Content uploaded by Hiroshan Hettiarachchi

Author content

All content in this area was uploaded by Hiroshan Hettiarachchi on Mar 18, 2015

Content may be subject to copyright.

ASCE Geotechnical Special Publication No. 179, ISBN 978-0-7844-0972-5

Page

1

Estimating Shear Strength Properties of Soils Using SPT Blow Counts: An

Energy Balance Approach

Timothy Brown1 and Hiroshan Hettiarachchi2

1 Geotechnical Engineer, Commonwealth Assoc. Inc., 2700 West Argyle Street, Jackson, MI 49202;

tsbrown@cai-engr.com

2 Assistant Professor, Department of Civil Engineering, Lawrence Technological University, 21000

West Ten Mile Road, Southfield MI 48075; hiroshan@ltu.edu

ABSTRACT: The subsurface exploration of a site is often the aspect of a project that

gets overlooked during the design process. Many clients will get standard soil

borings, but do not want to pay for a full laboratory analysis. Lack of data forces the

designer to estimate important engineering properties of the soil. Very often the

Standard Penetration Test (SPT) blow counts are used to estimate the shear strength

properties of soil in foundation designs. Few correlations are widely used. However,

no clear explanation is found to justify the selection most of these mathematical

equations. This manuscript describes a new approach to estimate the shear strength

parameters based on the SPT blow counts. In this method, the standard penetration

test is treated analogous to driving a miniature pipe pile. The energy input to the soil

is used to correlate the SPT blow count to the shear strength parameters of the soil at

the depth of testing. Soil boring records from few different sites were analyzed and a

statistical analysis revealed that the proposed method can provide a better estimation

than the widely used existing correlations.

INTRODUCTION

A combination of soil borings and laboratory testing is the most reliable method

available to obtain accurate shear strength properties for subsurface soils. Many

projects, due to limited budgets, tight schedules, or lack of concern, do not usually

have the luxury of getting laboratory recommendations. In many cases, the only

subsurface exploration performed consists of soil borings with a log recording the soil

type and classification, depth of water table and SPT blow counts. Lack of lab data

forces the designer to estimate the properties of the soil.

When laboratory data is not available, it is a common practice to estimate the

shear parameters from the of the SPT results. There are many charts and tables

available to make direct correlations between the SPT blow count (N) and the angle

of internal friction (

) and undrained cohesion (cu). These estimations should be

made by individuals who have a thorough understanding of soil behaviors. It has

ASCE Geotechnical Special Publication No. 179, ISBN 978-0-7844-0972-5

Page

2

been the authors’ experience that this is often times not the case. Engineers with little

or no experience in evaluating soil borings and estimating

and cu are sometimes

expected to design foundations. It is very common for an inexperienced designer to

use a design chart which is not fully understood. It is this practice that shows a strong

need for a reliable tool to assist in design when a complete laboratory analysis is

unavailable.

EXISTING N60 -

AND N60 - cu CORRELATIONS

A brief review of widely accepted correlations of N value to

and cu are

presented herein. The two most common types of SPT hammers used in the US are

safety and donut hammers. Energy studies have revealed that the efficiency of safety

and donut hammers are about 60% and 45% respectively. When SPT results are

presented it is customary to modify the blow counts to the 60% efficiency levels

(N60). Almost all the correlations are hence based on N60. It is important to note that

the factors such as borehole diameter, sampling method, and rod length are also

incorporated into this standardization process.

Early work on estimating

from the N60 value attempted to make direct

correlations. Meyerhof (1956) and Peck et al. (1974) tabulated recommended values

for estimating

. Peck et al. (1974) published a graphical representation which was

later approximated by the following equation by Wolff (1989).

2

6060 00054.03.01.27 NN

(1)

Results from a laboratory research by Gibbs and Holtz (1957) showed that

overburden pressure could significantly affect the SPT blow count. Schmertmann

(1975) considered overburden pressure to develop a relationship between N60 and

.

This correlation can be mathematically approximated as follows (Kulhawy and

Mayne, 1990) where

is the effective overburden pressure and pa is the

atmospheric pressure.

