

Ip = -0 .0 9 0

qc-o-v (k g /c m ^ )

bg

100 200 300 400

C em ented (C.C. 1 %)

-0 .1 7 5-0 .1 3 5-0 .0 9 0

Figure 7.32: C hart for E stim ating i/» for Cem ented Sands


264

40

QQQQDC.C. 0 SB □DQOD C.C. 1 SB a a a a a C.C. 2 SB38

34

32

30 C-t 0.05 0.15 0.200.10

State Parameter

50

40QQQQOC.C. 0 SB □□□□□ C.C. 1 SB AAAAA C.C. 2 SB

OSCL,

ÜO'm0)-0ou

30

20 -

10

> t I 1 I 1 I ^ t t I I I I-1- I I /r, I t0.05

‘ O ' I ' ' I I I e0.10 0.15State Parameter

0.20

Figure 7.33: V’ Versus Cohesion and Friction Angles


265

7.4 In fluence o f V ertical C onfining Pressure on C em en ta

tio n

T ip resistance increases w ith an increase in cem entation. This increase can

be observed significantly a t shallow dep ths (small overburden pressures). However,

a fter certa in dep th , th e influence of cem entation on tip resistance can be disregarded

since its con tribu tion to tip resistance is less than 4 %. To investigate th is aspect, the

bearing capacity results a t various depths for different cem entations are considered.

F igures 7.34 (cem entation 1 %) and 7.35 (cem entation 2 %) are p lo tted w ith the

ra tio of th e difference betw een th e cem ented tip resistance and th e uncem ented tip

resistance as a percentage of th e uncem ented tip resistance, _ ] ) on x - axis Qc,uncem '

and vertical confining pressure on y - axis. This ra tio decreases w ith the increase

in confining pressure. T he ra tio is closer to 0.04 a t confining pressures of 600 kP a ,

suggesting th a t the cem entation contribution is insignificant. This implies th a t for

confining pressures of m agnitude 600 hP a (equivalent to 60 m of dry soil or 120 m of

sa tu ra ted soil), sem i-em pirical or em pirical charts for uncem ented sands can be used

for in terp re ta tions.

7.5 D iscussion

An a tte m p t is m ade in th is section to com pare th e results of this study w ith the

various classification charts proposed by different investigators. Classification charts

by Schm ertm ann (1978), Douglas and Olsen (1981), Tum ay (1985) and R obertson

and C am panella (1985) are used. Figures 7.36, 7.37, 7.38 and 7.39 present these

charts. Schm ertm ann (1678) charts is in agreem ent w ith present results. C hart

by Tum ay (1985) classify dense cem ented results close to the present test results,

however it classifies the m edium to loose cem ented sands as loose sands. Similar

observations are noted in C am panella’s chart. C em ented sand results overlap the

zones 1, 6, 7, 8 and 9 in th is chart (R obertson and Cam panella, 1985). T he in teresting

aspect of th is chart is th a t th e cem ented sands lie in a zone of possible liquefiable

soils which signifies the im portance of identifying cem ented deposits.

F igure 7.37 shows a relatively m ore com prehensive chart developed by Douglas

and Olsen (1981). This chart not only uses the norm alized param eters bu t also


266

Oo

e(Uuü3_o

CT*

S(Uüo

cr*

100

Cernent Content 1 %

QQQQO Dr 5 0 %□□□□□ Dr 70 55 A A A A A Dr 85 56

10

0 200 4 0 0 6 0 0 8 0 0

Effective Vertical Stress (kPa)

Figure 7.34: Influence of Cem entation (1 %) and Confining P ressure on T ip Resis

tance


267

1000

Cernent Con ten t 2

OO

QQQQO Dr 50 %□□□QP Dr 70 % A A A A A Dr 85 %

B0)üa3CJ

cr*

B«uu

CJ*

8 0 00 200 4 0 0 6 0 0Effective Vertical Stress (kPa)

Figure 7.35: Influence of C em entation (2 %) and Confining Pressure on T ip Resis

tance


268

recognizes the change in n value w ith the type of soil. C em entation is not included

in th is chart. T he au tho r upda ted th a t region of the chart as given in Figure 7.37.

I t was earlier discussed th a t Dr in te rp re ta tion charts can not be valid for ce

m ented sands since cem entation has pronounced effect on cone resistance. In ad

d ition , charts are not developed for determ ining unconfined com pression s treng th

and cohesion in tercept of cem ented sands. This research has accom plished th a t goal

by providing em pirical and theoretical evaluation m ethods. These m ethods can be

validated and also u pda ted by conducting field tests on cem ented deposits.

7.6 Sum m ary

T heoretical schemes such as bearing capacity theories and cavity expansion th e

ory are used to predict tip resistance. Bearing capacity theories proposed by Dur-

gunoglu and M itchell (1973) and Janbu and Senneset (1974) provided reasonable

estim ates of tip resistances. Two procedures are used for sim ulating cavity expan

sion. T he first procedure of cavity expansion theory predicts lim iting pressure which

can be correlated w ith m easured tip resistances. Spherical cavity expansion p re

dicted a lim iting pressure which is closer to tip resistance than the cylindrical cavity

expansion. This theory still needs the actual expansion th a t takes place under the

cone. T he second procedure of cavity expansion theory is based on a soil m odel

proposed by Ju ran and Beech (1987) and it predicts lower lim iting pressures than

the first theory. Differences are a ttr ib u ted to the assum ptions involving the dilation

angles. T he K values required for m easured friction resistances show th a t they are

g rea ter th an Ko- This is due to the dilation characteristics of th e sand.

A nev/ chart is developed for estim ating the streng th properties using bearing ca

pacity theories. Com parisons w ith present results show good predictions of s treng th

properties. However, th is m ethod still needs to be checked in the field.

Two em pirical m ethods are also presented for estim ating the relative density and

unconfined compression s treng th in cem ented deposits. T he first m ethod is based

on th e a m ethod and can be used in two approaches. C harts for various confin

ing pressures are presented in the first approach. Block sam pling and unconfined

com pression testing are required for th is approach. W hen block sam pling is not fea

sible, th e second approach can be used. The second em pirical m ethod is based on


269

oû .

1 0 08 060

4 0

ocr

>HO<û-<oo

or<ÜJOÛ

ÜJzooXoh—Z)Q

- V E R Y SHELLY SAN DS LIMEROCK

20

08

0.80.6

0 .4

0.20

From Bergemann ( 1965) and based on correlations

North Central Floridain

DENSE OR GEtdENTED

S A N D

LOOSE

S I L T - S A N D MIXTURES

CLAYEY SANDS, AND SILTS

'INSENSITIVE 'NON-FISSURED

INORGANIC CLAYS

-VgRY S T IF F ,

S T I F F

M E D I U M

S O F T

V E R Y SOFT________ I__________ L

ORGANIC CLAYS AND

MIXED SOILS

2 3 4 5FRICTION RATIO ( % )

7

P R E S E N T STUDY

CZ) CEMENTED S A N D S

EH UN CE ME NT ED SANDS

Figure 7.36: Classification C hart (Schm ertm ann, 1978)


270

GRAVELA N D

S A N D S

S A ND S (Dean

cn 8 0

S A N D M I X T U R E S ( S i l t y S a n d

C l a y e y S a n d S a n d y S i l t ]

= 20

O Ü J » -

S IL T M I X T U R E S ( S a n d y S i l t

C l a y e y Silt S i l ty CIc

C o n e e x p o n e n t v a l u e

C L A Y S ( C l a y

Si l ty Cloy]

Equi va l entusesr a n g e

0 . 2 0 . 4 0 .6 I

CORRE CT E D FRICTION RATIO ( % ) IN T E R M S OF t s f

f s IFR 100

Figure 7.37; Classification C hart (Douglas and Olsen, 1981)


271

Ll(f)

ÜJOZ<k—œO)ÜJtrÜJzoo

< CE CO

Q

Z ÜJ LU Q S

LU O

SA N D

CLAy

SOFT I N O R G . CLAY

VERY SOFT INORG. CLAY

2 3 4

FRICTION RATIO ( % )

P R E S E N T STUDY

El CEMENTED SANDS

E l U N C E M E N T ED SANDS

Figure 7.38: Classification C hart (Tumay, 1985)


272

1 0 0 0

P R E S E N T STUDY

E3 CEMENTED S A N D S

m UN CEMENTED SANDSk.a

100k—cr

OZ£T<LUCO

LUZoo

FRICTION RATIO ( % )

ZO NE q y / N S O I L BEHAVIOUR TY P E

1 2 S E N S I T I V E FINE GR A IN E D2 I ORGANIC M A T E R I A L3 I CLAY4 1 . 5 SI LTY CLAY TO CLAY5 2 CLAYEY SI LT TO SILTY CLAY6 2 . 5 S A N D Y SILT TO CLAY EY SILT7 3 S I L T Y S A N D TO SANDY SI LT8 4 S A N D TO SILTY SAND9 5 S A N D

10 6 GRAVELLY SAND TO SA N D11 I V E R Y S T I F F FINE GR AI N E D12 2 S AN D TO CLA Y EY SAND

- - - L iq u e f ia b le Z o n e

Figure 7.39: Classification C h art (R obertson and C am panella, 1986)


273

th e steady s ta te concept. This approach requires a prior knowledge of cem entation.

Hence, a classification chart which provides a zone for cem ented m ateria ls is p re

pared . Any cone tes t resu lt th a t lie w ithin the zone defined by the chart is likely

to ind ica te cem entation. Once cem entation is known, %l> can be determ ined by using

th e p roper charts. T he ^ - streng th correlations provides streng th param eters. B oth

these m ethods are need to be assessed in the field. The existing classification charts

are also updated to include cemented sand zones.


Chapter 8

SUMMARY AND CONCLUSIONS

8.1 Sum m ary

N atu ra l cem ented deposits are very common throughout different s ta tes in US

and also different pa rts of the world. Generally, cem entation effect on streng th

properties of sand is neglected since cem entation often improves them . However,

recent studies indicate th a t neglecting cem entation, particu larly the sm aller degree

of cem entation bonds results in overestim ation of the liquefaction resistance and u n

derestim ation of th e stab ility of slopes and strength properties. This signifies the

im portance of identifying cem entation in the natu ra l deposits. Difficulty in sam pling

n a tu ra lly cem ented deposits prom pts the need of using insitu testing m ethods. Cone

p ene tra tion testing is one such m ethod which is gaining wide acceptance and use

in the USA and th e world due to its repeatability , economy and capability to p ro

vide accurate, repeatab le vertical soil profiles and pertinen t engineering param eters

rela ted to th e sounded deposits.

Prelim inary studies indicated th a t cem entation resulted in an increase of tip and

friction resistance. T he existing in terpreta tions which are developed for clean sands

would be invalid for cem ented sands. A calibration cham ber study was conducted.

M onterey No. 0/30 sand and ordinary portland cem ent (T ype I) were used

in specim en preparation. P luviation m ethod was adopted since it best sim ulates

the n a tu ra l cem ented specim en structu re . S trength tests (triax ial and unconfined

compression tests) and calibration cham ber tests were conducted. T riaxial tests

showed th a t cem entation induces cohesion in tercept, thereby increasing the overall

s treng th . Unconfined compression tests showed th a t denser cem ented specim ens

display a m axim um qj value of 50 kPa. These results are la ter used in in terp re ting

cone penetration te s t results.

T he cone tests were conducted in a large scale calibration cham ber, which can

house a sam ple of 0.53 m in d iam eter and 0.79 m in height. Specim ens were prepared

a t th ree ranges of relative densities (45 - 55, 65 - 75 and above 80 %) and tested at

274


275

th ree confining stresses (100, 200 and 300 kPa). A m iniature quasi-static pene trom

e te r of 1.27 cm in d iam eter was used in th e tests to give a d iam eter ra tio of 42. For

cem ented specim ens, a curing period of 7 days was adopted.

Test results on uncem ented sands were first assessed for repeatab ility and accu

racy. T hen , these resu lts were com pared w ith th ree previous investigations conducted

on sim ilar type of sand. These com parisons showed th a t boundary conditions, cham

ber size effects, cem entation , grain size and shape influence the test results.

An em pirical in te rp re ta tio n scheme was suggested by norm alizing th e cone test

resu lts w ith respect to octahedral stresses. This approach was then used in p repar

ing two prediction m ethods; one in which unconfined compression stren g th of the

tes ted sand is known through block sam pling and the second when such sam pling

procedures are not available. These procedures need to be validated by field tests.

T he param eters used in th e above norm alization, a and n were also investigated to

assess th e influence of various param eters on cone test results.

Two hearing capacity theories and sm all strain cavity expansion theory were

used in pred icting tip resistances. D urgunoglu and M itchell (1973) and Jan b u and

Senneset (1974) bearing capacity theories predicted m easured tip resistances quite

closely. T he lim iting pressure pred icted by spherical cavity expansion was close to

tip resistance. Cylindrical cavity expansion predicted lim iting pressures lower than

m easured tip resistance. T he cavity under the cone which is different from the

idealized cavity m ay resu lt in th is difference. The K values back calcu lated from

sleeve friction values in testing were generally greater than Ko- T he trend displays

K being closer to Ko a t lower densities. Bearing capacity theories were used in

form ulating an approach to predict th e s treng th param eters of cem ented sands.

C ritical s ta te concept was used in preparing another em pirical approach in p re

diction. A classification chart is provided which gives cem entation value when cone

test resu lts are known. Once it is established th a t there is cem entation, p roper fig

ures have to be used for estim ating the B een’s param eter, ip. T he ip> value provides

streng th param eters from the correlations.


276

8.2 C onclusions

T he following conclusions are draw n from th is study.

1. P luv ia tion was found to be th e best process in specimen p repara tion of bo th

cem ented and uncem ented specim ens. P luv ia tion or rain ing process showed

th a t the sieve sizes, sh u tte r porosities and th e height of fall (d istance betw een

th e end of th e sieve to th e top of deposited sand) affect th e relative densities

of th e specim ens. Specim ens of 40 to 90 % relative density were p repared by

varying these variables. D ensity tests along th e depth of the specim en ind ica ted

th a t uniform specim ens can be prepared w ith th e pluviation process. Scanning

m icroscope photographs of th e cem ented sam ples a t different dep ths of th e

specim en showed th a t cem ent bonding developed all along the dep th . T he

segregation of cem ent during pluviation was insignificant.

2. Tests conducted w ith a piezocone in a specim en prepared a t 85 % rela tive

density and 2 % cem ent content showed th a t there is no excess pore pressure

developed during penetration . This implied th a t the reduction in hydraulic

conductiv ity due to cem entation will not result in undrained conditions during

cone penetration .

