of a rocky residual soil l

68
L L L L MISSOURI COOPERATIVE HIGHWAY RESEARCH PROGRAM FINAL REPORT STRENGTH AND DRAINAGE PROPERTIES OF A ROCKY RESIDUAL SOIL MISSOURI STATE HIGHWAY DEPARTM

Upload: others

Post on 22-Dec-2021

2 views

Category:

Documents


0 download

TRANSCRIPT

L L

L

L

MISSOURI COOPERATIVE HIGHWAY RESEARCH PROGRAM

FINAL REPORT

STRENGTH AND DRAINAGE PROPERTIES OF A

ROCKY RESIDUAL SOIL

MISSOURI STATE HIGHWAY DEPARTM

1

l L L L L L L

L L L

L J

r L

l L L r

STRENGTH AND DRAINAGE PROPERTIES OF A ROCKY RESIDUAL SOIL

STUDY NO. 75-1

Prepared by

MISSOURI STATE HIGHWAY DEPARTMENT

Division of Materials and Research

Final Report March 1976

in cooperation with

U.S, DEPARTMENT OF TRANSPORTATION

Fede ral Highway Administration

The opinions, findings, and conclusions expressed in this publication are not necessarily those or the Federal H.ighway Administration.

l l l l L L l l L l L

l L l L L L

-

ABSTRACT

The quantity of granular material in a Clarksville residual clay was varied to

determine the effect on strength parameters as determined by triaxial test procedures.

Consolidation and permeability characteristics were also studied. Some general conclusions

are drawn on the relationship of granular rootent to permeability and pore pressure

development in embankment construction.

,-

L

L L L l L L r ~

f

~

L ,

I L

,

l

Introduction

Conclusions

Implementation

Scope

The Study Soil

Preparation and

Analysis of Data

TABLE OF CONTENTS

Testing

2

3

4

5

8

10

LIST OF FIGURES

Figure

I. Compaction Curves for Clarksville Gravelly Clays

2. Gradations of Natural Clarksville Gravelly Oays

3. Gradations of Mixtures Prepared for Triaxial Testing (2.75 Inch Diameter)

4. Consolidation Curves for 2.75 Inch Diameter Triaxial Specimens at 60 psi Lateral Pressure

5. Stress - Strain Relationships of 1.4 Inch Diameter Specimens with 10 Percent Retained on Number 10 Sieve

6. Triaxial Shear Strength Envelope for 1.4 Inch Diameter Specimens with 10 Percent Retained on Number 10 Sieve

7. Stress - Stress Relationships of 1.4 Inch Diameter Specimens with 20 Percent Retained on Number 10 Sieve

8. Triaxial Shear Strength Envelope for 1.4 Inch Diameter Specimens with 20 Percent Retained on Number 10 Sieve

9. Stress - Strain Relationships of 1.4 Inch Diameter Specimens with 25 Percent Retained on Number 10 Sieve

10. Triaxial Shear Strength Envelope for 1.4 Inch Diameter Specimens with 25 Percent Retained on Number 10 Sieve

II. Stress· Strain Relationships of 1.4 Inch Diameter Specimens with 30 Percent Retained on Number 10 Sieve

12. Triaxial Shear Strength Envelope for 1.4 Inch Diameter Specimens with 30 Percent Retained on Number 10 Sieve

13. Stress· Strain Relationships of 1.4 Inch Diameter Specimens with 35 Percent Retained on Number 10 Sieve

14. Triaxial Shear Strength Envelope for 1.4 Inch Diameter Specimens with 35 Percent Retained on Number 10 Sieve

IS. Stress· Strain Relationships of 1.4 Inch Diameter Specimens with 40 Percent Retained on Number 10 Sieve

16. Triaxial Shear Strength Envelope for 1.4 Inch Diameter Specimens with 40 Percent Retained on Number 10 Sieve

6

7

9

13

20

21

22

23

24

2S

26

27

28

29

30

31

L [ Figure Page

L 17. Stress - Strain Relationships of 2.0 lnch Diameter

Specimens with 20 Percent Retained on Number 10 Sieve 32

18. Triaxial Shear Strength Envelope for 2.0 Inch Diameter

l Specimens with 20 Percent Retained on Number 10 Sieve 33

19. Stress - Strain Relationships of 2.0 Inch Diameter

l Specimens with 25 Percent Retained on Number 10 Sieve 34

20. Triax ial Shear Strength Envelope for 2.0 Inch Diameter

[ Specimens with 25 Percent Retained on Number 10 Sieve 35

2l. Stress - Strain Relationships of 2.0 Inch Diameter Specimens with 30 Percent Retained on Number 10 Sieve 36

L 22. Triaxial Shear Strength Envelope for 2.0 Inch Diameter Specimens with 30 Percent Retained on Number 10 Sieve 37

l 23. Stress - Strain Relationships of 2.0 Inch Diameter Specimens with 35 Percent Retained on Number )0 Sieve 38

l 24. Triaxial Shear Strength Envelope for 2.0 Inch Diameter Specimens with 35 Percent Retained on Number 10 Sieve 39

l 25. Stress - St rain Relationships of 2.0 Inch Diameter Specimens with 45 Percent Retained on Number 10 Sieve 40

[ 26. Triaxial Shear Strength Envelope for 2.0 Inch Diameter

Specimens with 45 Percent Retained on Number 10 Sieve 41

27. Stress - Strain Relationships of 2.75 Inch Diameter

l Specimens with 10 Percent Retained on Number 10 Sieve 42

28. Triaxial Shear Strength Envelope for 2.75 Inch Diameter

[ Specimens with 10 Percent Retained on Number 10 Sieve 43

29. Stress - Strain Relationships of 2.75 Inch Diameter

l Specimens with 20 Percent Retained on Number 10 Sieve 44

30. Triaxial Shear Strength Envelope ror 2.75 Inch Diameter Specimens with 20 Percent Retained on Number 10 Sieve 45

l 31. Stress - Stra in Relationships of 2.75 Inch Diameter Specimens with 25 Percent Retained on Number 10 Sieve 46

l 32 Triaxial Shear Strength Envelope for 2.75 Inch Diameter Specimens with 25 Percent Retained on Number 10 Sieve 47

L 33. Stress - Stra in Relationships of 2.75 Inch Diameter Specimens with 30 Percent Retained on Number 10 Sieve 48

l {

Figure Page

34. Triaxial Shear Strength Envelope for 2.75 Inch Diameter Specimens with 30 Percent Retained on Number 10 Sieve 49

35. Stress - Strain Relationships of 2.75 Inch Diameter Specimens with 35 Percent Retained on Number IO Sieve 50

36. Triaxial Shear Strength Envelope for 2.75 Inch Diameter Specimens with 35 Percent Retained on Number 10 Sieve 51

37. Stress - Strain Relationships of 2.75 Inch Diameter Specimens with 40 Percent Retained on Number 10 Sieve 52

38. Triaxial Shear Strength Envelope for 2.75 Inch Diameter Specimens with 40 Percent Retained on Number 10 Sieve 53

39. Stress - Strain Relationships of 2.75 Inch Diameter Specimens with 50 Percent Retained on Number 10 Sieve 54

40. Triaxial Shear Strength Envelope for 2.75 Inch Diameter Specimens with 50 Percent Retained on Number 10 Sieve 55

LIST OF TABLES

Table

l. Summary of Triaxial Test Data

2. Summary of Permeability Test Data Determined by One~Dimensional Consolidation Tests

3.

4.

5.

6.