34.0

60

13.202.12/tan

a

p

N

(2)

Despite the research shown above, there have been few other attempts to correlate

directly to N60 without considering overburden pressure (Peck et al., 1953, and

Japan Road Assoc., 1990). Hatanaka and Uchida (1996) tested high quality,

undisturbed frozen samples from few sites in a standard triaxial apparatus and the

friction angles were compared against the corresponding N60. They proposed the

following equation to estimate

where CN is a factor to correct N60 to a standard

overburden pressure (100 kPa).

2020 60

NCN

(3)

ASCE Geotechnical Special Publication No. 179, ISBN 978-0-7844-0972-5

Page

3

Correlating cu to N60 has been attempted many times. Efforts have been made to

find a general relationship for all clay types. The following equation presented by

Terzaghi and Peck (1967) is one of the more commonly used methods of estimating

cu for all clay types.

60

06.0 Npc au

(4)

Some believe it is unlikely a generally accepted relationship between cu and N60

will be found and a realistic correlation between cu and N60 may be possible for clays

within the same geology. The following equation by Hara et al. (1974) is an example

for one such effort.

72.0

60

29.0 Npc au

(5)

PROPOSED METHOD BASED ON ENERGY BALANCE

Modified version of Mohr-Coulomb failure criterion is typically used to estimate

the shear resistance between soil and pile material such as steel, concrete, etc.

Therefore, the shear resistance (

f) between soil and the SPT sampler is modeled by

the following equation where ca and

are adhesion and angle of friction between soil

and the sampler. K is defined as the coefficient of lateral earth pressure.

tan

Kcaf

(6)

Driving the sampler is analogous to driving a pipe pile. Assuming no plug

formation, the resisting force is the

f multiplied by the surface area of the sampler

both inner and outer. The work done by the sampler to overcome the

f of the soil

(E1) can be estimated by resisting force times the distance traveled (d).

dKcAKcAE innerainnerouteraouter )tantan

1

(7)

It is assumed that the lateral pressure on the inside of the sampler is zero. Inside

surface area of a standard sampler is approximately 70% of the outside. Therefore,

equation 7 can be simplified to:

tan7.17.0tan

1

KcdAdcAKcAE aouteraouterouteraouter

(8)

The energy transferred by the hammer to the soil (E2) is the total work done by

the hammer times the hammer efficiency (

). However, it is convenient to use

standardized N60 instead of the field N which results in the following equation where

W is the hammer weight and h is the drop height.

WhNNWhE 6026.0

(9)

ASCE Geotechnical Special Publication No. 179, ISBN 978-0-7844-0972-5

Page

4

Assuming that there is no other energy lost to the system, equation 8 and 9 can be

set equal to each other to find N60 as a function of shear strength parameters. D is the

outer diameter of the sampler. Shear strength parameters are made non-dimensional

by dividing them by pa.

)tan7.1(

6.0

)tan7.1(

6.0

2

60

aa

aa

a

outer p

K

p

c

Wh

pDd

Kc

Wh

dA

N

(10)

The parameters pa, D, d, W, and h are constants and hence can be replaced by a

constant (B) to form a general equation.

)tan7.1(

60

aa

ap

K

p

c

BN

(11)

0.5

in30lb 1406.0 in

lb

144

2000

in12in2

6.0

2

2

2

Wh

pDd

Ba

(12)

PROPOSED N60 -

CORRELATION FOR COHESIONLESS SOILS

For granular soils, adhesion (ca) is zero. Angle of friction between soil and pile

material (steel in this case) is typically assumed to be proportional to soil friction i.e.,

=

where

is the constant of proportionality. Reese et al. (2006) proposed to use

K=0.8 for open ended pipe piles which are driven unplugged. Therefore, when the

soil is granular the general equation can be deduced to the following.

tan4tan8.05tan

60 aaa ppp

KBN

(13)

a

p

N60

125.0tan

1

(14)

Results from 36 standard penetration tests conducted at 24 different boreholes in

Oconto and Marinette County, WI, in 2005 were used to estimate

parameter in Eq.