3. A ccuracy was assessed by com paring the uncem ented results w ith th ree o ther

investigations (Villet and M itchell, 1981; Schm ertm ann, 1978; B aldi, 1981).

V ariations between th e te s t results of this study and those of o ther investiga

tions are a ttr ib u ted to th e boundary conditions, K q values, d iam eter ratio , size

and shape of th e aggregates and specim en preparation procedures.

4. Two specim ens of sim ilar density were tested under two different boundary

conditions 1 (constant stress) and 3 (zero lateral displacem ent). B oth produced

sim ilar results and th is was a ttr ib u ted to the d iam eter ra tio in the present tes ts

which is around 42. A study on the influence of diam eter ra tio on various cone

te s t results showed th a t d iam eter ra tio is less im portan t when the relative

density is lower th an 40%. However, d iam eter ratios of 40 and above 50 are

needed while testing specim ens w ith 80 and 90 % relative density respectively.


277

5. C em entation increased tip resistance due to developm ent of cohesion and fric

tion resistance due to dilation.

6. In th e theoretical approaches, two bearing capacity theories were used. D ur

gunoglu and M itchell (1973) predictions showed a good agreem ent w ith m ea

sured values. T he rigid p lasticity assum ption was used in th e bearing capacity

theories. C em ented and uncem ented sands showed th is rigid p lastic behavior

a t and around peak stress, hence the theory which used peak streng th param e

ters qu ite well p red icted the m easured resistances. Janbu and Senneset (1974)

predictions depend upon the plastification angle. E stim ation of p lastification

angles are form alized by providing a correlation betw een the plastification an

gle and the m axim um dilation angle. This theory also rendered qu ite good

com parisons.

7. Back calculations from sleeve friction values showed th a t a K value higher than

Ko is needed to m atch the m easured friction values. D uring cone pene tration ,

th e d ilation around the cone is restrained. As a consequence, the confinem ent

a t th e interface increases resulting in an increase in the m easured K value.

8. B earing capacity theories and sleeve friction predictions are used in form ulat

ing a semi-em pirical approach to predict cem ented soil characteristics. This

approach proposes a m ethod to predict cohesion and relative density based on

the norm alized cone tip resistance and friction ratio . Once the relative density

is ob tained , friction angle can be estim ated.

9. Two cavity expansion theories w ith different soil m odels are used. The first

theory assum es the soil m edium to be elastic - perfectly plastic. T he second

theory assum es elasto - p lastic soil behavior. D ilation angles are properly

represented in th e second theory whereas the first theory assum es a constan t

d ilation angle. Therefore, lower lim iting pressures are pred icted in the second

theory. The ra tio of tip resistances to lim iting pressures are com pared w ith the

form factor suggested by Vesic (1977). There seems to be reasonable agreem ent

betw een these values a t relative densities of 45 - 55 % and 65 - 75 % range.

However, there is variation at higher relative densities (above 80 %) and the

reasons for th is are well established.


278

10. An em pirical approach based on the norm alization of bo th tip and friction

resistances is proposed. Norm alized param eters, a and /3 increase w ith both

cem entation and relative density. O ther param eters, n and tii are considered as

constan ts for a particu lar sand. Com parison of the results w ith various sands

showed th a t n and rii values depend on the angularity of th e sand particles

and decrease w ith cem ent content.

8.3 R ecom m endations for Future Studies

T he following topics are recom m ended for fu ture studies in th is area.

1. T he accuracy of the proposed schemes can be improved if tests are conducted

a t higher cem entation levels (4 and 6 %), lower confining stresses (less than

100 kPa).

2. It seems th a t there is very little study on sleeve friction. It is necessary to

b e tte r understand the developm ent of sleeve friction.

3. T his study shows th a t the effect of cem entation on cone penetration testing

can be predicted reasonably well w ith proposed theoretical models. However,

field tests are necessary to validate the findings of th is study.


R eferences

® Acar, Y .B ., “Piezocone P enetra ting Testing in Soft Cohesive Soils,” Louisiana S ta te University, D epartm ent of Civil Engineering, Fugro Postdoctoral Fellowship, Activity R eport No. 4, 1981.

0 Acar, Y .B ., Durgunoglu, T. and Tumay, M .T ., “Interface Properties of Sand,” Journal of Geotechnical Engineering, ASCE, Vol. 108, GT 4, pp. 648-654, 1982.

• Acar, Y.B. and Tum ay, M .T., “Strainfield Around Cones in Steady P enetratio n ,” Journal of Geotechnical Engineering, ASCE, Vol.112, No.2, pp 207-213, February, 1986.

• Acar, Y.B. and El-Tahir, E.A ., “Low Strain Dynam ic P roperties of Artificially C em ented Sand,” Journal of Geotechnical Engineering, ASCE, Vol. 112, No. 11, pp. 1001-1015, November, 1986.

• Acar, Y .B ., “T im e D ependent S trength Gain in Freshly Deposited or Densified Sand,” by Jam es K. M itchell and Zoltan V .Solym ar, Journal of Geotechnical Engineering, Vol. 113, No.2, February, 1987.

o Akili, W . and Nabil M.A. A l-Joulani, “Cone Penetration Tests on Artificially C em ented Sands,” Proceedings of the F irst In ternational Symposium on Pene tra tion Testing, Volume.2, lSO PT -1, Orlando. Florida, pp 607-614, March 1988.

0 A l-Ghanem , .\bdu lhak im , M .E .,“Factors Affecting the S trength C haracteristics of Calcium C arbonate-C em ented Soils,” Ph.D . D issertation, Lhe University of Arizona, 1989.

» Alm eida, M.S.S., Jamiolkowski, M. and Peterson, R .W ., “Prelim inary Results of CRT Tests in Calcareous Quiou Sand,” Proceedings of the F irst International Sym posium on C alibration C ham ber Testing (ISOCCTT), Potsdam , NY. pp. 41-54, June 1991.

9 Arslan, S., “Cone Penetration in C em ented Deposits; A Field S tudy,” M asters Thesis, August, 1993

® Am erican Society for Testing and M aterial S tandards D 3441, “S tandard M ethod for Deep Q uasi-static Cone and Frict ion Cone Penetration Tests of Soils,” 1979.

8 A vram idis, A. and Saxena, S.K ., “Behavior of Cem ent Stabilized Sands IT d e r S tatic and Dynam ic Loads,” R eport No. 11T-CF85-03, D epartm ent of Civil Engineering, Illinois Inst itu te of Technology. Septem ber, 1985.

279


280

• Bachus, R .C ., Clough, G .W ., S itar, N., Shafii-Rad, N ., Crosby, J ., and K aboli, P., “Behavior of W eakly Cem ented Soil Slopes U nder S ta tic and Seismic Loading C onditions,” Volume II, U .S.G.S., C ontract No. USGS 14-08.0001-18380, R eport No.51, July, 1981.

• Baldi, G ., B ello tti, R ., G hionna, V., Jam iolkowski, M. and Pasqualini, E ., “Cone R esistance of a Dry M edium Sand,” 10th In ternational Conference on Soil M echanics and Foundation Engineering, Stockholm , Vol.2, pp. 427-432., 1981.

® Baldi, G ., B ello tti, R ., G hionna, V., Jamiolkowski, M. and Pasqualini, E ., “Cone R esistance in Dry N.C. and O.C. Sands,” A SCE Geotechnical Division, Sym posium on Cone P ene tra tion Testing and Experience, St. Louis, pp. 145- 177, 1981.

« Baligh, M. M. and Levadoux, J . N ., “Pore P ressure D issipation A fter Cone P enetra tion , Sea G ran t R eport No. 80 - 13, M assachusetts In stitu te of Technology, C am bridge, M ass., 1980.

o Baligh, M. M and Scott, R .R .,“Q uasi-static Deep P ene tra tion in Clays,” Jo u rnal of G eotechnical Engineering division. Vol.101, Novem ber, 1975.

o Baligh, M. M, Azzouz, A.S., W issa, A .Z.E., M artin , R .T . and M orrison, M .J., “T he Piezocone P ene trom eter,” ASCE, Geotechnical Division, Sym posium on Cone P ene tra tion Testing and Experience, St. Louis, pp. 247-263, 1981.

» Baligh, M. M., “U ndrained Deep P enetration , I: Shear Stresses,” Geotcchnique, Vol. 36, No. 4, 1986.

® Beckwith, G. H. and Hansen, L. A., “T he Calcareous Soils of the Southw estern U nited S ta tes ,” Presented at ASTM Sym posium on Calcareous Soils, F t. Lauderdale, Florida, Jan . 20, 1981.

8 Been, K ., Crooks, J . H. A., Becker, D .E., and Jefferies, M .G., “T he Cone P ene tra tion Test in Sands, P a r t 1: S ta te Param eter In te rp re ta tio n ,” Geotechnique, Vdl: 36, pp. 1 8- 132,1986.

® Been, K ., Jefferies, J . H., Crooks, A. and R othenburg, L, “T he Cone P en e tra tion Test in Sands, P a rt II: General Inference of S ta te ,” Geotechnique Vol: 37, pp. :%%L299, 1987.

8 Been, K ., Crooks, J . H. A and R othenburg, L., “A C ritical Appraisal of C P T C alibration C ham ber T ests,” Proceedings of th e F irst In ternational Sym posium on P enetra tion Testing, ISO PT -l,V olum e.2, O rlando, Florida, pp. 651-660, M arch 1988.

® Been, K., Jefferies, M. G., and Hachey, J ., “The C ritical S ta te of Sands,” G eotechnique 41, No. 3, pp. 365-381, 1991.


281

• B ello tti, R., Bizzi, G. and G hionna, V. N., “Design, C onstruction and Use of a Calibration C ham ber,” Proceedings, 2nd European Sym posium P enetra tion Testing, A m sterdam , pp. 439-446, 1982.

9 B ello tti, R., C rippa, V ., Pedroni, S., Baldi, C ., F re tti, C ., O stricati, D., C hionna, V., Jamiolkowski, M ., Pasqualini, E ., “L aboratory V alidation of In- s itu T ests,” in Geotechnical Engineering in Italy (Assoc. C eotecnica Ita liana), published for ISSM FE Golden Jubilee), pp. 251-270, 1985.

® B ello tti, R., C rippa, V., Pedroni, V. and C hionna, V .N ., “S aturation of Sand Specim en for C alibration C ham ber T ests,” Proceedings of the F irst In te rnational Sym posium on P enetra tion Testing, Volume.2, ISO PT -1, O rlando, F lorida, pp. 661-672, M arch 1988.

• B ello tti, C, F retti, V. N. C hionna and Pedroni, S., “Com pressibility and C rushability of Sands a t High Stresses,” Proceedings of the F irst In ternational Sym posium on C alibration C ham ber Testing (IS O C C T l), Potsdam , NY, pp. 79-90, June 1991.

• Beringen, S. L., Koike, H. J and W indle, D .,“Cone P ene tra tion and L aborato ry Testing in M arine Calcareous Sedim ents,” Geotechnical P roperties, Behavior and Perform ance of Calcarious Soils, ASTM Sym posium , F t. Lauderdale, F lorida, January, 1982.

9 Bishop, A. W. and Henkel, D. J ., “T he Triaxial Test,” Edw ard Arnold, London, 1962.

« B jerrum , “Allowable Settlem ents of S tructu res,” Proceedings, European Conference on Soil Mechanics and Foundation Engineering, W iesbaden, Germany, Vol III, pp. 135-137, 1960.

® Brinch, H. J ., “A General Form ula for B earing C apacity ,” Bulletin No. 11, D anish Geotechnical In stitu te , Copenhagen, pp. 38-46, 1961.

• B ryan Sweeney., “Liquefaction Evaluation Using A M iniature Cone P enetrom e te r and A Large Scale C alibration C ham ber,” Ph.D . D issertation, S tanford University.

® Cam panella, R. G and R obertson, P. K., “A pplied Cone Research,” Sym posium on Cone P enetration Testing and Experience, Geotechnical Engineering Div., ASCE, O ctober, 1981, pp. 343-362.

» C am panella, R. G and R obertson, P. K., “C urrent S ta tus of Piezocone T est,” Proceedings of the F irst In ternational Sym posium on P enetra tion Testing, Volume. 1, ISO PT-1, O rlando, F lorida,pp 93-116, M arch, 1988.

® Canon, J ., Hachem, M. EL, K a ttan , A. and Ju ran ,!., “M ini Piezocone (M- C PT U ) Investigation Related to Sand Liquefaction A nalysis,” Proceedings of


282

the F irs t In ternational Sym posium on P enetration Testing, Volume.2, ISO PT- 1, O rlando, Florida, pp 699-706, M arch, 1988.

• C arter, J . P., Booker, J . R ., Yeung, S. K., “C avity Expansion in Cohesive Frictional Soils,” G eotechnique, Vol. 36, pp. 349-358, 1986

• Chang, T. S., ’’Dynam ic Behavior of Cem ented Sand,” Ph.D . Thesis, U niversity of M ichigan, June, 1986.

• Chang, C. S., M isra, A. and Sundaram , S. S., “M icrom echanical M odelling of Cem ented Sands U nder Low A m plitude O scillations,” G eotechnique 40, No. 2, pp. 251-263, 1990.

» C hapm an, G. A., “A C alibration C ham ber for Field Test E qu ipm en t,” P ro ceedings, European Sym posium on P enetration Testing, Stockholm , Vol.2.2, pp.59-65, 1974.

« Chiang, Y. C. and Chae, Y. S .,“Dynam ic P roperties of Cem ent T reated Soils,” Highway Research Record, N ational Research Council, No. 379, 1972, pp. 39-51. '

® Chung, R. M ., Yokel, F. Y., and W echsler, H., “Pore Pressure B uildup in Resonant Colum n T ests,” Journal of Geotechnical Engineering, ASCE, Vol. 110, NO. 2, pp. 247 - 261.

o Clough, G. W ., S itar, N., Bachus, R. C., “Cem ented Sands U nder S ta tic Loading,” Journal of Geotechnical Engineering, Vol. 107, GT6, pp. 799-818, June, 1981.

® de Lim a, D.C. “Developm ent, Fabrication and Verification of the LSU In-situ Testing C alibration C ham ber,” Ph.D . D issertation, Louisiana S ta te University, B aton Rouge, Louisiana, 1990.

® Dario de L im a (1989), Personal Com m unication.