Averages of Pore Granular Content

u-u Pressure Factor 1 ~

6 -6 vs. Varying

d1-u Effective Stress Factor 3 vs. Varying Granular

6 - u Content for 1.4 Inch Diameter Triaxial Specimens

cj l_u Effect ive Stress Factor 6 3 'IS. Varying Granular

-u Content for 2 Inch Diameter Triaxial Specimens

1 Effect ive Stress Factor 6 j -u vs. Varying Granular

d -u Content, 2. 75 Inch Diameter Triaxial Specimens

14

15

16

17

18

19

c

c'

" "0

LIST OF ABBREVIATIONS AND SYMBOLS

apparent cohesion (y intercept of MOhr's total stress strength envelope)

cohesion (y intercept of Mohr's effective stress strength envelope)

change in height

consolidated undrained triaxial test with pore pressure measurements

total pore pressure

pore pressure resulting from 6 3 confining pressure

induced pore pressure during shear

strain

angle of internal friction based on total stress

angle of internal friction based on effective stress

effective angle of internal friction based on deviator

effective angle of internal friction based on deviator

effective angle of internal friction based on deviator

stress

stress

stress

deviator stress (total axial stress minus confining pressure)

of 15 psi

of 30 psi

of 60 psi

J

INTRODUCTION

Many of the highest roadway fiUs in Missouri are constructed of residual clays

from the Clarksville series which contain varying amounts of admixed chert ranging in

size from sand to boulders. Slope designs for such high fills have been based on experience

gained from fills of lesser heights. Equipment restrictions have limited previous

consolidation and direct shear testing to samples with the granular fraction larger than

sand size removed.

In a previous research report entitled Moisture, Density and Slope Reguirements

in High Fills, it was concluded that such soi ls, when encountered without significant

granular content, should be limited by stability considerations to about 40 feet in height

with 2 to I slopes. This conclusion was based upon interpretation of direct shear tests

of essentially granular free soil. However, it was acknowledged that , since granular-free

occurrences were rare, such slope and height restrictions were inconsistent with

demonstrated performance of the normal occurrences of the soil. The present study was

designed to overcome these limitations and to develop data on strength, consolidation

and permeability characteristics of the soil with granular contents more typical of normal

occurrences. It was concluded that triaxial testing was the most practical means of

achieving these objectives for this soil.

- 1 -

CONCLUSIONS

1. With less than about 40 percent retained on the No. 10 sieve, the study

soil was indicated by triaxial tests to have an effective angle of internal friction averaging

32 degrees. The average effective angle increased to a maximum of 34 degrees with 50

percent granular content, the maximum tested.

Total strength parameters show a large increase in apparent cohesion at granular

contents exceeding 40 percent.

2. Triaxial and consolidation test data indicate the study soil to be highly

impermeable at granular contents of less than about 40 percent. At granular contents

over 40 percent, the permeability increases as much as fifty fold.

3. Triaxial testing, while slow and expensive, is judged to provide more realistic

strength parameters than were previously determined by direct shear testing for this soil

Similar testing would appear appropriate for any other gravelly soil to be used in

embankments of such height that stability problems could be likely or of serious

consequences should they occur.

4. Within the ranges tested, no significant differences in effective strength

parameters were found as a result of varying triaxial specimen sizes or the top size of

the included granular material

- 2 -

I L.

L L L L L L

f L

L L L L L

IMPLEMENTATION

Consideration should be given to requiring detennination during the soil survey

of the granular content of residual clays. This would be most useful in design of fills

built of such soils. A determination that the granular content exceeds about 40 percent

would suggest free drainage, rapid settlement and minimal stability problems. Those soils

with granular contents less than about 40 percent should receive specia l consideration

of settlement and stability problems where embankment heights exceed about 40 feet,

the limitation suggested by a previous study of this soil with minimal granular content.

Triaxial shear parameters, in the light of past experience, appear to provide a

realistic basis for predicting behavior of gravelly clay residual soils. As most of the residual

soil series in Missouri usually have high granular content, triaxial testing should be used

in future studies of slope requirements for high fiUs of such soils.

Both experience and the data derived in this study confirm that steeper fill

slopes are practical with such gravelly residual soils than are possible with most other

soils in the state.

- 3 -

SCOPE

A residual gravelly clay commonly encountered in heavy grading for highway

embankments in southern Missouri was modified with varying amounts of granular material

from a minimum of 10% to a maximum of 50% retained on the No. 10 sieve. These

varying mixtures were tested for determination of shear strength, consolidation and

permeability characteristics.

Triaxial test specimens were graded to match natural gravelly clays, statically

compacted wet of optimum and back pressured to saturation. Shear tests were performed

by consolidated, undrained (R) procedures to permit evaluation of both effective and total

stresses.

Specimen sizes of 1.4 inch, 2 inch and 2.75 inch diameters, with varying sizes

of gravel added. were tested for evaluation of the effect of top size and gradation of

the granular fraction.

Permeability data derived from both triaxial and consolidation tests was evaluated

to determine the granular content at which internal drainage would permit rapid

consolidation and dissipation of pore pressures.

- 4 -

l l L L L L L L l L L L I L

l L

TIlE STUDY SOIL

The study soil is a residual clay from the Oarksville series derived from the

decomposition of cherty dolomitic limestones. It is found throughout the Ozark Region

of southern Missouri. The soil sample used in the study is from Route 21 in Reynolds

County and was used in a previous study entitled Moisture, Density and Slope Requirements

of High Fills. It is relatively free of granular content with 10 percent retained on the

No. 10 sieve and 78 percent passing the No. 200 sieve. As sampled, it is non-typical

for the series although representative of chert free pockets frequently encountered. A

liquid limit of 69 and a plastic limit of 41 were determined. Soils of this series arc

found with a liquid limit range of 25 to 85 and a plastic index range of 9 to 54.

Compaction records from past projects show Clarksville series clays to exhibit

a wide range in granular contents and maximum densities. Typical compaction curves

and gradations are shown in Figures I and 2. In Figure I , Group A represents 6 samples

having less than 40 percent passing the No. 200 sieve and Group B 16 samples having

between 40 and 60 percents passing the No. 200 sieve. As these curves wouJd suggest,

considerable difficulty is experienced in construction control since the applicable maximum

density varies widely with the amount , size and specific gravity of the granular fraction.

Frequently the chert content is such that the soil is judged "too rocky to test" by

inspectors.

The known range of specific gravity of the admixed chert varies from 1.85 to

2.65. That added in preparation of test specimens had a specific gravity of 2.26 and

was composed of typical angular fragments obtained by washing of natural Clarksville

gravelly clay mixtures.

- 5 -

,e>

I

--~ " , , " •

u

"

u,

'"

..

,

~

~ Gcoup A

<. 401t. p"' .. i"9 n_< 100 .hwe

" " " .. " '" " ""'inure Conunt

Figure 1.

uo

C<O"P a

< 60l' but) 40% """'''9 the n"-" 200 .tev.

u, ---~ ; ] • , '"

a-.... ..

.. " " .. " •

" Moi.nuu CD"...,.

compaction Curves for Clarksville Gravelly Clays

o

~

" "

,

-.J

'F I 10

, ,

"" " -;;

I : ~

;., I " • o

• :< ! ~

I I "II iT )

IIII ' '

1m

... 100

II Tn

1\

~~ ~ R III III ~ -..: 111

AV·~.!1e ---=

~ "-

"W

III III Ii .. 10 . 1

Size l.n 111m

Figure 2. Gradation of Natural Clarksville Gravelly Clays

..

-

PREPARATION AND TESTING

Gravel and clay mixtures were prepared with the amounts retained on the No.

10 sieve varied in increments of 5 percentage points from 10 to 50 percent. Additional

fine granular content was added to produce gradations approximating natural Clarksville

soils as closely as the maximum allowable particJe sizes would permit. (See Figure 3.)

The maximum particle size was limited to that passing the 1/4 inch sieve and retained

on the No.4 for 1.4 inch and 2 inch diameter triaxial test specimens, and to that passing

the 3/8 inch and retained on the 1/4 inch sieve for the 2.75 inch diameter specimen.

The resulting gradation curves for the 2.75 inch diameter specimens are shown in Figure

3.

Test specimens were statically compacted to the maximum extent practical on

the wet side of optimum to obtain a nearly saturated specimen at approximately 95 percent

of AASHTO T-99 maximum density for the soil fraction only (less the admixed granular

portion). Thirty-three percent moisture was added based on weight of the soil less the

admixed granular portion. The granular material was soaked for 24 hours, drained and

blotted free of water before blending. All mixtures were cured 24 hours after adding

water and before compaction. Triaxial specimens were prepared in 1.4 inch, 2 inch and

2.75 inch diameters, all having a 2 to I height to diameter ratio. Consolidated, undrained

triaxial tests with pore pressure measurements (R) were performed in conformance with

AASHTO T-234-70 except that the ratio of specimen diameter to maximum particle size

was about 7 to I instead of 10 to I for the 1.4 inch diameter specimens.

Differential confining pressures varied from 15 to 60 psi. Porous stones were

used at both end caps with discontinuous ftIter strips to facilitate saturation by

backpressuring in 5 psi increments with a 3 psi differential maintained across the membrane.

Pore pressures were monitored using low displacement transducers. Saturation was

considered complete when an immediate resp:mse in pore pressure was measured by the

transducer after an increase in back pressure.

Consolidation was measured by both changes in the heights of the specimens

and changes in burette readings with drainage.

Rate of shear was less than 0.002 inch per minute except that a rate of 0.0027

inch per minute was used with the 2.75 inch diameter specimens to permit achieving

15 percent strain in a normal work period encompassing 9 hours.

Membrane puncture by the sharp chert particles limited successful testing to

40 percent retained on the No. IO sieve for 1.4 inch diameter specimens, to 45 percent

for the 2 inch specimens and to 50 percent for the 2.75 inch specimens. Forty-five

- 8 -

,

L

l L L L L L

psi was the maximum differential confming pressure successfully applied to the 2.75 inch

diameter specimens with 50 percent retained on the No. J 0 sieve.

Triaxial test data was resolved by a computer program with automatic plotting

of Mohr's circles and stress-strain data.

Specimens with 20,30 and 45 percent retained on the No. 10 sieve were statically

molded to one inch thickness for consolidation testing, with moisture content and density

similar to that of the triaxial test specimens, for determination of permeability.

•• , - , .

~. ::1+J-H±+--= t r--.::mt ~:,"~~4-~~~~~~~,

-'"

• ~~~~mtt++~#H~'n" , ""~~p-

1 fIl.I-I+H-+-H+I+H-H­

"~~~~H-~+H_III~I4-~H+_1

Figure 3. Gradation of Mixtures Prepared For 2.75 Inch Diameter

Triaxial Specimens

- 9 -

ANALYSIS OF DATA

General

Results of triaxial tests are presented as Figures 5 through 40. Mohr's circles

and plots of stress-strain relationships are included as alternate figures. Angles of effective

and total stress envelopes for varying granular contents and specimen sizes are summarized

in Table I. Both averages and standard deviations are also listed. Tabulation of stress-strain

relationship factors for varying granular contents and specimen sizes are presented on Tables

3, 4, 5 and 6.

Consolidation curves obtained during conditioning of 2.75 inch diameter triaxial

specimens with varying granular contents are shown on Figure 4.

A summary of permeabilities computed from one dimensional consolidation tests

performed with varying granular contents are shown in Table 2.

In the following discussion, "granuJar content" is used to mean that portion

retained on the No. I 0 sieve.

Effective Stresses

The effective angle of internal friction ($') varies from 29 degrees to 34 degrees

for the 1.4 inch diameter triaxial specimens over a range of I 0 to 40 percent granular

content. $' varies from 31 to 34 degrees for the 2 inch diameter specimens over a range

of 20 to 45 percent granular content and from 30 to 34 degrees for the 2.75 inch diameter

specimen over a range of 10 to 50 percent granular content.

Also tabulated on Table I are the effective angles (~') determined by striking

lines tangent to individual Mohr's circles back to zero. The angles determined for 15

psi confrning pressures are highly variable, probably due to the apparent pre-consolidation

resulting from sample preparation. At confining pressures of 30 and 60 psi , the average

angle is 30.9 degrees with granular content of less than 40 percent and 33.5 degrees with

granuJar contents of 40 percent or greater.

Comparison of the effect of confining pressure on ~' indicates a consistent

variation. With granular contents of less than 40 percent, the average angles are 31.6

degrees for 30 psi confining pressures and 30.2 degrees at 60 psi. For granular contents

of 40 percent or greater, the averages are 34.5 degrees at 30 psi and 32.5 degrees at

60 psi. This average 1.5 degree variation as a function of confining pressure is believed

due to apparent pre-consolidation effects at the lower confining pressures and, at higlter

pressures, to development on the shear plane of locally high pore pressures which cannot

be accurately measured at the specimen ends.

The average angle (~') is 31.8 degrees for all tests at granular contents of less

than 40 percent. For 40 percent granular content, ~' is 32.5 degrees and, for 45 and

- 10 -

50 percent combined, the average is 35.3 degrees. The standard deviations of the test

results are 2.7, 2.2 and 1.9 degrees respectively.

Total Stresses

Comparison of total stress envelopes indicates the available shear resistance of

varying granular contents. The total stress strength increases dramatically at 40 percent

granular content with an average increase in apparent cohesion (c) of 7.7 psi. The total

stress angle (Itl) averages 12.5 degrees, apparently without significant variation as a function

of granular content. More varia tion is evident as a function of test specimen size with

2.75 inch diameter specimens indicating an average value of Ib of 4.9 degrees grea ter than

was determined for 1.4 inch specimens.

The increase in total stress occurring at approximately 40 percent granular

content and above is believed due to internal pressure dissipation within the shear plane

from dilatancy and to pressure equalization within the specimen as a function of the

greatly increased permeability.

Triaxial Stress-Stra in Related Factors

Factor U llO is the ratio of the internal pore pressure measured to the d l -d3

added axial load applied to the specimen. This factor , included in the plots in alternate

Figures 5 through 39, averages 0.8 to 1.0 through the 0 to 30 percent granu lar range

but, significantly, drops as low as 0.3 for tests on specimens where granular content exceeds

about 40 percent. At the higher contents, dilatancy effects are probably tending to relieve

internal pore pressures

The factor

at the shear plane in combination with increased permeability.

d~-U is the ratio of the effective axial load to the effective 6 -u

confining load on the specimen. Plots of this factor are included in alternate Figures

5 through 39, for varying specimen sizes. There are no significant patterns to the variations

of this factor throughout the range of tests. This suggests that rate of shear, drainage

and monitored pressures are not significant variables through the range of specimen sizes

and granular contents tested.

Permeability

Permeability data, obtained from consolidation tests on specimens with varying

granular content, are tabulated on Table 2. The permeabilities shown are calculated from

coefficients of consolidation and compressibility and from compression indices derived

from one-dimensional consolidation tests.

This data shows the permeability of the specimen with 45 percent retained on

the No. to sieve t.o have a 50 fold increase in permeability over a sample having 30

percent retained on the No. 10 sieve.