9 de Lim a, D.C and Tumay, M .T ., “Calibration and Im plem entation of M iniature Electric Cone Penetrom eter and Development, Fabrication and Verification of the LSU In Situ Testing C alibration C ham ber (LSU /C A LC H A S),” Research R eport No. GE-92/08, S ubm itted to Louisiana T ransporta tion Research Center, B aton Rouge, Louisiana, 1992.

® de Lim a, D. C. and Tumay, M. T ., “Scale Effects in Cone P ene tra tion T ests,” Geotechnical Engineering Congress 1991, GT D iv/A SC E, Boulder, CO, 1991, pp. 38 - 51.

0 D atta , M., G ulhati, S. K. and Rao, G. V., “Engineering Behavior of C arbonate Soils of Ind ia and Some Observations on Classification of Such Soils,” Geotechnical Properties, Behavior and Perform ance of Calcarious Soils, ASTM Sym posium , F t. Lauderdale, Florida, January, 1982.


283

c DiM aggio, F. L., and Sandler, I. S., “M aterial Model for G ranular Soils,” J. Eng. Mech., ASCE, Vol. 97, EM 3, pp. 935-950, 1971

• Douglas, B. J ., and Olsen, R. S., “Soil Classification Using E lectric Cone Pene trom ete r,” Proceedings, Cone P enetra tion Testing and Experience, ASCE, St. Louis, MO, 1981, pp. 209 - 227.

9 D upas, J. M and Pecker, A., “S tatic and D ynam ic P roperties of Sand-cem ent,” Jou rna l of Geotechnical Engineering Division, ASCE, Vol. 105, No. GT3, M arch, 1979.

9 D urgunoglu, H. T and M itchell, J . K., “S tatic P enetra tion R esistance of Soils," Space Sciences Laboratory, University of California, Berkeley, April, 1973.

9 E l-T ahir, E. A., “Low S train D ynam ic P roperties of Artificially C em ented M onterey No. 0/30 Sand,” M aster’s Thesis, Louisiana S ta te University, B aton Rouge, 1985, 96 pages.

9 E id, W . K., “Scaling Effect in Cone P enetra tion Testing in Sand,” Ph .D . Disserta tio n , Civil Engineering, V irginia Polytechnic In stitu te and S tate University, V irginia, 1987.

« Fioravante, V ., Jamiolkowski, M ., Tanizawa, F ., and Tatsuoka, F ., “R esults of C P T ’s in Toyoura Q uartz Sand,” Proceedings of the F irst In ternational Sym posium on C alibration C ham ber Testing (IS O C C T l), Po tsdam , NY, pp. 135-146, June 1991.

• Frydm an, S., “Calcareous Sands of the Israeli Coastal P la in , Geotechnical P roperties, Behavior and Perform ance of Calcareous Soils,” ASTM Sym posium , F t. Lauderdale, F lorida, January , 1981

® Frydm an, S., H endron, N. D., Horn, H., Steinbach, J ., Baker, R and Shaal, B, “Liquefaction Study of Cem ented Sands,” Journal of Geotechnical Engineering Division, ASCE, Vol. 106, No. GT3, M arch, 1980.

0 G hionna, V. N., and Jamiolkowski, M., “A Critical A ppraisal of C alibration C ham ber Testing of Sands” , Proceedings of the F irst In ternational Sym posium on C alibration C ham ber Testing (IS O C C T l), Potsdam , NY, pp. 13-40, June 1991.

9 Greeuw, G., Sm its, F . P., van Driel, P., “Cone Penetration Tests in D ry Ooster- schelde Sand and th e Relation w ith Cavity Expansion M odel,” Proceedings of th e F irs t In ternational Sym posium on P enetra tion Testing, Volume. 2, ISO PT- 1, O rlando, Florida, pp. 771-776, M arch, 1988.

9 G uerm azi, A., “Analyse Théorique e t Expérim entale du C om portem ent d ’un Sol Renforce par Colonnes B allastees,” Ph. D. Thesis, Paris V l-E N PC , 1986.


284

• H arm an, D. E., “A S ta tistica l S tudy of S tatic Cone Bearing C apacity , Vertical Effective Stress, and Relative Density of Dry and S a tu ra ted F ine Sands in a Large, Triaxial Test C ham ber,” M aster’s Thesis, University of F lorida, Gainesville, Florida, 1976, 180 pages.

• H am el, J . V., “Large Scale L aboratory D irect Shear Tests on D esert A lluvium ,” Jou rna l of Rock M echanics, pp. 385-414, 1973.

« Holden, J . C., “T he C alibration of Electrical Penetrom eters in S and ,” Norwegian Geotechnical In stitu te , Int. Rep. 52108-2, 1977.

® Holden, J . C., “L aboratory Research on S ta tic Cone P ene trom eters ,” R eport No. CE-SM-71-1, D epartm ent of Civil Engineering, University of F lorida, 1971.

® Houlsby, G. T and H itchm an, R ., “Calibration C ham ber Tests of a Cone P ene trom eter in Sand,” G eotechnique 38, N o.l, p p .39-44, 1988.

e Hughes, J . M. O., W roth, C. P. and W indle, D., “Pressurem eter Tests in Sands,” Geotechnique, Vol. 27, pp. 455-477, 1977.

» Jaky, J ., “T he Coefficient of E a rth Pressure a t R est,” Journal of the Society of H ungarian A rchitects and Engineers, pp. 355-358, 1944.

9 Jam iolkow ski, M .,“ New Developments in Field and L aboratory Testing of Soils,” Them e lecture, 11th In ternational Conference on Soil M echanics and Foundation Engineering, San Fransico, 1985.

® Jan b u , N and Senneset, K., “Effective Stress In te rp re ta tion of In-situ S tatic P ene tra tion Tests,” Proceedings of the E uropean Sym posium on P ene tra tion Testing, E SO PT I, Stockholm , Sweden, Vol. 2.2, pp. 181-193, 1974.

® Ju ran , I and Beech, J .F ., “Effective Stress Analysis of Soil Response in a P ressurem eter Test,” The Pressurem eter and Its M arine Applications: Second In ternational Sym posium , ASTM ST P 959, J.-L . Briaud and J.M .E . A ud ibert, E ditors, ASTM , 1986.

® Ju ran , I. and G uerm azi, A., “Settlem ent Response of Soft Soils Reinforced by C om pacted Sand Colum ns,” ASCE Journal of Geotechnical Engineering, Vol. 114, pp. 930-943, 1988.

e Ju ra n , I. and M ahm oodzadegan, B., “An In te rp re ta tion Procedure for P ressurem eter Tests in Sand,” ASCE, Journal of Geotechnical Engineering, 1989.

® Ju ran , I and Tumay, M .T ., “Soil S tratification Using D ual Pore P ressure P iezocone Test (D P C P T ),” Published in T ransportation Research Board 68th A nnual M eeting, January, 1989, pp. 22-26, W ashington D.C.

9 K rynine, D. P and Judd , W . R ., “Principles of Engineering Geology and G eotechnics,” M cGraw Hill, New York, 1957.


285

« Kiousis, P. D .,“ Large S train Theory as A pplied to P ene tra tion M echanism in Soils,” Ph.D . D issertation, Civil Engineering departm en t, Louisiana S ta te University, B aton Rouge, A ugust, 1985, 97 pages.

• Kiousis, P. D., Voyiadjis, G. Z., Tumay, M. T . ,“A Large S train T heory and its A pplication in the Analysis of Cone P ene tra tion M echanism ,” In ternational Journal of th e Num erical and A nalytical M ethods in Geom echanics, Vol. 12, No. 1, February 1988, pp. 45-60.

9 Ladanyi, B., “Expansion of a Cavity in a S a tu ra ted Clay M edium ,” Jou rna l of th e Soil M echanics and Foundations Division, ASCE, Vol. 89, No. SM4, pp. 127-161, July, 1963.

» Ladd, C., “Shear S trength of Cohesive Soils,” Illinois L ecture Series, 1966.

9 Lee, K. L., “Form ation of Adhesion Bonds in Sands a t High P ressure ,” R eport No. UCLA. Eng. 7586, UCLA School of Engineering and A pplied Science, Los Angeles, California, O ctober, 1975.

9 Lhuer, J . M ., “An E xperim ental Study of Q uasi-S tatic Cone P en e tra tio n in S atu ra ted Sands,” M aster’s Thesis, Civil Engineering, U niversity of F lorida, Gainesville, Florida, 1976.

9 Li L ore tta and M itchell, R .J ., “Effect of Reinforcing E lem ents on the Behavior of W eakly Cem ented Sands,” Canadian Geotechnical Jou rnal, Vol.24, 1988.

® Lunne, T . and Christoffersen, “In te rp re ta tion of Cone D ata for Offshore Sands,” 15th Offshore Technology Conference, Houston, Texas, Proceedings, Vol. 1, pp. 181-192, 1983.

9 Lunne, T ., “P ractical Use of C P T Correlations in Sand Based on C alibration C ham ber Tests,” Proceedings of the F irst In ternational Sym posium on Calibration C ham ber Testing (IS O C C T l), Po tsdam , NY, pp. 225-235, Ju n e 1991.

0 M anassero, M, “C alibration C ham ber C orrelations for H orizontal In situ Stress Assessment Using Self-Boring P ressurem eter and Cone P ene tra tion T ests,” Proceedings of th e F irst In ternational Sym posium on C alibration C ham ber Testing (IS O C C T l), Potsdam , NY, pp. 237-248, June 1991.

9 M ayne, P. W . and Kulhawy, F. H., “C alibration C ham ber D atabase and B oundary Effects Corrections for C P T D ata ,” Proceedings of th e F irst In ternational Symposium on C alibration C ham ber Testing (IS O C C T l), Potsdam , NY, pp. 257-264, June 1991.

9 Mckown, A. F and Ladd, C. C., “Effect of C em entation on th e Com pressibility of P ierre Shale,” Geotechnical P roperties, Behavior and Perform ance of Calcareous Soils, ASTM Sym posium , F t. Lauderdale, F lorida, January , 1981.


286

e M eyerhof, G. G .,“Some Recent Research on the Bearing C apacity of Cohesion- less Soils,” C anadian Geotechnical Journal, Vol. 1, No. 1, pp. 16-26, 1963.

® M itchell, J. K ., “Fundam entals of Soil Behavior,” John W iley & Sons, Inc., New York, N.Y., 1976.

• M itchell, J. K ., Cuzikowski, F and V illet, W . C. B., “T he M easurem ent of Soil P roperties In -situ ,” P repared for US D epartm ent of Energy, D epartm ent of Civil Engineering, U niversity of California, Berkeley, M arch, 1978.

• M itchell, J. K ., “T he P roperties of Cem ent-stabilized Soils,” P aper for W orkshop on M aterials and M ethod for Low Cost Road, Rail and R eclam ation W ork, Leuro, A ustralia , Septem ber, 1980.

# M itchell, J . K ., and Solymar, Z. V., “T im e D ependent S treng th Cain in Freshly Deposited or Densified Sands,” Journal of Geotechnical Engineering, ASCE, Vol. 110, No. 11, pp. 1559-1576, 1984.

® M itchell, R. J and Stone D. M., “Stability of Reinforced C em ented Backfills,” C anadian Geotechnical Journal, Vol. 24, pp. 189 - 197, 1987.

* N u tt, N .R .F. and Houlsby, C .T ., “C alibration Tests on the Cone P ressurem eter in C arbonate Sand,” Proceedings of the F irst In ternational Sym posium on C alibration C ham ber Testing (IS O C C T l), Potsdam , NY, pp. 265-276, June 1991.

e 0 ’ Rourke, T . D and Crespo, E., “Geotechnical P roperties of Cem ented Volcanic Soil,” Journal of Geotechnical Engineering, Vol. 114, No. 10, O ctober, 1988.

® Park in , A. K. and Lunne, T ., “B oundary Effects in the L aboratory C alibration of Cone Penetrom eter for Sand,” Proceedings, 2nd European Sym posium Penetration Testing, A m sterdam , pp. 761-768, 1982.

9 Parkin , A. K., “Evaluation of Calibration C ham ber D ata ,” Norwegian Ceotech- nical In stitu te , R eport No. 52108-18, 1986, 31 pages.

0 Parkin , A. K., “T he C alibration of Cone P enetrom eters,” Proceedings of the F irst In ternational Sym posium on Penetration Testing, V olum e.], pp .221-244, ISO PT-1, Orlando, Florida, M arch, 1988.

9 Poulos, H. C ., “Behavior of a Laterally Loaded Piles N ear a C ut or Slope,” A ustralian Ceom echanics Journal, Vol. C6, no. 1, pp. 6 - 12, 1976.

0 Poulos, H. G., “A Review of the Behavior and Engineering P roperties of C arbonate Soils,” Research R eport No. R 381, The University of Sydney, School of Civil Engineering, December, 1980.


287

• P uppala , A .J., Acar, Y .B ., and Tum ay, M .T ., “M iniature C P T Tests in Dense M onterey No. 0 /30 Sand in a Flexible Double-walled C alibration C ham ber,” Proceedings of the F irst In ternational Sym posium on C alibration C ham ber Testing (IS O C C T l), Potsdam , NY, pp . 339-350, June 1991.

» P uppala , A .J., Acar, Y.B. and Senneset, K., “Cone P enetra tion in Cem ented Sands: In te rp re ta tion by Two Bearing Capacity Theories,” ASCE, Journal of Geotechnical Engineering, November 1993 (in press).

0 R ad, N. S., and Clough, G. W ., “T he Influence of C em entation on the S tatic and D ynam ic Behavior of Sands,” R eport No.59, T he John A. B lum e E a rth quake Engineering Center, Stanford University, December, 1982.

• R ad, N. S., “Effect of C em entation on P enetration Resistance of Sand,” F inal R eport S ubm itted to Fugro In ternational, D epartm ent of Civil Engineering, R eport No. GE-84/01, Louisiana S ta te University, June 1984, 222 pages.

» R ad, N. S. and Tumay, M. T ., “Effect of C em entation on P ene tra tion Resistance of Sand - A Model S tudy,” Geotechnical Testing Journal, Vol. 9, No.3, Septem ber, 1986, pp. 117-125.

• R ad , N. S. and Tumay, M. T ., “Factors Affecting Sand Specimen P repara tion by R ain ing ,” Geotechnical Testing Journal, Vol.lO, N um ber 1, M arch, 1987, pp. 31-37.

« Riccobono, 0 . , “Experim ental Study to Analyze th e Influence of C em entation on th e Behavior of Sand Colum ns for Reinforcing a Soft Foundation Soil,” M.S. Thesis, Louisiana S tate University, 1985, 113 pages.

® Rix, G .J. and Stokoe, K .H ., “Correlation of In itial Tangent M odulus and Cone P ene tra tion R esistance,” Proceedings of the F irst In ternational Sym posium on C alibration C ham ber Testing (IS O C C T l), Potsdam , NY, pp. 351-362, June 1991.

® R obertson, P. K., and Cam panella, R . G., “Guidelines for Use and In terp reta tio n of the E lectric Cone P ene tra tion T est,” University of B ritish Colum bia, Civil Engineering D epartm ent, Soil M echanics Series, 1984.

® R obertson, P. K., and C am panella, R . G., “Liquefaction Poten tial of Sands Using the C P T ,” Journal of Geotechnical Engineering, Vol. I l l , No. 3, 1985, pp. 384 - 403.

® Sandven, R., “Strength and Deform ation P roperties of Fine G rained Soils Obtained From Piezocone T ests,” University of Trondheim , Trondheim , Norway, 1990.

9 Saxena, S. K. and Lastrico, R. M., “Static P roperties of Lightly Cem ented Sands,” Journal of Geotechnical Engineering, ASCE, Vol. 104, No. GT12, Decem ber 1978, pp. 1449-1464.


288

• Saxena, S. K ., Reddy, R. K. and A vram idis, A., “Liquefaction R esistance of Artificially Cem ented Sand,” Journal of G eotechnical Engineering, Vol. 114, No. 12, Decem ber, 1988.

a Schm ertm ann, J . H., “An U pdated C orrelation between R elative D ensity and Fugro T ype E lectric Cone B earing,” C ontract R eport DACW 9-76-M, 6646, W aterw ays E xperim ental S tation , V icksburg, MS, 1976.

a Schm ertm ann, J. H., “Guidelines for Cone P en e tra tio n Test Perform ance and Design,” Federal Highway A dm inistra tion , R eport FHW A-TS-78-209, W ashington, 1977.

a Schm ertm ann, J. H., “Guidelines for Cone P ene tra tion Test Perform ance and Design,” R eport. FHW A-TS-78-209, Federal Highway A dm instration , W ashington, 1978, 145 pages.

a Schm ertm ann, J. H., “T he M echanical Aging of Soils,” The T w enty-F ifth K arl Terzaghi Lecture, Jou rnal of Geotechnical Engineering, Vol. 117, No. 9, Sep, 1991, pp. 1288 - 1330.

a Schnaid, F and Houlsby, G. T . ,“An Assessment of Cham ber Size Effects in the C alibration of In S itu Tests in Sand,” Geotechnique, Vol. 41, No. 3, pp. 437-445, 1991.

e Salomone, L.A., Singh, H. and Fisher, J.A ., “Cyclic Shear S treng th of Variably Cem ented Sands,” Proceedings of the ASCE, Geotechnical Engineering Division, Specialty Conference on E arthquake Engineering and Soil D ynam ics, Pasadena, California, Vol.2, June , 1978, pp. 819-835.

9 Senneset, K ., “P enetra tion Testing in Norway,” Proceedings of the F irs t E uropean Sym posium on P ene tra tion Testing, Stockholm , Vol. 1, pp 85-95, 1974.

® Senneset, K ., Janbu , N ., and Svano, G., “S trength and Deform ation P aram eters from Cone P enetra tion T ests,” Proceedings of the Second E uropean Sym posium on P enetra tion Testing, A m sterdam , Vol. 2, pp 863-870, 1982.

e Senneset, K ., Sandven, R ., and Janbu , N ., “T he Evaluation of Soil P aram eters from Piezocone T ests,” Bulletin of T ransportation Research, Record NO. 1235, N ational Research Council, T ransporta tion Research Board, 34 pages, 1989.

® S itar, N., “Seismic Response of Steep Slopes in W eakly Cem ented Sands and G ravels,” Proceedings of H. Bolton Seed M emorial Sym posium , Ed. J.M . D uncan, Vol. 2, pp. 67-82, 1990.

® Sm its, F .P ., “P ene tra tion Pore Pressure M easured with P iezom eter Cones,” Proceedings of the Second E uropean Sym posium on Penetration Testing, E SO PT II, A m sterdam , Vol.2, pp. 877-881, 1981.


289

® Sowers, G.B. and Sowers, G .F .,“In troductory Soil M echanics and Foundation: G eotechnical Engineering,” 4 th E d., M cM illan Publishing Co., Inc., New York, 1979.

• Teh, C.I. and Houlsby, G .T ., “A n A nalytical S tudy of th e Cone P enetra tion Test in Clay,” G eotechnique 41, No. 1, pp. 17-34, 1991.

• Terzaghi, K .,“Theoretical Soil M echanics,” W iley, New York, 1943.

• Tum ay, M .T ., “R elative Density Versus C P T Cone B earing C orrelations in Cohesionless Soils,” (in Turkish) Bogazici U niversity P ress, Istanbul, Turkey, 1976.

® Tum ay, M. T ., “Fundam ental A spects of the P enetra tion M echanism in Soft Soils,” Research P roposal, December, 1980.

• Tum ay, M .T ., Boggess, R.L. and Acar, Y. B., “Subsurface Investigations w ith Piezocone P enetrom eter,” in ‘Cone P enetra tion Testing and Experience’ edited by G.M. Norris and R.D . Holtz, Proc. Session Sponsored by G eotechnical Engineering Division at ASCE, N ational Convention, St. Louis (M issouri), O ct. 1981.

o Tum ay, M .T ., “Field C alibration of Electric Cone P ene trom eter in Soft Soil - Executive Sum m ary,” Published by Louisiana T ransporta tion Research C enter, 1985.

® Tum ay, M .T ., Acar, Y .B ., M urat H. Cekirge, N arayanan Ram esh, “Flow Field A round Cones in S teady P ene tra tion ,” Journal of G eotechnical Engineering, Vol. I l l , No. 2, February, 1985.

« Torstensson, B .A ., “Pore Pressure Sounding In stru m en t,” Proceedings, ASCE Spec. Conference on In-situ M easurem ent of Soil P roperties, Vol. II, Raleigh, N .C ., pp.48-54, 1975.

9 Veismanis, A., “L aborato ry Investigation of Electrical Friction Cone P ene trom eters in Sands,” Proceedings, European Sym posium on P ene tra tion Testing, Stockholm , Vol. 2.2, pp. 407-419, 1974.

o Vesic, A.S., “Expansion of Cavities in Infinite Soil M ass,” Journal of Soil Mechanics, Foundation Div., ASCE, Vol. 98, pp. 265-290, 1972.

o V illet, W .C.B and M itchell, J .K ., “Cone Resistance, R elative D ensity and Friction Angle. Cone P enetra tion Testing and Experience,” Proceedings, Cone P ene tra tion Testing and Experience, ASCE N ational Convention, St. Louis, M o., pp. 178-208, 1981.

9 W issa, A.E.Z. and Ladd, C .C .,“Effective S tress-strength Behavior of Com pacted Stabilized Soils,” Research R eport R64-32, Soils P ublication No. 164, D epartm ent of Civil Engineering, M .I.T ., July, 1964.


290

9 W issa, A.E.Z. and Ladd, C.C., “Shear S trength G eneration in S tabilized Soils,” Research R eport R65-17, Soils Publication No. 173, D epartm en t of Civil E ngineering, M .I.T ., June, 1965.

e W issa, A .E .Z., M artin , R .T . and G arlanger, J .E .,“T he P iezom eter P ro b e ,” Proceedings ASCE Spec. Conference on In-situ m easurem ent of Soil P roperties, Raleigh, N .C ., Vol. 1, pp. 536-545, 1975.

9 Yamanouchi, T ., M ochinaga, R ., Cotoh, K. and M urata , H., “S ta tu s of C u toff Slopes in a Pum ice Flow Soil Deposit and their A pplications to the Design S tandards for an Expressway,” Proceedings of th e 9 th In te rnationa l Conference on Soil M echanics and Foundation Engineering, Tokyo, 1977.

« Yu, H.S. and Houlsby, C .T ., “F in ite Cavity Expansion in D ila tan t Soils: Loading A nalysis,” Ceotechnique 41, NO. 2, pp. 173-183, 1991.

9 de R uiter, J ., “T he S tatic Cone P enetra tion Test S tate-of-the-A rt R ep o rt,” P ro ceedings of th e Second European Sym posium on P enetra tion Testing, E S O P T II, A m sterdam , May 1982, Vol. 2, pp. 389-405.


A ppendix A

A .l L iterature on C em ented Sands

In this section, various investigations and their findings are presented. Table 1 presents the various variables studied in each investigation. Table 2 presents the sum m ary of conclusions of each investigation.

291


C D■DOQ.C

gQ .

■DCD Table A.1. A Summary of Geotechnical Studies Conducted on Cemented Sands

C/)wo'3O

8■D

i3CD

3.3"CD

CD■DOQ.CaO3

■DO

CDQ.

■DCD

C / )C / )

Reference Type of Sand Cementing AgentSpecimen

Preparation/Retrieval

Curing Period Types of Tests Conducted

Wissa, el al. (1964, 1965)

Coarse Ottawa uniform sand and medium Ottawa sand

Portland cement dry pluviation 3 days CU and CD triaxial tests

Chiang and Chae (1972)

Uniform sand and silty clay Cement, lima, fly ash Compaction 14 days Resonant column tests

Hamel (1973) Natural soils Calcium carbonates — — Direct shear tests

Mitchell (1976) Monterey No. 0 Portland Cement Compaction NA Indirect tension flexure

Yamanouchi, el al. (1977)

Natural Thermal welding Shirasu cutter (5 cm diameter tubes)

— Several types

Salomone, et al. (1978)

Natural Carbonates 76 m Denison sampler

— Undrained triaxial tests, cyclic triaxial tests

Saxena and Lastrico (1978)

Natural cemented sand near Vinceton, New Jersey

Carbonates, calcite cement

NA -- Isotropic consolidated triaxial tests, stress controlled cyclic triaxial tests

Dupas and Pecker (1979)

Medium-dense sands Portland Cement Compaction 7 days CD triaxial tests, dynamic triaxial tests, longitudinal forced vibration studies

Frydman, et al. (1980)

Kurkar deposits near Israel Calcareous materials Block samples, freezing

— (1)SPT(2) Cyclic triaxial tests

Poulos (1980) Natural Carbonates Different methods — Strength tests

Clough, et al. (1981)

Monterey #0 Monterey #20

Silicates, ironoscias and Portland Cement

Compaction 14 days Unconfined compression and drained triaxial corrpression tests

COs

C D■DOQ .C

gQ .

■DCD

C/)Wo"3O

8

ci'

Table A.1 (continued)

33"CD

CD■DOQ .CaO3

■DO

CDQ .

■DCD

C / )C / )




Datta, et al. (1982)

Natural Carbonates NA — Isotropic compression tests, drained triaxial shear tests

Beringen, Kolk and Windle (1982)

Calcareous sediments at several off-shore locations

Calcareous material Coring — (1 ) Cone penetration testing(2) Direct shear tests(3) Particle cmshing tests

McKown and Ladd (1982)

Pierry shales in Nebraska Calcium carbonate Core drilling - (1 ) Leaching tests (2) Consolidation tests

Rad and Clough (1982)

Monterey #0 and natural deposits

Carbonates (natural) Portland Cement

PluviationCompaction

14 days Drained and undrained triaxial tests

Rad and Tumay (1984)

Monterey #0 Portland cement Pluviation 7 to 14 days Static penetration tests

Avramidis and Saxena (1985)

Monterey #0 Portland cement Under compaction 15 days to 6 months

(1 ) Drained triaxial tests(2) Brazilian tests(3) Unconfined compression tests(4) Resonant column tests

Acar and Taliir (1986)

Monterey #0 Portland cement Pluviation 14 days Resonant column tests

Ghang (1986) Ottawa 20-30 Muskegon sand, mortar sand, medium sand

Sodium silicate, Portland cement, fly ash, lime

Injection and mixcompactionmethods

— NA

Mitchell and Stone (1986)

Mortar sand Cement NA 14 days Pullout tests

Loretta Li and Mitchell (1987)

Fine to medium sand Portland cement Mechanical mixer 14 days Plane strain teststococo

C D■DOQ .C

gQ .

■DCD Table A.1 (continued)

c/)wo"3O

8■D

CD

3.3"CD

CD■DOQ .C

aO3

■DO

CDQ .

■DCD

C / )C / )




Riccobono (1987) Louisiana river sand Portland cement Compaction 21 days Triaxial

O'Rourke and Crespo (1988)

fvtoderately cement fine sand and silt sized particles. Natural near the Andes of Ecuador and Colurrtiia

Amorphous silicate NA NA Uniaxial and triaxial compressive strengths, Brasilian tensile strength tests

Saxena, et al. (1988)

Artificial Monterey #0 Portland Cement Under compaction 15 to 60 days (1 ) Cyclic triaxial tests (2) Resonant column tests

5

C D■DOQ .C

gQ .

■DCD

C/)C/)

8■D

Table A.2. Synthesis of Data Reported in Studies Investigating Cemented Sands

3.3"CD

CD■DOQ.C

aO3"Oo

CDQ.

Reference Parameters Studied Conclusions from the Studies

Wissa, et ai. (1964, 1965)

D„ C.C., t,,, shear resistance

(1) At low strains, the shearing resistance was due to cementation between grains.(2) At higher strains, cementation between grains were completely destroyed and effective stress-strength

cun/e converge towards the origin on p vs. q plot.

Chiang and Chae (1972)

Cement content, confining pressure, shear strain amplitude, and moisture content

(1 ) Dynamic shear modulus and danrping can be greatly increased by adding a small amount of cement. (2) Effect of cement content is more pronounced in cement treated cohesionless soils.

Hamel(1976)

Shear strength parameters

The insitu peak strength parameters of this desert alluvium probably lie between the peak values detennined for granular specimens and specimens containing cemented lumps.

Saxena andLastrico(1978)

n. Yd, 4 , Sy (1) The stress-strain behavior, the pore pressure response, the stress paths versus strain plots, the relation between undrained strength versus the consolidating confining pressures indicate that the strength behavior of the natural cemented soils are strain dependent.

(2) At lower strains, the cohesion caused by the calcite cement bonding between particles is the major component of strength. At higher axial strains (around 1%), the cohesive strength is destroyed and then the frictional strength is predominant.

(3) It was also obsen/ed that a high hydrostatic confining pressure can destroy cementation.

Dupas andPecker(1979)

C.C., t„ E, G, C, K, cyclic strength

(1) Investigated on the minimum amount of cement content required for a particular type of sand so that it will be in stable under both static and dynamic loading.