- 11 -

Time curves, plotted from consolidation of 2.75 inch diameter triaxial specimens

at varying granular contents, are shown in Figure 4. Due to the complex drainage and

indeterminate time factors associated with use of discontinuous filter strips, coefficients

of consolidation were not calculated. However. a comparison of these curves indicates

that the time for 50 percent consolidation, which is inversely proportionate to the

coefficient of consolidation, decreases as the percent granular content increases. At 40

percent granular content, the time for 50 percent consolidation is in the one to ten minute

range as compared to 50 minutes for specimens with lower granular content. This

corresponds to the changes in permeability computed from consolidation data.

All data and observations indicate that at about 40 percent granular content,

a substantial increase occurs in permeability.

,

- 12 -

• 0 .~, .' • " .• • I ~ ,

5J , I

0 , .... • w

I

, " ". 1000

~I """"" ,,,,,-

"" ~ 'P" .""

"" "'-t"~ ""- ,,.

''''' ''''' ''''' ,,.

· Pe r cente snown sr. amount retalned on ~lO aieve

, " , .. 1000

Tl:ne in Minut ••

Figure 4. Consolidation Curves for 2.75 Inch Diameter Triaxial Specimens at 60 psi Lateral Pressure

r

.,. ...

-.

X G.ranular Content ••

10 -20

" 30

,

" 40

45

50

Average

S tandard Deviation

1.4~

SpecilDen Oialll'te, Effective Stre •• Total

Strel. ," · '1 ~2 .'3 0 c{psi)

,,' 3.' ,,' ,,' ll' 2-1

30' 30.5' 34' 30' ll' 3_1

31' ,,' 30.5' 29' ,,' 5,2

,,' 34.5' '" ,,' 11' ._3

,.' 30' 34' '0' 11' .,7

34' 34' 3.' 30' 5' 15.2

31' ,,' 33' 29.f

, 1. 75" 2 . )3' 1.96' 1. 75'

TABLE 1

S~&y Of Triaxial Teat D&t&

2.0 · S2!c~men Dl&meter

EfCectiye Strel' Total

.," .'1 .~ ., ~tre ••

• c(pai)

31' ,,' 31' 31' ll' 1..

,,' 30' ,,' 30' 0' • '" 37' '" 31' 1" 3,0

,,' 3.' '0' ,,' ll' 3,'

34' 30' 37' 34' g' 18.7

,,' 3" ,,' '" 1.1' 2.1' 3.2' loS'

" Aver age tangent to Mohr's circle, e f f ective stre •• (ca s e c' - OJ ~'1 Confi nlng p ressur e 15 pli (c'· 0) ., Confining p r essu r e • 30 pai (c'· 0)

2.75 · S2!clmen Qlam.ter

Effectiye Stre •• Total ;tr, ..

.," ,'1 O~ " • c (ptli)

31' 31' 31' 31' 10' 0

30' 3.' 31' 2.' 14' 0,.

31' 29' 31' 2.' ll' 0,5

,,' 33' '" '" '" 1.7

,,' ,,' 30' ,,' 10' 1.2

'" '" '" 31' lB' 2-0

34' ,,' ,,' ,,' 17' .,. ,,' 33' 31' 31'

1.25' 3.2' .98" 2 . 06"

"3 Conf i n ing pressure - 60 ps i {except for 50% g ra nu lar whe Le the confining pre •• ure • 45 psi ) (c ' . 0)

"" Amoun t retained on :.0 . 10 lieve

Aver' e. 0' 0 c(pd)

,,' 15.5' 1.1

,,' 13.3' 1.0

31' 11' 3,.

'" '" 3_.

31' 14' 3,0

,,' 11.5" 0,'

34'

34 '

'" 2 . 8"

• r-

.... '"

r- r- r- r- r'- , r- r- r- r- r- r- r- r-

Summary of

Granular Content·

",,. 20%

30%

45"

TABLE 2

Permeability Data Determined Consolidation Tests

Coefficient of Con!olidation .. r~ /O_ay x 10-2

2

1.5

1.5

60

*Amount reta i ned on No. 10 sieve **4 . a and 16 Taf loading average

.**At 4 ton loading in consolidometer

by One-Dimensional

Permeabi 11 ty ••• X10:-:'~ ,_~~

7.5

5.9

4.7

250

r- r- r- r-

.... '"

TABLE )

Averages Of Pore Pressure Factor (u-Uo) Va. Varying Granular content

(0'1-1$')1

% Granular Content*

10' 20 25 )0 JS

(1" 1 - 0') At )ox; Strain

15 psi .925 .94 . 89 ." .0)

)0 psi .925 .90 1.0 .87 .80

60 psi 1.025 1.08 1.1 1.11 .9

At 8% Strain

15 psi .81 .79 .0' .44 .44

30 psi .8 .91 .81 .75 .82

60 psi .97 1.0 .98 1.1) .94

• 1.4" and 2.75" diameter specimens only •• 2" diameter specimen only

... 2.75" diameter specimen on1:( * Amount retained on No. 10 SLeve ** (),,1-d3 = 45 psi

40 45·· SO···

.4' .1 .)

.0 .JS .4

.78 .4 .• **

.29 0 .0'

.>1 .18 .01

.70 .12 .42

r- r r r r r r- r- r~ r- r- r- r- r-

TABLE 4

Effective Stress Factor ( ~l_u ) Vs . Varying Granular Content (d3~)

For 1.4 Inch Diameter Triaxial Specimens

% Granular Content#

Standard 10 20 25 30 J5 40 Average Deviation

ef 1 - ~3 At 3% Strain

15 PSl. 3.2 2.5 3.3 3.75 2.7 3.3 3.13 .45

30 psi 2.9 3.35 3.05 3.55 3 . 4 3.9 2 . 93 1.24

60 psi 2.4 2.9 2.7 2.5 2.6 2.9 2.63 .16

Average 2.9 2.88 3.02 3.27 2.' 3 . 3 3.03 I-' ...,

Standard Deviation 0.4 .43 .3 .67 .44 .5 . 46

At B% Strain

15 psi 3 . 55 2.6 3.3 3.5 2.9 3.3 3.18 .3'

30 psi 3.3 3.5 3.1 3.6 3.6 3.5 3.3 . 32

60 psi 2.55 2.85 2.9 2.9 2.7 2 •• 2.78 .14

Average 3.13 2.98 3.1 3.3 3.1 3.23 3.14

Standard Deviation .52 .46 . 25 .4' .44 .31 .41

# Amount retained on No . 10 sieve

TABLE S

Effective Stress Factor ( ~ l_u) Va. Varying Granular Content ((r1"~)

For 2.0 Inch Diameter Triaxial Specimens

% Granular Content*

Standard 20 25 30 3S 45 Average Deviation

(1'1 - 6 3 At 3% Strain

15 psi 2.2 4.3 4.0 4.2 4.0 3.7 .87

30 psi 2.8 3.1 3.1 2.5 3.. 3.2 .48

60 psi 2.4 2.7 2.8 3.0 3.4 2.' .37

Average 2.47 2.1 3.4 3.3 3.2 2 . '

.... Standard ex> Deviation . 31 1.44 .83 .8 .7 • .8

At B% Strain

15 psi 2.3 2.8 3.4 3.4 3 •• 3.1 . 54

30 psi 3.2 3.0 2.8 2.7 1.5 3.0 .32

60 psi 2.7 2 . ' 3.0 3 . 2 3.2 3.0 .21

Average 2 .73 2.' 3.1 3.70 1.4 3.0

Standard Deviation .45 .10 .31 .2. .21 . 27

# Amount retained on No. 10 sieve

I r-- r-- r-- r-- r-- r-- .... .... .... .....

TABLE 6

Effective Stre.s Factor (~l-ul VB. Varying Granular Content ( d'3-u)

Foc 2.7S Inch Diameter Triaxial Specimen.

% Granular Contant*

Standard. 10 22 2S 30 35 'Q ~Qtt 6verage t!~iatiQn

crl - c1 l At 3% StEain

IS psi 3. 0 3.' 2.0 3.2 3 •• 3.2 3 • • 3.3 .28

30 psi 2.0 2.7 3.1 2.0 2.0 3.3 3.3 3.0 .23

60 psi 2.7 2.7 2 . 7 2.' 3.0 3.0 3.5* 2.86 .35

Average 2, '87 2.93 2.0 2.8 3.2 3.2 3.S 3.1

.... Standard

'" Deviation .15 0.' .2 •• .28 .15 .15 .25

At 8% Suain

15 psi 2.7 '.2 2.7 3.2 3.S 3.2 3.' 3.3 .S

30 psi 3.0 3.0 3.1 3.2 3.3 3.3 3.2 3.2 .13

60 psi 3.0 3.0 2.0 2 .S 3.3 3.1 3.S 3.0 .32

Average 3.0 3 •• 2.' 3.0 3 •• 3.2 3 •• 3.1

Standard Deviation .0. .7 .2 •• .12 .1 .15 .2S

.45 pai loading only *Amount retained on NO. 10 8ieve

N 0

")""0 - IS pol

,

"f: 1'!!. • , " • f! , • a:! • • M. • •

~.

" -~ L

" .. ,- 10 1/: Z~ .. M~ • , •

Figure 5

-. '" '.

.. " . " .. 7. ~

,

...... .. "". '" ..

" 7. " •

Stress with

")""0 -)0 p.l 11]""0 - n P.l IIl_ ..... ~O p.1

• • ,

• I

.' .......... .. .. - ... ·"11 .. • k'" • • .. ..... II • • , • , , t .... ",0' • " ..... ... .... " ....

• • • • .. , " • .. ,

" • .. , "

• • • .., .. , .. .. .. ' -... .. . " . • • . .. ... " . . ' • •

" "l " · • .- . ;;:::; . . .. '.

~ ~ L ... -- L ' • L

" .' , " ..... " . " • , , " 10 1- "

• , • , • • .. 7. " • " 7. " • .. j( " • • •

Strain 10

Relationships of Retained On

1.4 Inch Diameter 10 Sieve

Specimens Percent Number

r

~

~

I ". ..... .... '"

,,'

o L, /' 0

EFF ECTIVE ANGLE . 320

c· .. 0 TOTAL ANGLE ~ 13° c ~ 2.1

/" , .r- ~ 7- -,. -

r

, , ,

/ 1 __ '::,--

:>o.--~'\ •

• , J • , " ~ "

r

r

----

• •

-~ .- '$- -.

"

o

r- r- r- r- r-

."

, ~"""""~ , , " '

"

, , , , o '\

\

psi

II

."

• o

o

" ~ ~

r-

Figure 6. Triaxial Shear Strength Envelope For 1.4 Inch Diameter Specimens with 10 Percent Retained On Number 10 Sieve

r-- r-- r-

~ ,.

elL..., _ lS pd

--. , .. --2-2-

" ,._ .... _e 0"; .. '-

, • 2.2-.. ........ ... <0 • , ,

" ,

" • '" '" ",

,,' ,:1

- L ,. I" • .,. U ,. ... '. '" 2.:e .. <0

o I. ~ lO

Figure 7

eI).~O - )0 pel. ell."" _ 10 pol d'."" _ .~ pol "1.",, • 60 yO>

• 1

• 1

1 "o .... ee. h ". • ~U ......

~e"'O a e. 00 • a e"" ....... 0 0

" • J .. •

l _· ___ ....... . .. -. "

, " "

, " "

, "

, 10 ~. ~O

", ", ", 00 ...

I.;" -. '. eo e •

.. " .... D '. 10 i: " ':1 " • , ;;b Do •• D ...

'. -" '" " ." " , • " , .. ,: I "

..aU ... , " .. .. .t' • • " '. " , ,

L I

" I " ,

" " , , j , ,

" ,

" " , .. ,

" ,

" ,

" ,

" • •

1.4 Specimens Stress With

Strain 20

Relationships of Percent Retained On Number

Inch Diameter 10 Sieve

~

N

"-m 0.

'" ... "

.l-•

I

EffECT I VE RNGLE _300

C' · 0 TOTAL liNGLE· 13° C • 3.1

-- ---o

4. - .... ------:.-)1'" ' ----

/' -- . . - ' , - , ~ , - /1 '" ... -.::::-=:-• -?'- ~..- -.

_1- :>0-

~

.... ..~ ,

,

e ' 1- •

'. L \r.l , '/" 1 I, • • , I

" N ~ ..

r-' ,--.

~-, ,

\

' \

.. psi

"

, \

~

"

r- r"'" r"'" r- r-

--~

.. M ..

Figure 8. Triaxial Shear Strength Envelope For 1.4 Inch Diameter Specimens With 20 Percent Retained On Number 10 Sieve

r- r- r-

.. '"

.,.L ..... _ 15 pool

; r --~!. -- ...... " • 'f] ~!.

•• . '" ... '" -. • • " • " • " ,

'" ... .. .. • "

"~ -f! '" '" ~.!

•• -. • 0 ,0 X ...

Figure 9

01.. ... 0 - 21 pol .,.'_ .... - l O po l ell ....... _ 4 ! pol ,,1_ .... _ .0 ,.1

I . " . ..... ..... 0. o.

e .L • Ir ·0 0. 0 • . , ..

' " ., o "", ••

• , • ,olIO • " • " o ,e ... • " " • ,. • • o 0 .

0 .. , ", .. , "I '

r····' +:" " " .. .. .. . 0 0. o. - .' , . . ' .. ...... .. • .. .

• • • • • . . • · "t • • • ;.., . " • • ' .. '.

.... ,.

"L ,,[ " " • I • • , • " • • • "

, • " '." • " " •

1.4 Spec i mens Stress With

Strain 25 Number Percent

Relationships of Retained On

Inch Diameter 10 Sieve

~

~

• .<

'" ll.

'" "

I

EffECTIVE ANGLE . 31° C' • 0 0 TOTRL RNCLE • 12 (. 5.2

• "/ "' ,

?- --- ~ /.

r

, ""'". _s- __ _ ,

" ~ ~-- ./---.....

•• , ,

"

, ,

r- r-

------•

• -;:;~ - . , , .

'. "-\ \

\ , • \

'" psi

r- r- r- r- r-

~------,

, , \

• \

'" • ~ ~

Figure 10 . Triaxial Shear Strength Envelope For 1.4 Inch Diameter Specimens With 25 Percent Retained On Number 10 Sieve

r- r- r-

,~

d' ....... - 1) pel d' ...... o - 2) pool .,] ........ - )g 1'81 '; . "0 - 60 pel

0 , • , • •

l: ..... · .. ·· • "i> .- . .. ..... .. ..... o. • -- • • ~~ • • "

.. • I: .... · .. 0 ••

0> , , , , .. , '-~~

g r ........... ,.... . .... " r .••••• •• 0 •• • • _0 • • . -• o 0 10 X • • "

, " o 10 X ZO " • " % " ..., • • • •

'" •• 00 •••••• . , ,", ,", ~O , ..

• '"

" ~ .. , .. ' • , • o •• '"

• " " • "0.· • - .

~ ~ ~ . ~ L o ••• L /# L L

• • • " .' " • lO '. " • • .. .. ..

r I, .--- '" . . -010 " " "r " .- " . .!~ · . -; ... • • _0 • • • • • • • "

, " • " % " • " % " • " % " • • • •

Figure 11. Stress - Strain Rel ationships of 1.4 Inch Diameter Specimens With 30 Percent Retained On Number 10 Sieve

-

'" "

.~

B.

r-

N

"

r- r- r-

EFFECTIVE RNGLE • 329 c· • 0 0 TOTAL ANGLE .11 c. 6.2

,--, r- r- r- r- r- r- r-

------------" • - 0.:;.::;--__ ..... ,~

"" , -"". " .-- -.,....-~--.... ..-- , ..-, ....-'.. _ ......... , , _7__ "

> ,

, • ,

\

r- r- r-

~ .---_ ........