(2) Another observation of this work was that only a small amount of cement is required to prevent liquefaction.

(3) Methods have been developed for the interpretation of static and dynamic moduli from the static and dynamic triaxial tests.

■DCD

C / )C / )

Footnote: S - static tests; D - dynamic tests; D, - relative density; C.C. - cement content; t - curing period; - confining pressure; M, - deformation modulus;- effective friction angle; S - undrained shear strength; K - permeability; E - Young's modulus; G - shear modulus; % - dry density; - volume change; -

unconfined compressive strength; - cone resistance; f, - frictional resistance; f - friction ratio; e - void ratiotoCOor

C D■DOQ .C

gQ .

■DCD

C/)C/)

8■D

CD

3.3"CD

CD■DOQ .C

aO3"OO

CDQ .

■DCD

C / )C / )



Frydman, et al. (1980)

cyclic strength parameters, stress ratio, Y

(1 ) The cyclic strength of intact specimens may be similar to that of reconstituted specimens compacted to the same infened relative density calculated from the insitu blow counts using the Gibbs and Holtz conelations.

(2) The block sarrpling and freezing techniques were found to be satisfactory methods for preparing intact specimens of granular soils.

Clough, et al. (1981)

C.C., D„ grain arrangement, strength of deformation parameters

(1 ) A v/eaWy cemented sand shows a brittle failure mode at low confining pressures with a transition to ductile failure at higher confining pressures.

(2) Volumetric strain increases during shear occur at a faster rate and at a smaller strain for cemented sand than uncemented sands.

(3) The residual strength of a cemented sand is close to that of an uncemented sand, although some degree of residual cohesion was obsen/ed for all the cemented sands investigated.

(4) The tensile strength of a cemented sand is about 10% of the unconfined compressive strength.

Beringen, et al. (1982)

w . Ip, carbonate content, Ey, shearing resistance, Pp, f„ f,

The study which conducted both cone penetration testing and lab testing in marine calcareous sediments revealed that the insitu testing (cone penetration testing) can dramatically improve the soil classification. The tests also showed that the cone penetration test results from the cemented (carbonate) soils can be interpreted using the principles established for noncarbonate soils. Many examples are quoted in this study to show the irrportance of performing cone penetration testing when engineering strength parameters are needed for design.

Datta, et al. (1982)

Op, (j), crushing coefficient

This study of engineering behavior of carbonate soils of India reveals that the crushing of carbonate particles and cementation by carbonate materials are the two most dominating factors which influence the engineering behavior of carbonate soils.

McKown and Ladd (1982)

e-log p curves, calcium carbonate content

The following are some conclusions from the consolidation and leaching tests performed on undisturbed specimens from a deposit of Pierre shale located in Northeast Nebraska:(1) The results support that natural cementation can have a significant effect on the apparent maximum past

pressure.(2) Reduction of CaCOg due to leaching caused an increase in compressibility during recompression and a

lower measured apparent maximum past pressure.

Footnote: S - static tests; D - dynamic tests; D, - relative density; C.C. - cernent content; t . - curing period; - confining pressure; - deformation modulus;([)' - effective friction angle; - undrained shear strength; K - permeability; E - Young's modulus; G - shear modulus; - dry density: - volume change; -

unconfined com pressive strength; - con e resistance; f, - frictional resistance; f, - friction ratio; e - void ratiotoCD05

C D■DOQ.C

gQ.

■DCD Table A5 (continued)

(/)c/)

8■D

CD

3 .3"CD

CD■DOQ.C

ao3T3O

(DQ.

Reference

Rad andClough(1982)

Rad andTumay(1984)

Parameters Studied

C.C. M,, Ev, consolidation pressure, strength parameters, liquefaction, O.C.R.

C.C., D„ q„ f„ f„ o,, strength components

Conclusions from the Studies

(1) The response of cemented sands to load is a function of the level of cementation, relative density and confining pressure.

(2) Increasing the cement content increases the cohesion intercept, while it has little effect on the friction angle.

(3) Cyclic shear resistance curves for cemented sands have essentially the same form as for uncemented sand.

(4) Available models to predict pore-pressure build-up in pure sands under repeated loading may need to be modified for cemented sands.

This investigation provides an insight into the effect of cementation on the cone penetration resistance ofsands. The major conclusions from this study are:(1) Cementation has a pronounced effect on the cone penetration resistance of sand. Increasing the cement

content increases the tip resistance and the sleeve friction, while decreasing the friction ratio. This behavior is similar to that of the relative density on uncemented sands. This increase in tip resistance and sleeve friction is attributed to the increase in cohesion intercept in cemented sand.

(2) The correlation between the internal friction angle and the cone penetration resistance of cemented sands depends strongly on the cement content. Specimens with similar friction angles but different cement contents show higher tip resistances and sleeve frictions and lower friction ratios.

(3) The effect of cementation on the cone penetration resistance of sand is similar to that of relative density. Utilizing the available correlations for uncemented sands to estimate the relative density or internal friction angle of naturally deposits possibly cemented sands can be possibly misleading. Generally existing correlations would suggest values of relative density and internal frictional angle higher than those actually available for the cemented sand.

T3CD

(/)(/)

Footnote: S - static tests; D - dynamic tests; D, - relative density; C.C. - cem ent content; t,. - curing period; o, - confining pressure; M, - deformation modulus;4)' - effective friction angle; S„ - undrained shear strength; K - permeability; E - Young's modulus; G - shear modulus; - dty density; Ey - volume change; q„ -unconfined com pressive strength; q - con e resistance; f, - frictional resistance; f, - friction ratio; e - void ratio

toCO

C D■DOQ.C

gQ.

■DCD

C/)C/)

8■D

CD

3.3"CD

CD■DOQ.C

aO3"OO

CDQ.

■DCD

C / )C / )



Acar and Tahir (1986)

C.C., D„ q„, stiffness ratio

(1) The tarrping method used in specimen preparation scheme leads to higher unconfined strength than in the tapping scheme.

(2) The increase in dynamic shear modulus of artificially cemented specimens at low levels of cementation is due to an increase in stiffness coefficient.

(3) Cementation leads to a decrease in damping ratio at all levels of strain.(4) It is detennined that an increase in the degree of cementation leads to a rapid decay of modulus with

increasing strains. It is observed that this modulus decay is more dominant in specimens with high stiffness ratios.

(5) The relative increase in the stiffness coefficient with cementation could be expressed with stiffness ratio, R. This ratio is nonlinearfy related to both the degree of cementation and void ratio.

Avramidis and Saxena (1986)

Strength parameters, dilatancy, pore pressure, cyclic strength

(1 ) Cohesion, angle of internal friction increases with the increase in cement content.(2) For cemented sands, the peak strength is reached when the summation of all strength components

reaches its maximum whereas for uncemented sands, the peak strength is reached when the rate of dilatancy is maximum.

(3) Small amount of cement increases significantly the cyclic strength of uncemented sands. This cyclic strength increases with relative density and curing period.

(4) For cemented sands before an pore water pressure generation, the cementation bond has to break. This requires a certain number of loading cycles.

Chang(1986)

G, C.C., danping ratio (Could not locate the original paper.)

Mitchell and Stone (1986)

C.C., q„ (t) In this work, the use of reinforcements in cemented fill to reduce the cement usage is studied. It is found that strong cemented layers at typical spacings of about 3 m in a low cement content bulk fill can reinforce the fill and reduce the overall cement usage.

Li andMitchell(1987)

C.C., stress-strain relationships, alignment or orientations of mesh elements

This study investigated the role of mesh element reinforcements and the anchored reinforcements in increasing the strength and ductility of sandfills. The major conclusions from this study are: The reinforcements are effective in increasing the strength and ductility of the cemented sand fills. But the other type of reinforcements, the smooth and deformed failure reinforcements were not as effective.

Footnote: S - static tests; D - dynamic tests; D, - relative density; C.C. - cernent content; t - curing period; - confining pressure; M, - deformation modulus;(J)' - effective friction angle; - undrained shear strength; K - permeability; E - Young’s modulus; G - shear modulus; - dry density; - volume change; -unconfined com pressive strength; - con e resistance; f, - frictional resistance; - friction ratio; o - void ratio

toCO00

C D■DOQ.C

gQ.

■DCD

C/)C/)

8■D

3.3"CD

CD■DOQ.C

aO3"OO

CDQ.

■DCD

C / )C / )



Riccobono(1987)

C.C., stress-strain curves

(1 ) At low axial strain, the cementation results in an elastic behavior. Cementation also induces an apparent cohesion of approximately 120 kPa and a slight increase in internal friction angle.

(2) At large strains, the cementation between grains is completely destroyed and therefore the residual shearing resistance of the cemented sand approaches that of sand.

(3) As a result of coupled reinforcement and drainage effect, the cementation reduces the settlement of the reinforced soft soil by about 50%

O'Rourke and Crespo (1988)

Index properties uniaxial compressive strength, Brazil tensile strength, peak and residual strengths, S„, M,, principal stress- strain curves

The study which has focused on the geotechnical properties of volcanielastic formation has yielded the following conclusions:(1) The Brazil tensile strength is usually high and is 18 to 29% of the uniaxial compressive strength.(2) This formation exhibits brittle failure mode at low confining pressure with a transition to ductile at high

confining stresses.(3) Increasing degrees of saturation cause a shift from brittle to ductile failure at constant confining pressure.(4) Material properties such as tensile strength and fracture toughness play an important role in explaining

and evaluating slope failures in the cemented formations found in the Andes of Ecuador and Colorrtria.

Saxena, et al. (1988)

D„ C.C., t , o,., stress ratio, e, dynamic modulus, cyclic strength

This study is devoted to discussing the factors affecting liquefaction resistance and to investigate the correlationbetween the dynamic moduli and cyclic strength of cemented sands.(1) Small amount of cement increases significantly the cyclic strength of uncemented sands. The cyclic

strength and the pore pressure development curves are corresponding to cemented sands are similar to those for uncemented sands. The cyclic strength in cemented sands increases with relative density and curing periods.

(2) Nondimensional empirical relationships are developed for dynamic shear and Young's modulus and damping ratio.

(3) Damping ratio initially increase and then decrease as cemant content increases from zero percent to eight percent.

Footnote: S - static tests; D - dynamic tests; - relative density; C.C. - cem ent content; t - curing period; - confining pressure; - deformation modulus;- effective friction angle; - undrained shear strength; K - permeability; E - Young's modulus; G - shear modulus; % - dry density; - volume change; -

unconfined com pressive strength; - con e resistance; f, - frictional resistance; - friction ratio; e - void ratiotoCOCO

300

A .2 D ynam ic P rop erties { G m a x )

Figure A .l and A.2 presents the variation of m axim um shear m odulus versus confining stress for cem ented specim ens of relative densities 35 and 80 % respectively.

Monterey No. O S e n d D, » 255&o. 1000

«9 5 0 0

O

oc5 «00X</>

5 0

Acar and E l - T a h ir (1986)

100 100050 5 0 0

CO N FI N I NG P RE SS URE * flfeCkPfl)

Figure A .l: T he V ariation of Gmax Versus Confining Stress for Cem ented Specim ens


301

Monterey No. O Sond Of = 80?ta. lOOO " 1 6 .0 7 k N / m

Ccmentotcon %

C9 5 0 0v>=>

5 0

Acar an d E l - T a h ir (1986)

5 0 100 SOOO5 0 0CONFINING PR ESS UR E, «sr f k P o )

Figure A.2: T he V ariation of Gmax Versus Confining Stress for C em ented Specimens


302

A 3 C one Test R esu lts on O ther Sands

In th is section, various cone tes t results of the investigations presented in section2.3.3 are presented.

Table A .3: Test R esults - E id (1987)

Dr (^v 9c D.R.(%) (kPa) (ksc) (ksc)

1 64.2 100 19.19 1.24 652 64.1 100 19.55 1.22 653 61.4 100 18.89 1.45 654 65.1 100 19.01 1.22 655 60.0 200 35.08 2.07 656 60.9 100 15.62 0.94 427 62.7 200 25.77 2.16 428 61.3 100 11.13 0.48 349 60.3 200 15.63 0.82 34

10 23.0 100 5.02 0.27 3411 21.3 200 8.72 0.56 3412 24.0 100 7.00 1713 24.0 100 7.00 0.51 1714 24.0 100 5.62 1715 24.0 100 7.12 0.67 1716 24.0 200 12.32 0.98 1717 24.0 200 12.45 1.18 1718 24.0 200 12.56 1.11 1719 24.5 100 6.51 0.55 4220 24.5 100 2.64 0.32 4221 24.5 100 3.77 0.44 4222 24.6 200 11.17 1.03 4223 21.9 200 10.16 42


303

Table A.4: Test Results - Baldi (1981)

Dr Ov 9c fs D.R.(%) (kPa) (ksc) (ksc)

1 40.4 137.8 13.6 &L8 432 50.6 231.6 20.5 61.3 433 52.2 324.5 25.5 54.6 434 50.6 116.0 9.3 127.7 435 5 1 2 65.0 5.7 142.2 436 44.5 118.0 7.1 9&5 437 47.0 2&L5 15.7 46.7 438 45.4 317.0 10.6 53.0 439 39.3 66.0 4.6 113.7 4310 42.1 729.0 20.3 42.6 4311 71.2 317.0 26.6 135.4 4312 69.4 116.0 15.9 224.2 4313 71.2 524.0 35.1 106.4 4314 71.6 730.0 41.5 89.5 4315 67.5 67.2 11.1 271.4 4316 69.3 118.0 13.8 191.6 4317 74.0 117.0 12.3 171.5 4318 75.2 319.0 22.5 11&8 4319 80.6 519.0 32.3 97.6 4320 72.5 69.0 8.9 213.3 4321 74.2 729.0 37.8 82.9 4322 9&9 525.0 47.4 145.6 4323 93.6 320.0 38.9 199.6 4324 93.6 118.0 24.3 3TA2 4325 93.6 319.0 36.9 190.6 4326 87.6 67.0 18.8 45^9 4327 96.6 522.0 44.5 137.1 4328 96.5 124.0 2L3 284.5 4329 90.5 70.0 12.2 298.2 4330 8R9 318.0 34.3 170.5 43


304

Table A.5: Test Results - V illet and M itchell (1981)

Dr o-v 9c D.R.

(%) (kPa) (M Pa) (kPa)1 29 296 13.8 6R9 202 24 136 6.5 70.6 203 31 408 18.5 67.0 204 54 296 23.1 116.0 205 63 250 24.0 109.7 206 53 69 8.5 122.1 207 52 68 8.9 129.8 20


305

Table A.6: Test Results - Harman (1976)

D r f s D.R.

(%) (kPa) (MPa) (kPa)1 33.9 59 2.1 53.5 342 37.2 272 7.6 41.4 343 38.0 263 7.5 42.2 344 34.5 61 1.9 519 345 22.0 61 1.5 37.1 346 20.0 61 1.5 312 347 31.3 61 2.4 61.1 348 35.5 51 1.6 49.7 349 31.5 61 1.9 49.7 3410 36.1 61 2.3 57.0 3411 30.1 61 1.8 46.2 3412 36.0 272 6.8 314 3413 40.6 272 8.1 410 3414 47.3 271 1&3 109.0 3415 30.2 272 6.8 316 3416 2&8 272 5.7 318 3417 27.5 272 5.8 31.6 3418 6&6 271 18.9 113.5 3419 61.1 60 6.3 170.8 3420 64.6 271 216 1314 3421 60.2 60 6.0 165.6 3422 6&6 271 20.7 128.5 3423 54.9 60 7.2 197.6 3424 74.7 60 10.4 278.6 3425 86T 270 23.5 140.5 3426 76.5 271 210 169.4 3427 8&4 60 10.2 271.6 3428 7&5 52 8.5 2B17 3429 75.4 60 10.2 2812 3430 77.9 52 8.7 270.8 3431 79.1 52 7.1 224.3 3432 7&3 60 9.4 256.7 3433 81.0 271 41.7 :%i7 3434 8&4 271 24.4 141.7 3435 76.4 60 15.0 423.5 34


306

Table A .7: Test R esults - F ioravante (1992)

Dr (Tv 9c D.R.

(%) (kPa) (M Pa) (kPa)1 41.4 111 6.0 8L 2 602 41.9 61 4.3 106.8 603 56.5 61 8.6 221.2 604 56.7 61 9.5 229.0 605 57.0 61 8.2 207.8 3&66 56.7 60 7.5 159.8 3&67 62.1 111 15.2 173.9 33.68 63.4 111 19.4 190.3 33.69 62.6 111 12.9 181.5 33.610 6&9 111 1&3 216.3 6011 61.0 111 15.2 202.4 6012 56.7 111 13.0 182.4 6013 74.5 111 19.2 223.3 3R 614 74.5 111 16.2 224.0 33.615 74.7 111 20.0 208.0 33.616 86.0 111 1&3 260.4 33.617 8&6 111 2&5 249.0 33.618 8 4 J 111 2 4 ^ 227.4 33.619 8 4 J 111 2&8 339.0 6020 8&6 112 26.5 304.4 6021 8&2 112 3&9 343.2 6022 84.1 112 27.1 2 5 4 J 33.623 8&3 113 29.0 253.4 33.624 8&8 112 2TA 339.5 6025 91.1 114 40.4 360.5 60


307

Table A.8: Test Results - Manassero (1992)

Dr 9c !s D.R.(%) (kPa) (M Pa) (kPa)

1 50 116.7 6.8 91.9 34.32 75 101.0 13.3 191.7 34.33 86 495.4 41.9 131.5 34.34 89 113.8 18.1 265.9 34.35 89 112.8 25.6 239.9 34.3

Table A .9: Test Results - N u tt and Houlsby (1992)

Dr dv 9c A D.R.(%) (kPa) (M Pa) (kPa)

1 6.7 4&6 2 j # 45.0 282 2&4 5&2 0.74 18.0 283 16.5 148.8 3.48 3AA 284 2&4 5R2 2jG 7&9 285 23.0 23.0 2 j j 62T 286 23.9 3&6 1.29 3&8 287 22.1 9R6 3.99 3&6 288 2&7 2&2 2jW 56.6 289 28.1 60.8 47.0 2810 31.0 148.9 5.16 3&8 2811 27.5 5R2 1.49 3&2 2812 2&4 149.6 3j& &L9 2813 32.7 39.4 2.07 5&6 2814 29.0 148.9 3^1 37.2 28


308

Table A. 10: Test Results - Lhuer (1976)

Dr CTv Qc fs D.R.(%) (kPa) (M Pa) (kPa)

1 34 59.90 2.0 34.22 34 61.90 1.9 44.0 34.23 82 60.90 10.1 23&3 34.24 75 60.90 10.1 232.8 34.25 37 276.10 7.5 3TA 34.26 77 275.18 28.0 141.6 34.27 78 275.17 2&9 146.2 34.28 82 275.17 30.3 153.1 34.29 36 51.76 1.8 4&6 34.210 78 52.76 8.5 224.8 34.211 78 52.76 8.7 230.1 34.212 38 266.96 7.5 3&6 &T213 79 28R31 30.4 146.8 34.214 80 295.45 30.2 142.2 34.215 26 62.26 1.6 3&0 34^16 24 62.47 1.8 41.1 34^17 29 276.48 6.5 3L9 34^18 27 276.04 6.9 34.3 3A219 25 27R54 4.1 19.58 34^20 79 61.26 7.9 179.77 34^21 81 275.46 24j 122.66 34^22 79 61.04 7.8 179.03 34^23 81 275.67 2&8 130.19 3A224 27 67.83 1.7 34.77 34^25 28 67.90 1.9 39.87 34^26 30 267.82 6.1 30.92 34^27 29 267.82 6.3 32.02 34^28 81 7&39 7.6 147.12 34.229 79 67.25 9.8 203.88 34^30 81 267.03 25.5 13&86 34^31 81 267.03 2&6 117.64 34^


Appendix B

B .l CU Tests

The results of consolidation undrained triax ia l tests are presented in this section. Cem ented results (C.C. 2 %) are reported in th is section.

2000MS-C2-UNDR50

BGGoo 100 kPa t>t>> 200 kPa

300 kPo1500

œMK

1000u

500

0 -tfr

S 150

[/]œ -150 K -300 ^ -450S -600

Figure B .l: Undrained Triaxial Test on C em ented M onterey No. 0/30 Sand (Dr 50 %; C.C. 2 % )

300


310

cdDh

mmwKHœ

o*—4KO

<>wQ

2 0 0 0MS-C2-UNDR65

1500

1000

oB-BDo 1 00 kPa 200 kPa 300 kPa

500

00 8 124 16

AXIAL STRAIN { %)

cCDh

150 -W K

0U1œ -150wK -300 ^ -450

-600

S— G

i - i r I I . . I I I I I I I I I I I I I I I r 1 I r

O 0CL,

4 8 12 16AXIAL STRAIN (%)

Figure B.2: U ndrained Triaxial Test on C em ented M onterey No. 0 /30 Sand {Dr 65 %; C.C . 2 %)


311

3000M S-C2-UNDR85

oooDo 100 kPa 200 kPo

U1

M 2000

1000

0 84 12 16

AXIAL STR A IN, Si(%)

CÜ

4 8 12AXIAL STRAIN (%)

Figure B.3: U ndrained Triaxial Test on Cem ented M onterey No. 0/30 Sand {Dr 85 %; C.C. 2 %)


312

B .2 Stress P aths

Stress pa th s of the undrained test results are shown in th e figures in th is section.

1500

T.S.P.E.S.P. MS-C0-UNDR45

eoooB 100 kPa 200 kPa

Go o o o 300 kPo1000

O h

500

500 1000(kPa)

1500r

Figure B.4; Stress P a th s of CU Test {Dr = 45 %; C.C. 0 % )


313

1500

T.S.P. M S -C 1 -U N D R 4 5

A A A A A 2 0 0 k P oM-M-0 300 kPa1000

aPh

500

550 1 100

p' ( k P a )

Figure B.5; Stress P a ths of CU Test {Dr = 45 %; C.C . 1 % )


314

1 5 0 0

T .S .P . E .S .P .

M S -C 2 -U N D R 4 5

— 1 0 0 k P a iSr-A A A A 2 0 0 k P a1000

cr5 0 0

6 0 0 1200( k P a )f

F igure B.6: Stress P a ths of CU Test {Dr = 45 %; C.C. 2 % )


315

1500

T.S.P.E.S.P. MS-C0-UNDR65

Qoooo 1 00 kPo Ù.-ÙS-Ù.-Ü A 2 0 0 kPa o o o e o 300 kPo

1000

500

500 1000(kPa)

1500

Figure B.7: Stress P a th s of CU Test {D^ = 65 %; C.C. 0 % )


316

1500

T.S.P. M S -C 1 -U N D R 6 5

0 0 -0 0 0 1 00 kPa AAAiWi 200 kPa1000

c r

500

600 1200(kPa)r

Figure B.8: Stress P aths of CU Test {Dr = 65 %; C.C. 1 % )


317

1500

T .S .P . M S -C 2 -U N D R 6 5

08000 100 kPaI3BBBQ 200 kPa AAAAA 300 kPo000

500

1200600p, p' (kPa)

Figure B.9: Stress Paths of CU Test {Dr — 65 %; C.C. 2 % )


318

B .3 F low C harts and L istings

T he flow chart and listing of the three program s used in th e chap ter 4 are presented in th is section. T he first program reads the drained d a ta and calculates stress ra tio and dilation angle. T he second program sim ulates the drained triax ia l behavior. T he th ird program sim ulates undrained triax ial behavior.

F lo w C h a rt: P r o g r a m 1

START

END

Calculate

d e j , de j, dy"

Output Param eters

Input P aram eters from Drained S et

n ' 0' M

7 = E, - e,

cr^+2a,

Figure B.IO: Flow C hart of Program 1


319

L istin g O f P rogram 1

C Program 1 (Anand Puppala)

DIMENSION EPS1(20),RATSIG(20),EPSV(20),SIG1(20),Q(20),PE(20),& GAMA(20),EPS3(20),A1(20),A2(20),A3(20),& DEPSV(20),DGAMA(20),DPE(20),DQ(20),DEPSVE(20),DGAMAE(20), & DEPSVP(20),DGAMAP(20),A4(20),A5(20)

OPEN(8,FILE='GEOIP.OUT \ STATUS='NEW')WRITE(8,15)15 FORMAT(5X,'THE FOLOWING ARE DRAINED TEST RESULTS',

& //,11X,'I',4X,'SIG1(I)',9X,'Q(I)',11X,'PE(I)',10X,& 'GAMA(I)',8X,/)

epsl(l)=0.5 epsl(2)=l. epsl(3)=1.75 epsl(4)=3.5 epsl(5)=5.25 epsl(6)=7.5 epsl(7)=10.5 epsl(8)=13.5

EPSV(1)=0.2EPSV(2)=0.22EPSV(3)=0.3EPSV(4)=0.25EPSV(5)=0.06EPSV(6)=-0.09EPSV(7)=-0.45EPSV(8)=-0.57B

RATSIG(1)=2.25 RATSIG(2)=2.88 RATSIG(3)=3.03 RATSIG(4)=3.34 RATSIG(5)=3.43 RATSIG(6)=3.50 RATSIG(7)=3.57 RATSIG(8)=3.57

SIG0=100.

DO 10 1=1,8


320

SIGl(I)=RATSIG(I)*SIGOQ(I)=SIG1(I)-SIG0PE(I)=(SIGl(I)+2*SIG0)/3EPS3(I)=(EPSV(I)-EPSl(I))/2GAMA(I)=EPS1(I)-EPS3(I)

A1(I)=Q(I)/PE(I)A2(I)=GAMA(I)/A1(I)A3(I)=EPS3(I)/EPS1(I)10 WRITE(8,*)I,SIG1(I),Q(I),PE(I),GAMA(I)

HRITE(8,45)45 F0RMAT(//,3X,'EPS3(I)',10X,'Q/PE',8X,'GAMA/(Q/PE)',4X,

& 'EPS3/EPS1',/)

DO 40 1=1,840 HRITE(8,*)EPS3(I),A1(I),A2(I),A3(I)

AM=.25G=400E=1000

WRITE(8,55)55 FORMAT(//,2X,'DEPSVP(I)',9X,'DGAMAP(I)',7X,'(DQ/DPE)',4X,

& 'DEPSVP/DGAMAP',/)

DO 50 1=1,8IFd.EQ.DTHENDEPSV(I)=EPSV(1)DGAMA(I)=GAMA(1)DPE(I)=PE(1)-SIG0DQ(I)=Q(1)ELSEDEPSV(I)=EPSV(I)-EPSV(I~1)DGAMACI)=GAMA(I)-GAMA(I-1)DPE(I)=PE(I)-PE(I-1)DQ(I)=Q(I)-Q(I-1)ENDIFDEPSVE(I)=((1-2*AM)/E)*3*DPE(I)DGAMAE(I)=DQ(I)/(2*G)DEPSVP(I)=DEPSV(I)-DEPSVE(I)DGAMAP(I)=DGAMA(I)-DGAMAE(I)A4(I)=DEPSVP(I)/DGAMAP(I)A5(I)=DQ(I)/DPE(I)


321

50 WRITE(8,*)DEPSVP(I),DGAMAP(I),A5(I),A4(I)STOPEND


322

Flow Chart: Program 2

( START ^

No

7 g 0.15No

Yes

Do Results com pare with rexperimants? No

Yes,

END

Calculate h(7)

Calculate q, p'

0.005, Y=TT+ d')'

OUTPUT MODEL PARAMETERS

Compute dc. and c,= c + d e .