----- , , , , \

,

• •

Figure 12. Triaxial Shear Strength Envelope For 1 . 4 Inch Diameter Specimens With 30 Percent Retained On Number 10 Sieve

r- r- r-

" -~ . , , " " .!.! :::::::::: o. · , '" z-.! • • "" IV

ex>

-~ n '" Z-~

• • ""

Figure

6 3 -"'0 - 15 pd

• I • l!~' '" ...... '"

,

l'~ .. 0

.. ....... 0 " X "

""

301

,:1 ........ • •

.' • • •

" • ~.'

•• • • • 0

0 " X " •

13. Stress - Strain

d 3 _u • 30 p.,1 o

ll'" • ",'" . '"

,

r'" "" ...... 0

0 " X " •

"" .. '" e ..

"I .' • • •

1 ~ . " . E' •• .. '. . ••

• ,,1

0 0 " X "

Relationships of 1.4 With 35 Percent Retained On Number

6 1 -uO • 60 psi

, I ............. " .... '"

T_·" .... . ... ..... 0

0 " X " , "'

.. .. '" "'."'''' .... • • • • • • •

"

.. "" ..

" ;:;:; I. ~

"

"

0 0 " X "

Inch Diameter Specimens 10 Sieve

- r- r-

N

~

." '"

0 0.

'" "

D Ie

r- r- r- r- r- r- r- r- r- r- r- r-

..-! / . "

Ef f ECT/VE ANCLE • 29° C' • 0 0 TOTRL ANGLE . 11 C • 6.7

-- ---'::I'""-e , --.!",

'I·

, , - -,<,.-

, . , , Bo- -­, ..

~ , ' -/------'-'/,

, '" / "

'" \ . • .. ..

-~- ..

, ,

-- -

\

..

,

,

.. psi

,

..

------------- -

.. ..

Figure 14. Triaxial Shear Strength Envelope For 1.4 Inch Diameter Specimens With 35 Percent Retained On Number 10 Sieve

r-- r- r-

.. .M

IN o

11 --~:!. " "ioJ .!!. .. _m

n ~.!. .. -0

0' ....... U I'd

•• • • '''J' ..

• c~_ •••• • "z ..

• : •

.. • • ,.

.' •

" '0 ·

.':~

.. -· "'----,.-..-. 1,0110

Figure 15 .

"' .... 0 .. 21 pel

-~. "-• roo II '" ... • ,,:I.

. , . -,. • •

..

... "'-

'. '. ", · ,'---.-,,-;; o 'ox.

,,'""0 .. lO pOll

....... ' ..

· , ' ......... . I If X ..

• -• •

..

0" .. .....

• •

,'. .....

· ''---,.-,-;; • - •• X H

III' ...... 45 ,.1 • •

. ..... . · c· .... · '. "1 .. .. ." • -• •

..

• •

. •

• I •• _.

' .. ' .

.. • ..

Stress With

Strain 40 Percent

Relationships Retained On

of 1.4 Number

Inch 10

Diameter Sieve

dJ ...... ' 0 ~1 . .

I" .. 10 __

." • 'L_~:;-,-,O . - .0 Z 10

• -• •

0'· .. '. ,_

0' • '. "-

"L 0 0 ,0 f"IO

Specimens

~

w ....

." It

r-""

~

~

r- ,.......

EFfECTIYE RNGLE • C· ·0 0 TOUIL ANGLE. 5 C ·15.2

r-"""" ,.......

34°

_ _ .rd:-..;-r--=---

r- r- r-

--------- ...... -----S ---=-=-=-- -

r-"""" r- r-"" r-"" r- r--

---~ --~<:"-~

~

-------- ---• .,.,..;;..""" _ ......... , .... :fl,

. " • , ,

"

• ~ • "

• •

, , , • I .11 • f" , , , , ,

II ,

H " ..

, , , , \

• /\ •

~. PS,l.

..

, \

'. \

\

" ..

Figure 16. Triaxial Shear Strength Envelope For 1 . 4 Inch Diameter Specimens With 40 Percent Retained On Number 10 Sieve

r-- r-- r--

.. ,.

<:11 ""'0 • n ,d

,,1 ". :!:!

" ~'¥ ...... .. • ..

~:! .. ....... , -. , "

, "

~,

w " IV

" • •

f! L " . :;~ : ....... .. _0

" , •

Figure 17.

<:11 .... 0

• U ,.1 <:11 .... 0 • 10 pol ,,1""'0. 10 pol

, , • , , ,

". .. .. . . ..

~ .

, :~ ... - '"

,

," ........

" , • "

, • " , •

, ~, ~, '"

" " • " " " • • • • • .......... , • .:PI~~ ... - •

L " :' -. " , • ,. x ,.. ,

" ,

" " , •

Strain 2.0 Stress with 20 Percent

Relationships of Retained On Number

,,1 .... 0 • 60 pol

I .. ··· ....

L "

, "

'" .' •• ;.v= :

'.'

~~: •• •

"L , "

, •

Inch Diameter 10 Sieve

"''''0 • 60 pol

,

, ...... , ': • f,··· ....

F o~ "

, • . ' ..

" I .....

X' • "

"I , , I • "

, •

Specimens

- r--"

~,

" . ~ ~

w '" w

"

00

r- r- r r- r-"" r-"" r-"" r--" r-"" r-- ,--.

EFF[CliVE ANGLE· 31° C' • 0 lOTAL ANCLE • 13° c. 1.6

, , ",II, -

, , ,./ ,

---~ - -.Jo .;.",. --

" ' -' . . ... ... • •

, , , ' , '­. -~ --",

----t.,_ -,- '" ... ..

. ,' .. .. .. - . :,,\,

. , , ~ . , . ~ ~ . '

-'

\

\

--

" ,,"l \

-.<4;;-;, ~_#1=:i--' i,,'o :2if,-= . " U'~~_ . - , -

' M

" psi

---

• • ~ ~ ..

Figure 18. Triaxial Shear Strength Envelope For 2.0 Inch Diameter Specimens With 20 Percent Retained On Number 10 Sieve

..

r- r-

,.

ctl-o - 15 PIli ,

d-o-30pai ';-110 .60 pd

• , •

-- • . -, • • ~ T ~- • <'J<'J"'''', 2.~ ~ .... ~ • • " • ,. , , • , . ~~ .a:!. ......... --- ',;"" .. "--• • .'" 0 0 0

0 " % " 0 " ,

" 0 " % " • • • ,: ." ...... w "I "[ ''I-... e'" !lee

• .. " 'l • ~ ~ ~. ~ ~ ~ .. -" J"""'- 20 Jf "" " t- .. • • - " "L " t ... '? -

.a~

•• . " 0 0 0 " % " COlO f. 20 0 "

, " -• • •

Figure 19. Stress - Strain Relationships of 2.0 Inch Diameter Specimens With 25 Percent Retained On Number 10 Sieve

I 0" m we. '"

r

~

~

"

r

EfFECTIVE ANGLE .320

C' • 0 TOTAL ANGLE • aO

c • 6 . 0

-- -- ----- ... , ,

, .. , .,;. __ J_ ...

----~/----)~ .- 1' . \

I • '-• ,

, ,

• -... ..>- - --­-- -.-- ...

--~ \. ------ .......- . --....... \

, , •

r r ~ r- ~

----------.-

• •

. 1/ I I I Y . \ 1\:/ \ o 10 ~ 10 " 50 .., 70 10 .. ."

psi

Figure 20. Triaxial Shear Strength Envelope For 2 . 0 Inch Diameter Specimens With 25 Percent Retained On Number 10 Sieve

r-

... a-

6 1 -uO .. 15 pai 6 1 -1:10 .. ]0 l*i ~~ .. 60 ~i

, • • •

~ • l" , , ,

v"¥ .. .. ..• l"ee '21- I.aaaaa-e_

" " .,1 , , , e~ r" ( ........ J.-- .. _ • • "0 0

- ........ 0 0

0 " '- " 0 " '- " 0 " '- " , • • •

", ", ", . ' . • a.~

" t .. " " •

~ ~ ~ ~ ~ .a ...... ~

" " 0- .. " 0·-"- • • •• " " , - ~ . "l O~ " . " r~ .......

e~

• • "[:J 0 • 0

, 0

0 " '- " 0 " '- " 0 " '- " • • •

Figure 21. Stress - Strain Relationships of 2.0 Inch Diameter Specimens With 30 Percent Retained On Number 10 Sieve

.

-

w ....

r- r"""" r- r"""" r- r- r r-- r- ,.--.. r- r-

~

~

"" • "

"

--1/ • •

EFFECTIVE RNGLE .. 32° c· .. 0 TOTAL RNGLE C • 2.8

• 12°

-;.-;.-...

, __ J./ B / .

• •

"

-, -

,

... ~ .• -"," J(' ..

"'" .. - r- -~ __ = :;;:-..... , , , •

\ . \ • • •

----:---•

~ '0 .0

--------" ')._ ... --------- " ,

\

\ • " \ •

~ ~ " psi

Figure 22. Triaxial Shear Strength Envelope For 2.0 Inch Diameter with 30 Percent Retained On Number 10 Sieve

-Specimens

r- r- r-

- ,.

.,'-0 - 18 p-i 6'""\10 - '7 p.i til-v - 70 pd 0

• • •

~ - • • • f 1 • ~~ • "e • ." ~ __ . .£~ 0

" .1". - ... • /~

0 , , · '0 , 0 • ,~

2.~ •• 0" • . ~. .. ... • • • • . '" . ... " . , 0 , 1-0 " y. " 0 " y. 20 · , 0 00 " •••• • • .. "c •

• "" "" ", " •

• w .. I "I .. I: (XI . '"

::' 1 ..... ~r 0

~ ~ . ~

• • · .......... 20 ... 20" • " • 0

" 0

• • , -~

10 ~.e 00' 00 ,of '. ,~ " .... 2.~

• • .'" , • , • , , 00 y. " , 00 y. "

, 00 y. 20

• • •

Figure 23. Stress - Strain Relationships of 2.0 Inch Diameter Specimens With 35 Percent Retained On Number 10 Sieve

." w • '" '"

"

o rff[C l lVE !'INGLE' 32 C' • 0 lOll'll ANCU • 13° c · 3 .6

, , _-r

, , • ..J.._

,

/

", . , . , , ,

---, -.~

, ,

/"

\

~. -- - ---

,

,

I

,

, , , - '-­, , ,

, \

\

\

r r-" r-

----

\

,Cd:: , " • • ~ • • • • • ,. psi

Figure 24. Triaxial Shear Strength Envelope For 2.0 Inch Diameter Specimens With 35 Percent Retained On Number 10 Sieve

r- r-

cr-o - 15 pi ~'"'\Io - 30 pd

• •

~ - , - , H '. ~ '. ~~ CIt! 'fJ D 'fJ.fJ o 'fJO 0 a-~~ " " n , , ,~

.0 2.~ .. , •• • .'" • • • •• 0 0

0 " ,

" 0 • " ,

" .' ~ • • •

" • " • • ...

0 • •

" • " ~ ~ ~ • ~

" " ~.

- "1 •

.'Y • ,~ " • 2.E ,

" • . " • • • . ", •

0 0 I 0 " I,p y. " 0 "

, " . '-

• •

Figure 25. Stress - Strain Relationships of 2.0 With 45 Percent Retained On Number

Inch 10

0-3_ - 45 pd o

, ot6OttJee 'fJ DO e'fJ

• , ." • 0

• .0

• ...... " 0

0 " ,

" • •

"

.. ~

,:r···· '. '. "

0 0 "

, " •

Diameter Specimens Sieve

. ~ ... . ... 0,

0'-

"

,

EfFECT r~E IINCLE • 340

c· • 0 0 TOll'lL liNGLE· 9 r • 18.7

. -•

"j •

---

•• '/. ". , , ~7<

f

,

-- -e ~!!" -----

,.~y~

---. / \

• •

• •

------

, , , , , \

\

, , , , , , \

r

\

\

--

,--, r-"'

o 1 /" • * o ,. l(! , . •• .. Ie >0 • III ,au

psi

Figure 26. Triaxial Shear Strength Envelope For 2.0 Inch Diameter Specimens With 45 Percent Retained On Number 10 Sieve

r-'" r- r-

M _

V 1 "M

~~

" f1 Z~

• • x."

... '"

~ n '" 2.~

•• x."

Figure

6'_ .. 15 pd o

• IJPOooao a -• • ,

••••• , '-0 ,

" l. " •

"'

"

,: I ~i'" .-10 ... .. _

, 0

0 " l. " •

6'-\lo .. 30 pd

• I

,p'~ 00 0 0 0

• , • • ... ,- • •••• ,

0 " l. " •

"" ,.I .' . • • ~I . 2: , .. 0

, ,-" , , , , , , •

10 r.

0 , , ,0 l. "

27. Stress Strain Relationships of 2.75 With 10 Percent Retained On Number

6 3 -"0 .. 60 pel

,

~o"" 0 •• : , ........ ,

0 0 " l. "

• 1, ......

• "' •

" , • • , , , '0

]0 I "

~

~ I: "

"

0 0 .0 l. " •

Inch Diameter Specimens 10 Sieve

... w

.~

it

"