Output Param eters

Assume Different

Input Model Param etersMl I My p , My g y ,

7 = 0.0

Figure B . l l : Flow C hart of Program 2


323

L istin g o f P ro g ra m 2

C PROGRAM 2 DRAINED SIMULATION (Anand Puppala)

DIMENSION EPS1(100),SIG(100),EPSV(100),SIG1(100),0(100)& ,PE(100),GAMA(100),EPS3(100),H(100),D(100),AA1(100),aa2(100),& DEPSV(IOO),DPE(100),DQ(100),DEPSVE(100),DGAMAE(100),& DEPSVP(IOO).DGAMAP(100),A1(100),DEPS1(110),DEPS3(100),& GGAMA(IOO).EEPSV(IOO),QQ(100),EEPS1(100)

REAL MU,MU1,MU2,MPHI,MPHICVOPEN(8,FILE='yl80.dat',STATUS='NEW')v = 0 .3C E=2.0*G*(1.0+v)PHI=(38.0/180)*3.143 PHICV=(35.4/180)*3.143 CP=6.2 CR=7.50MPHI=(6*SIN(PHI))/(3-SIN(PHI))MPHICV=(6*SIN(PHICV))/(3-SIN(PHICV))C MPHI=1.55C MPHICV=1.4MU1=1.5 MU2=0.54 P1=CP/TAN(PHI)SIG0=100.0 G= 157.1 * SigO E=2*G*(l+v)DGAMA=.005RAT=1+SQRT(1-MPHICV/MPHI)A=-(2*SIG0*(MPHI*RAT)**2/(MPHICV*G))B=SIGO*MPHI*RAT/GC=MPHICVGAMA(1)=0H(1)=0q ( i )= oQQ(1)=0SIG1(1)=SIG0PE(1)=SIG0EPSV(1)=0EPS1(1)=0EPS3(1)=0D(1)=0A1(1)=0DO 10 1=2,50


324

GAMA(I)=GAMA(I-1)+DGAMAH(I)=C*GAMA(I)*(GAMA(I)-A)/((GAMA(I)+B)**2)IF(H(I).LE.MPHICV)THEW MU=MU1C P1=CP*MPHI/TAN(PHI)ELSEMU=MU2C P1=CR*MPHIC/TAN(PHICV)ENDIFQ(I)=3*SIG0*H(I)/(3-H(D)PE(I)=3*SIG0/(3-H(I))PE(I)=PE(I)-P1QQ(I)=Q(I)/PE(I)SIG1(I)=Q(I)+SIG0DQ(I)=q(I) -Q(I- l )DPE(I)=PE(I)-PE(I-1)DGAMAE(I)=DQ(I)/(2*G)DGAMAP(I)=DGAMA-DGAMAE(I)DEPSVE(I)=((l-2*v)/E)*3*DPE(I)C WRITE(*,*)'HI',DGAMAP(I),DGAMAE(I),DPE(I),'E',EDEPSVP(I)=DGAMAP(I)*(MPHICV-H(I))/MU C WRITE(*,*)'HI2',DEPSVE(I),DEPSVP(I)DEPSV(I)=DEPSVE(I)+DEPSVP(I)C WRITE(*,*)DEPSV(I)EPSV(I)=EPSV(I-1)+DEPSV(I)DEPS1(I)=(2*DGAMA+DEPSV(I))/3-DEPS3tI)=<DEPSVW=D6AMA>/3-----------------------------EPS1(I)=EPS1(I-1)+DEPS1(I)EEPS1(I)=EPS1(I)*100.0EPS3(I)=EPS3(I-1)+DEPS3(I)DCI)=DEPSVP(I)/DGAMAP(I)Ai(I)=Q(I)/PE(I)GGAMA(I)=GAMA(I)*100.0 EEPSV(I)=EPSV(I)*100.0 C WRITE(8,*)GAMA(I),Al(I),EPSV(I)10 CONTINUEDO 20 K=l,5020 WRITE(8,*)GGAMA(K),QQ(K),EEPSV(K)C20 HRITE(8,*)GGAMA(K),A1(K),EEPSV(K)020 WRITE(8,*)H(K),MPHICV,MU,EPSV(K)STOPEND


Flow Chart; Program 3

325

START 3

No

Yes

7 ë 0,15No

Yes,

No

Yes,

END

Calculate h(K)

(ty= 0.005, Y=7+ dy

OUTPUT MODEL PARAMETERS

Calculate q, p, p' Using e,= 0

Compute O ', , o .Pore Pressure U = tr ,- cr.

Assume Différent

Output Parameters

p .r .e .

Input Model ParametersGi Ml t Mz' ^ ,p '

Figure B.12: Flow C hart of P rogram 3


326

L isting o f P rogram 3

C PROGRAM 3 UNDRAINED SIMULATION (Anand Puppala)

C UNDRAINED SIMULATION PROGRAM 3DIMENSION EPSl(lOO),SIG1E(100),SIG3E(100),SIG1(100),Q(100),

& PE(IOO),GAMA(100),EPS3(100),H(100),D(100),DH(100),T(100),& SE(IOO),S(100),DPE(100),DQ(100),DEPSVE(100),DGAMAE(100),P(100), & DEPSVP(IOO),DGAMAP(100),A1(100),DEPS1(110),DEPS3(100),U(100) ,& RGAMA(100),AA1(100),REPS1(100),RQ(100),RAT0(100)

REAL MU,MU1,MU2,MPHI,MPHICVOPEN(8,FILE='U185.DAT',STATUS='NEW 0SIG0=100.0G= 193.0 * SIGOv=0.3E=2*G*(l+v)MPHI=1.75 MPHICV=i.52 MU1=0.55 MU2=0.96 DGAMA=.005RAT=1+SQRT(1-MPHICV/MPHI)A=-(2*SIG0*(MPHI*RAT)**2/(MPHICV*G))B=SIGO*MPHI**RAT/GC=MPHICVGAMA(1)=0RGAMA(1)=0.0REPS1(1)=0.0H(1)=0.0Q (1)=0.0RQ(1)=0.0RAT0(1)=0.0SIG1(1)=SIG0PE(1)=SIG0S(1)=SIG0SE(1)=SIG0U(1)=0.0T(1)=0.0EPS1(1)=0EPS3(1)=0DCl)=0Ai(l)=0DO 40 1=2,90GAMA(I)=GAMA(I-1)+DGAMA


327

H(I)=C*GAMA(I)*(GAMA(I)-A)/((GAMA(I)+B)**2)IF(H(I).LE.MPHICV)THENMU=MU1ELSEMU=MU2ENDIFDH(I)=H(I)-H(I-1)DPE(I)=E*(2*G*DGAMA-PE(I-1)*DH(I))*(MPHICV-H(I))/

& (6*G*MU*(2*v-l)+E*H(I)*(MPHICV-H(I))) IF(DPECI).LE.0.0)G0T015 GOTO 2515 DPE(I)=0.0C WRITE(8,*)i,'DP',DPE(i),'DH',Dh(i)25 DQ(I)=H(I)*DPE(I)+PE(I-1)*DH(I)C WRITE(8,*)i,DQ(I),'Dq HERE'Q(I)=Q(I-1)+DQ(I)PE(I)=PE(I-1)+DPE(I)RATO(I)=Q(I)/PE(I)SIGlE(I)=PE(I)+(2*Q(I)/3)SIG3E(I)=PE(I)-Q(I)/3U(I)=SIG0-SIG3E(I)SIG1(I)=SIG1E(I)+U(I)DGAMAE(I)=DQ(I)/(2*G)DGAMAP(I)=DGAMA-DGAMAE(I)DEPSl(I)=2*DGAMA/3DEPS3(I)=-DGAMA/3EPSl(I)=EPSl(I-i)+DEPSl(I)EPS3(I)=EPS3(I-1)+DEPS3CI)REPS1(1)=EPS1(I)*100.0S(I)=(SIGl(I)+SIG0)/2SE(I)=(SIGlE(I)+SIG3E(I))/2T(I)=(SIGl(I)-SIG0)/2P(I)=PE(I)+U(I)A1(I)=Q(I)/PE(I)RGAMA(I)=GAMA(I)*100.0 RQ(I)=Q(I)*2.0C HRITE(8,*)RGAMA(I),A1(I),U(I)40 CONTINUEC T(2)=0.DO 50 1=1,9050 HRITE(8,*)RGAMA(I),A1(I),U(I)C HRITE(8,45)C45 FORMAT(//,5X,'EPS1',12X,'U',14X,'S',14X,C & 'SE',14X,'T',/)


328

STOPEND


Appendix C

C .l Cone Test R esu lts

The MQSC test results conducted on both cemented and uncem ented specim ens are presented in th is section.

329


330

CN

OoCDOo oro

w

O CM

eio

Ü

oo'o

oorO

: CMTf

- 0- •4J: oCM a ”: O

II Q03;

Ï 1 1 1 1 1 1 ] 1 1 1 1 1 1 11 1 1 1 1 ; 1 1 1 1 1 1 1 1 1 1 1 1OCvJ

O oCD O00

bC o^ CNJ

u O jÿ O

-

-q

- CD- 11

- O'

1 1 1 1 1 1 1 1 1 .............til 1 1 1 1 1 1 1 1 IOCM OM" OCD OOO

( u i o )

F igure C .l: Cone P enetra tion Test Results on a Specim en [Dr = 48.7 %; e„ 0.85; Cmin = 0.56 and 100 kPa)


331

CN

OOCN

ON- OCO

o03

BO CN

bO

j w

;

indII

:

I 1 1 1 1 t 1 j t 1 1 t 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

o oN- OCD

O03O

O<-

N

o Oo cr

OOCN

ON- OCD

OOO

( U I o ) m d 9 Q

Figure C.2: Cone P enetra tion Test Results on a Specimen {Dr = 56.4 %; e„ 0.85; e-min = 0.56 and 200 kPa)


332

CN

OOCN

O OCD

O

CMCM

O oCN

ON- OCD

OO

en

C TO

O oCN

ON" OCD

OOO

( m o )

Figure C.3: Cone P ene tra tion Test Results on a Specim en [D-r = 54.8 %; =0.85; e-min = 0.56 and 300 kPa)


333

fO

CN

o I I _ _ _ _ I_ _ _ _ I_ _ _ _ I— !_ _ _ I_ _ _ I— _ _ _ I_ _ _ _ I_ _ _ I_ _ _ _ I_ _ _ _ I_ _ _ I_ _ _ _ I_ _ _ _ I_ _ _ _ I_ _ _ _ _ _ _ _ I I _ _ _ _ I _ _ _ _ I I_ _ _ _ I_ _ _ _ I I I

OCN (Û

na ua m

oo ooo

O '

o o o oCN CD

(mo) '\d9Q

Figure C.4: Cone P enetration Test R esults on a Specim en {Dr = 90.0 %; e„ 0.85; Cmin — 0.56 and 100 kPa)


rO

C\

-

-

-

.

/ 1 1 1 1 1 1 _1_ I 1 t I 1 1 1 t 1 1 J I 1 1 1 - .1. . . J— 1 . 1 . I .

toOCvJ

O oCD

CM

O

be CDd

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I ] I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I I I 1 I I I I I I I

CVJ

g da m

oo'j-

CM O

S Ru

oCM

OTf- OCD

W) O

o o çÿ O

o

-

oDO

CM

-

c r

XT 1 1 ] 1 1 1 1 1 1 j 1 r 1 1 1 1 1 1 1 1 1 1 1 1 1 1

oCM

oCD

( u i o ) q ^ id o Q

334

Figure C.5: Cone P enetra tion Test Results on a Specim en [Dr = 90.0 %; e„ 0.85; e-în — 0.56 and 200 kPa)


335

5rO

CM

Q I I T I I I I I I I I I I I I I I ] [ I I I I I I I I I I 1 I I I

o o o o oCMrO

.«UQ C P

B

o o oCM

on oo

o

cr

o o o o o

(ino) q:jd0 a

Figure C.6: Cone P enetra tion Test Results on a Specimen {Dr = 88.0 %; e„ 0.85; Cmin = 0.56 and 300 kPa)


336

CN

OCD

OO O

O oCM

O

Su

m

OO O

CDOO O

OC M

ON-OO

CM

(Xq H o

O OCN

ON" OCD

O00

( u i o ) q:jd0a

Figure C.7: Cone P enetra tion Test R esults on a Specim en {Dr = 86.0 %; 0.85; Cmin = 0.56 and 100 kPa)


337

ooCM

O O

N

CN

OO O O

CNOO

OCN

C T

^ O

o o oCN

On

( u i o )

F igure C.8: Cone P enetra tion Test R esults on a Specim en {Dr = 84.0 %; e„ 0.85; Cmin = 0.56 and 100 kPa)


338

C\'

k4M

-

-

-

I 1 I I 1 1 1 I 11

1 . . ! 1 t I t 1 1 I { \ { \ \ \ \ \ \ . . . I t 1 J . . . 1 l i l t

oCNJ

o-d-

OCD

OOO

so CM

tiB-iij

4h°

L DO)d

CV)

o

II k ”

a m

. L _ L _ L . . l . . i , 1 J _ 1 1 1 1 t 1 { 1 < 1 1 1 1 1 1 I 1 1 1 1 1 1 1 1 1 1 1 1 1

OO

N O

B °Ü

oCM

O oCD

O00

bfl S^ CM

c r §

-

C V J

m0 0CD

-11

Uc r

—

1 1 I 1 i 1 1 1 1 . - 1 1 i - J 1 J . 1 1 ,.L 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I

oCM

OTj-

oCD

OOO

( u i o )

Figure C.9: Cone Penetration Test Results on a Specimen {Dr = 86.0 %; C.C. = 1 %; Êmax = 0.85; tmin = 0.56 and 100 kPa)

R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.

339

oo o o o

CM CD

O fN

OO O

CMO OO

w o

CD

c r

o oCM

OM-

OCD

( m o )

Figure C.IO: Cone Penetration Test Results on a Specimen {Dr = 84.6 %; C.C. = 1 %; Cmai = 0.85; Cmin = 0.56 ctnd 200 kPa)


3 4 0

CN

O OCD

O OCN

en

oo o

CNoo

o

cr

oo o

CMoN-

( i i i o ) q : j d a a

Figure C .ll: Cone Penetration Test Results on a Specimen {Dr — 89.2 %; C.C. = 1 %; ^max = 0.85; e^.n = 0.56 and 300 kPa)


rO

341

CN

1 I T I I I I I I I I I I I I I I I I I I I ] r I I I I I I I I I I I I I I

rOOCN

ON-

OCD

OOO

S ô CN

tao oCN

O I T 1 - 1 I I - 1 - 1 1 - 1 I I I I I I I I I I I 1 I I I I I I I I I ............................

N'tl*

CQK 00

C0 W -

Qoa

CN CO

M

CO

O-

o oCN

O oCD

O

( m o ) q'i^doQ

Figure C.12: Cone Penetration Test Results on a Specimen {Dr = 48.8 %; C.C. = 1 %; Cmai = 0.85; Cmin = 0.56 and 100 kPa)


3 4 2

q-4

r )

CN

-

-

- / \

1 1 1 1 1 i 1 1 1 1 1 ( 1 1 1 1 1 1 ................................. l i t

h — \( ( 1 1 > 1 I [ 1

rO

o rN

oCN

o

tafi,h{

oooN-

W O

BOCO

oooCN

u Ocr O

I I I I I I I I I I I I I I r I I I I I I I I I I I 1 I I I r r I I I I I I I

CN . 2 3d-----6

eSUQOQ

OCN

Or)

oN-

-

- “CT)

- II■ u- c r

1 1 1 1 1 1 1 . . . . . . . . . . . . . . . . . . . . . . . . . ... .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

oCN

oCO

oN-

( u i o ) q i d a q

Figure C.13: Cone Penetration Test Results on a Specimen {Dr = 46.6 %; C.C. = 1 %; Cmax = 0.85; Cmin = 0.56 and 200 kPa)


3 4 3

r )

CN

-

ki 1 1 1 1 1 1 1 i 1 1 1 I 1 ) > 1 I 1 _ 1 . J ! 1 1 t ( J

COoCN

oN-

oCD

N

o (N

ÎM)

«t-f

:

N"

Nrf<

: . ^ . J

d

I I

c n

01 3

rrt r 'I

- J \ rQGQ

r 1 1 1 1 1 1 1 1 1 1 r 1 1 J 1 1 1 1 1 t 1 i > t 1 1

ooN-

OCN

OTf-

oCD

w oB otoÜ

ob f l o

CN

u Ocr O

-

CDrO

_ O-

- IIu

- cr

/VI J— I_J J 1 1 1 1 1 < 1 t 1 1 1 t 1 1 1 t 1 1 1 1O(N O

N"OCD

( u i o ) q : } d a a

Figure C.14: Cone Penetration Test Results on a Specimen {Dr = 53.2 %; C.C. = 1 %; Gmax = 0.85; Cmin — 0.56 and 300 kPa)


lO

CN

OO O OC N CD

CM

O

bo

O '— I— I— 1—1— I— I— I— 1— — 1—1_I I I I I I I I I I I I I I I L

CM

n

8JU5 m

ooN-

OorO

OOC N

OC N

ON-

OCD

-CDcsiO04

IIo

Ï' 1 ) 1 1 i 1 1 i 1 — L _ j __ . ,i I I 1 1 1 1 1 1 1 1 1 1

OC N

ON"

OCD

( u i o ) q q d o Q

344

Figure C.15: Cone Penetration Test Results on a Specimen {Dr = 88.2 %; C.C. = 2 %; Cmor = 0.85; Cmin = 0.56 aud 100 kPa)


345

CN

O OCN

O O O

M

So

bX)

o o oN-oo

N-

N

CT

ooC N

OT-O

oN-

( m o )

Figure C.16: Cone Penetration Test Results on a Specimen — 86.3 %; C.C. = 2 %; Cmox = 0.85; Cmtn = 0.56 and 200 kPa)


346

c\

o oCM

OCD

OM"

O

CN

O OCN

ON-

OCD

OOOO

N Q .

cr^ O

O O o o oCN

( n i D ) q ; d a ( ]CD OO

Figure C.17: Cône Penetration Test Results on a Specimen {D^ = 84.2 %; C.C. = 2 %; ^max = 0.85; ejnin = 0.56 and 300 kPa)


CN

Oo o o o o

n CN CD OO

N

o CN

bfl

4-1

N"d

o . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I I I J I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I

NN*

COCCCD

-CO-CJ -

o m

ooN-

OCN

ON" OCD

OOO

CM od oG nÜ

b fooCN

u ocr o

-

-d0 0

-

II

CT

k I 1 1 1 1 1 1 1 . 1 1 1 1-i . 1 1 1 1 J 1 1 1 t 1 1 t t 1 1 1 1 1 : • 1 1

o oCN

oN-

oCD

o00

( m o )

34 7

Figure C.18: Cone Penetration Test Results on a Specimen {Dr = 47.2 %; C.C. = 2 %; Cmar = 0.85; Cmin = 0.56 and 100 kPa)


348

M-)

CN

-

-

-

/ '1 1 1 t 1 t 1 1 1

L— '" S '— -

I 1 1 1 1 1 1 1 1 1 1 1 ! I i i 1 1 1 1 1 1 1 1 ( 1 1

CD 00

Ü CN

Oo o o

CDO

o

oCN

O oCN

ON-

OCD

O

( m o ) q^jdoQ

Figure C.19: Cone Penetration Test Results on a Specimen {Dr — 54.4 %; C.C. — 2% ; ^max = 0.85; Cmin = 0.56 and 200 kPa)


349

CN

U

o oro

O OCN

ON-

c

N

Sa

m

OO O O

CNO O

N-C

O

Oo

bB S^ CN D-

o

o o oCN

O o c

( m o ) m d a q

Figure C.20: Cone Penetration Test Results on a Specimen {D^ = 52.0 %; C.C. = 2 %; Gmax = 0.85; Cmin = 0.56 and 300 kPa)


Appendix D

D .l Program Listings for C avity E xpansion M odels

350


351

The listing of the cavity expansion simulation program is presented in this appendix. This is based on the closed form solutions proposed by Carter et. ai., 1986.

Program Listing

C Cavity Expansion Simulation (Anand Puppala)C Closed Form Solutions Carter et al., 1986C************************************************************** DIMENSION PL(SOO),ERR(600),RTERR(600),PL1(600) ,QC(600)REAL K,M,N,LEFT1,NU,K0 0PEN(8,FILE='z',STATUS='NEW')C Shear modulus from previous modelG=370000.0C Soil propertiesWRITE(*,*)'Cohesion ?'READ(*,*)C C C= 0.0WRITEC*,*)'PHI(PEAK) ? 'READ(*,*)PHII WRITE(*,*)'PHI(RES) ? 'READ(*,*)PHICIPHI=(PHII*3.1428/180.0)PHIC=(PHICI*3.1428/180.0)PC=(3.1428/4)+(PHIC/2)PC1=(3.1428/2)-PHICC MPHI=38.9C DIL=-33.0C Spherical cavityWRITE(*,*)'K = ? (1 FOR SPH OR 2 FOR CYL)'READ(*,*)K C K=1.0C Coeff. of earth pressure at restWRITE(*,*)'KO = ?'READ (*,»=) KO C K0=0.43C Initial confining pressureWRITE(*,*)'SIGO = ?>

READ(*,*)SIGOWRITE(*,*)'G/SIGO'READ(*,*)XXG=XX*SIGOC SIG0=200.0C Dilational angle


3 5 2

WRITE(*,*)'DIL ANGLE = ?'READ(*,*)DILA DIL=(DILA*3.14/180.0)C Octahedral stresses cind poisson ratioP 0 = ((1 + 2 * K 0 )/3 )* S IG 0NU=0.3C Initial limiting pressurePL(1)=20000.0 C EquationsM=(1+SIN(DIL))/(1-SIN(DID)N=(1+SIN(PHI))/(l-SIN(PHI))ALPHA=K/MBETA=1-K*((N-1)/N)GAMA=(1+ALPHA)/(1-BETA)SIGR=((1+K)/(N+K))*N*P0UP1=K*(1-NU)-K*NU*(M+N)+((K-2)*NU+1)*M*NDOWN1=((K-1)*NU+1)*M*NJAY=UP1/DDWN1Z=(K+1)JAY/(ALPHA+BETA)T=(K+1)+(1+(K*JAY/(ALPHA+BETA)))C WRITE(*,*)'A',ALPHA,BETA, 'HI' ,UP1 ,D0VJN1, JAYC WRITE(*,*)Z,T,'HERE',M,NC IterationsDO 10 1=2,590PL(I)= PL(I-l)+50.0P01=(P0+C/TAN(PHI))SIGR1= (SIGR+C/TAN(PHD)PL1(I)=PL(I)+(C/TAN(PHI))C WRITE(*,*) pOl,'HI',SIGR1C Calculating left and right side of the equations and C comparing for various limiting pressures LEFT1= 2*G*(N+K)/(P01*(N-1))RIGHT1=(T*((PL1(I)/SIGR1)**GAMA))-Z*(PLl(I)/SIGRl)ERR(I)=ABS(RIGHTl-LEFT1)RTERR(I)=SqRT(ERR(I))C IF (RTERR(I).LE.0.25) GOTO 40C GOTO 40C SHAPE FACTORS1=(TAN(PC)*TAN(PC))S2=l/(l+SIN(PHlO)S3=PC1*TAN(PHIC)S=S1*S2*EXP(S3)QC(I)=S*PL(I)WRITE(8, * ) I , P L ( I ) ,LEFTl,RIGHTl,S


353

C RTERR(I),ERR(I)10 CONTINUESTOP END


35 4

The following is the cavity expansion program based on the second procedure.

Program Listing

C Cavity Expansion Program - Anand J. PuppalaC Soil Model - Elasto-Plastic

DIMENSION EPSK200) ,SIG(200) ,EPSV(200) ,SIG1(200) ,Q(200) ,& PE(200),GAMA(200),EPS3(200),H(200),D(200),AA1(200),AA2(200),& DEPSV(200),DPE(200),DQ(200),DEPSVE(200),DGAMAE(200),& DEPSVPC200),DGAMAP(200),A1(200),DEPS1(200),DEPS3(200),& GGAMA(200),EEPSV(200),00(200),EEPS1(200) ,& EPTTC200),DIL(200),RAT1(200),RAT2(200),& PSI(200),SIGC(200),DSIG(200)

REAL MU,MU1,MU2OPEN(8,FILE='c.dat',STATUS='NEW') v=0.3WRITE(*,*)'SPHI ?'READ(*,*) SPHI WRITE(*,*)'SPHICV ?'READC*,*) SPHICV WRITE(*,*)'MU1 ?’READ(*,*) MUl WRITE(*,*)'HU2 ?'READ(*,*)MU2 WRITE(*,*)'SIGC ?'READ(*,*)SIGO WRITE(*,*)'G/SIGO ? ’

READ(*,*)CC WRITE(*,*)'C ?'READ(*,*)C G= GO * SigO E=2*G*(l+v)RAT=1+S0RT(1-SPHICV/SPHI)A=-(4*SIG0*((SPHI*RAT)**2)/(SPHICV*G))B=2*SIG0*SPHI*RAT/GC=SPHICVGAMA(1)=0H(1)=00(i)=ooq(i)=oSIG1(1)=SIG0PE(1)=SIG0


3 5 5

SIGC(1)=SIG0X=0.00318R=0.64EPTT(1)=0EPSV(1)=0EPS1(1)=0EPS3(1)=0PSI(1)=0DIL(1)=0D(1)=0A1(1)=0WRITE(8,*)'EPSILON THETA DSIGMA SIGNAC 'DO 10 1=2,200 DEPTT = X/R DGAMA=1.732*DEPTT EPTT(I)=EPTT(I-1)+DEPTT GAMA(I)=GAMA(I-l)+DGAMAH(I)=C*GAMA(I)*(GAMA(I)-A)/((GAMA(I)+B)**2)IF(H(I).LE.SPHICV)THENMU=MU1ELSEMU=MU2ENDIFC Dilation AnglesDIL(I)=((SPHICV-H(I))/MU)RAT1(I)=1/(1+GAMA(I)/DGAMA)RAT2(I)=GAMA(I)/DGAMA C The N ValuesPSI(I)=(DIL(I)+RAT2(I)*PSI(I-1))*RAT1(I)XX1=1-PSI(I)XX2=H(I)/(1+H(D)XX3=(1~EPTT(I))/(EPTT(I)*(1+EPTT(I)))C Incremental PressuresDSIG(I)=1.33*XX1*XX2*XX3*DEPTT*SIGC(I-1)+(C*C0S(SPHI)*XX1*XX3)SIGC(I)=SIGC(I-1)+DSIG(I)10 CONTINUEDO 20 K=l,20020 WRITE(8,*)EPTT(K),DSIG(K),SIGC(K)WRITE(8,*) ' No. 'H(gamma) ','Sphicv Dil Angle'DO 30 L=l,20030 WRITE(8,*)L,H(L),SPHICV,DIL(L)STOPEND


356

START ^

Yes

N o

Yes

U s e / I ,

X = X + 0.003

C alculate v , Y*

Assume X =0.003

Limiting P ressu re = £7

C a ’culate = Calculate h (r )

InputFrom CD Model

Figure D .l: Flowchart for Cavity Expansion Model (Procedure 2)


357

The ratios of tip resistance to limiting pressures are plotted in the following figures.

Cavity Expansion - Procedure 215.0

OCOOO Spherica l Cavity

100 KPa 2 00 KPa 5 0 0 KPa

Ücr0&C

5.0

0.020 40 60 80 1 0 0

R e l a t i v e D e n s i t y , D . (%)

Figure D.2: Ratios ( ^ ) Versus Effective Vertical Stresses (C.C. 0 %)


358

Cavity Expansion — Procedure 2

10.0

8.0

6.0

ÜGT

4.0

100 KPa 200 KPo 300 KPo

2.0

OOOOO Spherica l Cavity

0.080 1 000 20 40 50

R e l a t i v e D e n s i t y , Dj. (%)



10.0

Cavity Expansion - Procedure 2

359

C.C. 2 %

a,

ucr

8.0

6.0 -

4.0

2.0

0.0

A - B -

OOOOO Spherica l Cavity C —

' I l l ' l l . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I ■ I I I I ! ] I I I I I r I I I I I I 1 I

100 KPa 200 KPa 300 KPa

20 40 60 80 100

R e l a t i v e D e n s i t y , Dr (%)



T'btm

o f

are Prescj3B0

tbk

ro OiO ->

With^ 'P'/ss/on offhe °Pyright Owner,

re p ;« ,,ucf/on

Without ^ ''ooissior,

O'

Ü1

CO

%OQfo,

Pen'% /?

erer

"'ifecy0(/y

3/0»

362

cdb

* >b

( 9 .C * ^ v ) / ( * ^ a t m )6 0 04002 0 0

0

5 0 k P a

2

60 % 70 % 80 %3

4

Figure D.7; Approach 1 in Empirical Method (q j = 50 kPa)


Vita

Anand Jagadeesh Puppala was born in Rajahrnundry, India on May 8, 1964. He

received his B.S degree in Civil Engineering from Andhra University, Vizag, India

in May, 1985. Then, he received his M.S. degree in Civil Engineering from Indian

Institu te of Technology (IIT), Madras, India in February, 1987. Subsequent to re

ceiving his MS degree, he joined IIT as a research assistant. In .January, 1988. he

was adm itted to the Departm ent of Civil Engineering at Louisiana State University

as a graduate research assistant and a doctoral candidate. Mr. Puppala has suc

cessfully completed all requirements for the Degree of Doctor of Philosophy in Civil

Engineering on November 12, 1992 and he will receive the Degree in Spring, 1993

Commencement.

363


DOCTORAL EXAM INATION AND D ISSE R T A T IO N REPORT

Candidate: Anand Jagadeesh Puppala

Major Field: pivil Engineering

Title of Dissertation: Effect of Cementation on Cone Resistance in Sands:A Calibration Chamber Study

jor Professor and Chairman

Dean of the Graduate School

EXAMINING COMMITTEE:

(Co-Chairman)

Date of Examination:

Nov. 12, 1992


Documents

Effect of Cementation on Cone Resistance in Sands: A