o EffECTIVE IIIIClE • 31 c· · 0 0 'otlll RNGLE . 18 ,. 0

'. ,

, " .. ... ,

'. •

r

~~---------

~~~

~~

I ----.L --

.'.'. /'. ~~-----.~ / __ ..... >/;::-r I ---- j' " . ~~ . /' . . ' .• \ . \

'­~~.

~ .

• , , , , , \

\

\

r-.

,1..-?-1 'fI n f . 1 o 10 •• >II •• 50 ... ,. IICI 10 ,.

psi

Figure 28. Triaxial Shear Strength Envelope For 2.75 Inch Diameter Specimens With 10 Percent Retained On Number 10 Sieve

r- r-'"

d 3-uo • is peJ. d 3 _0

.. 30 peJ. d 3 -uo .. 60 p.J.

, , • I • •

" - • • • , • , , , , • " " ~.!. • "'e e ... • .... ••• I.,"' ''"' e aell. e " • • •

" , , •• , · , • '" Z.!. ...... " .......... ... """ ... ... f .................... • • • • 0 , , , ,

" ,

" ,

" ,

" ,

" ,

" • • • .., "I "I "I

• • .. .. .... , .., .,.,., . •• ••

0

" " ':1_ ~ 00 ~ ~ ~ ~

• " " •• :e:11 20 I • .....

• '. -" I .... "r ,'? 10 ,lAme lie •• : Ie " '" 2.2 • • . '" • , , • , ,

" ,

" ,

" ,

" ,

" ,

" • • •

Figure 29. Stress - Strain Relationships of 2.75 Inch Diameter Specimens With 20 Percent Retained On Number 10 Sieve

r I

"

" ,

. .< ~ 0.

"' .. I '"

, 0

0

r

fFFECllVE RNGLE • 30° c· .. 0 HITRL RNGLE C .. 0.6

,

_ 140

;.r:::---------/. --' -, -~", "

. '

. ---' ' 7- ----~-. .', --- --.. , . ----~- '

-- • \

" " .. ~

psi

r r-"

-----

'. \ •

•• " ~ ..

Figure 30. Triaxial Shear Strength Envelope For 2.75 Inch Diameter Specimens with 20 Percent Retained On Number 10 Sieve

r- r-"'

,.

d1"'\lo • 15 ~i d 1 -., _ 10 pi dl"'"'\,\O - 60 p.i

• • •

" -V? , • • ~" ~~ ( .... _. " ,p ...... ....... 1 .-- .. -.. '"f1 :~ .. -.. '" ,~

, , , Z~

• • ...... ".,," ." .. " .. " "i!I

• • , • • " r. " • " r. 20 • " r. " • • • .. .. .. ... . " .. " " " ... " _ D· ••. '" .--•

" " " I ~ ~ ~ ~ ~ ~

" 20 " ...... " II 20 . -. ~ •• D --1\> 10 1$11 •• n~~:: " I " ,~

z~

• • "'" • • I • • " r. " • " r. " • " r. 20

• • •

Figure 3l. Stress - Strain Relationships of 2.75 Inch Diameter Specimens With 25 Percent Retained On Number 10 Sieve

-

.. ...

r- r r r r- r-"' r-" r-" r- r-"'

• .<

• '"

K

~

I

"

EFFECT lYE ANGLE ... 31° C' .. 0 TOTAL RNGLE ... 13° c ... 0.5

------,

I . / .......

c-, . , . , ,

. , ,

-------

---, ---, , --,-, -' , -<: , \

• \

'/ ~:' \ •

. ~:-(;.;! '\' \ ! j 1 a LO ~ " .. " .. " .. ..

psJ

Figure 32. Triaxial Shear Str~ngth Envelope For 2.75 Inch Diameter Specimens with 25 Percent Retained On Number 10 Sieve

r- r-"

, .

d l __ 15 pd o

dl-vo - lO pd dl-u.o • 60 I'd

".< • .. 0.

'"

"

M

"

,

EFFECTl .... E RNGLE .. 32° c· a 0 0 TOTAL RNGLE . 12 c .. 1. 7

• /

.-+---/

i , '>-- -­-----; ....

, , -. --,

~ \ •

. -- - --

• •

-----•

, , .... --­_ -'C"

.------::-..... \

\

I

--

ole" "\\ T \\

r r r r r- r- r-

,

----- ------- -

• •

o 10 20 '0 .0 10 .., 10 to to I • •

psi

Figure 34. Triaxial Shear Strength Envelope For 2.75 Inch Diameter Specimens with 30 Percent ~etained On Number 10 Sieve

(11""'\,\0 - 15 pd ,r""'\,\ - 10 pd o d1-uo - 60 ~L

, , •

11 •

r·~·· -. •

l,···· "'-' •

"11. I_ dI - - -- - -

" • • 1] , , , • z.:c , , r···· . '" ... o t···· ...... -..... • • • Ol

0 o 0 LO;: 20 o 10;: 20 0 II ~.-- "

• • • of' •

"I "I • ~O I ..

'" •• 0 • • •• • •

• ,. " ,. .. ' Vila ~ ~ .' .

~ ~ .0' ~

• " " I·· .. " --~ -- • • ••• .. ' ••

Jj " F· "I " '" Z~ · ..... , , ." 0

, o i>- 0 0 " % " 0 " % " 0 " % "

• • • Figure 35. Stress - Strain Relationships of 2.75 Inch Diameter Specimens

With 35 Percent Retained On Number 10 Sieve

V> ....

• 0.<

Ii

"

[HEeTI¥[ AJ(Cll • 32° c· ·0 0 tOlAt ANelE . 18 c • 1.2

, • ..!_ ,J

" -

, , ,

I

.. I ........... ~ '"i''':::' ~'' ---r '~'- "

I ... ....

"'- - , y \

L r r r- r-

------. -..- , , ,

,

..,,-~ - .\ 7' · V-I I i l k' \

o ,0 III 10 •• 10

Figure 36. Triaxial Shear Strength Envelope For 2.75 Inch Diameter Specimens with 35 Percent Retained On Number 10 Sieve

r-

6]-110 .. 15 I'd , d ""'Uo .. 30 p.1 crl-uo .. 60 1'.1

• • , • I

" - •

r'~" .... : • ~.~.

• H • 11- ........ .. .... .......... " • • '0'; • , , , • , , • ,~

Z~

• • ,.. ••• x· t'" • ••• C·· X

! • OJ .' . 0

• • •••• 0 o '--

0 " ,

" 0 " ,

" 0.. 10 7. " • • • •

" " ''I" U>

'" ••••

" . " 0

", ", 000 " I. 00 ",0

,: I ~ 0 :c I. ~ .......... 0

" " ,pO ". • ••• • • 0 0 •

~ on "f " " ,~

'" .. " -- • ••••• • • 'OJ • • 0 0 0

0 " ,

" 0 " ,

" 0 " ,

" • • •

Figure 37. Stress - strain Relationships of 2.75 Inch Diameter Specimens with 40 Percent Retained On Number 10 Sieve

r

.~

• '" 0. w

r ,---. r- r- r-" r-

EffECTIVE A~CLE • 32° C· • 0 ISO H"IIL AN~l[ • , . 2.0

-~............... - ~ " .~,' .::.- .....

/" - , .-,. , . - , y" ., , , , ,

/,: ' -<-~ \ , --\

.---7-·. : .... --- • \

" ~'. '" . "

" I

f , "

, -;/ ( V:

.v • "

, • .. .. .. , • • , . psi

Figure 38. Triaxial Shear Strength Envelope For 2.75 Inch Diameter Specimens with 40 Percent Retained On Number 10 Sieve

r-" r-"

~ -~ 1 . ~ ~E.

" 11 ' . .=.~

• • "~

'" ..,

-n '-n --• • ""

Figure

0 1 _" .. U p_i o

,

• ~.~~ : ............ ,

of:..."" .. ! •••• o 10 j(

"I

,,' .0

000

",'

~I . 2: 8'

0

"

"L 00 10 ·°j( 20

(f1 _Ug .. 10 pd

,

• r' o

""

0

,

t:: ...... • , , " y. " •

"I o 00 0

00 0 • 0

0

0

" • ~

~

" •

" • • · '.

, f , "

y. " •

39. Stress - Strain Relationships of 2.75 with 50 Percent Retained On Number

d1-"o .. 60 pd

,

• 0 ~ ..........

0 0

0 , E 0

t· ..... .... , 12 ........ , 0 ;;;0

" y. " 0 •

" 1

0

" III L .00. •• • ~ . •

" I· "

.i , " y. " •

Inch Diameter Specimens 10 Sieve

M

.. , ...

e on", on

"

• V •

, ,

EffECl lYE PMGl[ • 34° c· ·0 TOTAl ANCLE • 17° c ·6.9

, , r

., ' , , , ,

, ---- . u' • ----. .--! -~-- -~.- :.. -.-~

/

, / ,

---, , ,

r

. ------- -

------.... - , , ,

--, , , , , , , , ,

r- r-

\

\

\

--- / . ., f \

/ • •

"

, "' / ,

~ .

J .. M • y • ~ • • psi

Figure 40 . Triaxial Shear Strength Envelope For 2.75 Inch Diameter Specimens With 50 Percent Retained On Number 10 Sieve

r- r-

,.