14. Details of these tests and the laboratory evaluated friction angles of soils obtained

at the same locations are reported in Brown (2007). These data produced an average

value of 1/

=0.3818 with a 0.018 standard deviation. Low standard deviation

indicates a reliable value for

.

PROPOSED N60

cu CORRELATION FOR COHESIVE SOILS

For cohesive soils

is zero. Adhesion between soil and pile material (steel in this

case) is typically assumed to be proportional to undrained cohesion i.e., ca=

cu where

is the constant of proportionality.

ASCE Geotechnical Special Publication No. 179, ISBN 978-0-7844-0972-5

Page

5

a

u

a

u

a

ap

c

p

c

p

c

BN

5.87.15)7.1(

60

(15)

60

5.8

1N

p

ca

u

(16)

Results from 14 standard penetration tests conducted at 9 different boreholes in

St. Clair, MI, in 1973 were used for

estimation (Eq. 16). Details of these tests and

the laboratory evaluated undrained cohesion values obtained at the same locations

can be found in Brown (2007). These data resulted an average value of 1/

=0.3535

with a 0.162 standard deviation. High standard deviation suggests that this

value

may not support a strong prediction from equation 16.

VERIFICATION AND DISCUSSION

The estimated

and

values were used to analyze 2 sets of data to verify the

usefulness of the proposed 2 equations. Data used for this verification are presented

in Tables 1 and 2.

The laboratory values of

were compared to those predicted by Equation 14 in

Figure 1. Predictions by equations 1, 2, and 3 were also included in Figure 1 for

comparison. Overburden pressure correction proposed by Liao and Whitman (1986)

was used with Hatanaka and Uchida (1996) method. Performance of all equations

was compared by the distribution of error which was defined as the percent deviation

of the calculated friction angle from the measured. This comparison is presented in

Table 3. With the lowest average and standard deviation in error, statistically, the

proposed equation does a better estimation than other equations. It is also noticed that

for the given set of data, the proposed equation generates more conservative results

(slightly underestimate), while other methods overestimate

. However, it has to be

tested with more sets of data to see if it is a general trend.

Figure 2 compares the laboratory measured cu to those predicted by Equation 16.

Predictions by Terzaghi and Peck (1967) equation are also included in Figure 2. Hara

et al. (1974) equation was not considered in the analysis as the geological history of

the soil was not known to make a fair comparison. Statistical distribution of percent

errors by both proposed and Terzaghi and Peck (1967) equations are also presented in

Table 3. High standard deviation in percent error indicates a less reliable correlation.

However, the proposed equation (Eq. 16) still does a better estimation than Terzaghi

and Peck (1967) method. In addition, the prediction by the proposed equation is

conservative (slightly underestimates).

When

in equation 16 is replaced by the estimated value, it produces

cu=0.04paN60 which is different from Terzaghi and Peck (1967) only by the

proportionality constant (0.04 instead of 0.06). In a way the proposed method

supports what Terzaghi and Peck (1967) suggested, i.e. N60 is directly proportional to

cu. However, the high standard deviation indicates that both methods perhaps lack

details specific to cohesive soils such as overconsoldated ratio and in-situ moisture

content.

ASCE Geotechnical Special Publication No. 179, ISBN 978-0-7844-0972-5

Page

6

Table 1. Data Used for Verification of Eq. 14

Borehole

No.

N60

Depth

(ft)

′

(psf)

′

Lab

′

Eq.14

Borehole

No.

N60

Depth

(ft)

′

(psf)

′

Lab

′

Eq.14

SB-8

5

5

525

30

29.84

23

27

28

2900

30

29.74

SB-8

11

13

1525

31

28.45

32

11

6

660

30

31.76

SB-8

25

20

1998

35

30.90

32

11

14

1100

30

30.05

SB-22

8

8

880

30

29.63

32

26

23

1640

32

31.62

SB-22

9

13

1455

30

27.53

44

24

8

960

35

32.62

12

16

5

550

30

32.87

44

31

45

3240

30

29.86

12

13

8

715

32

31.97

44

45

50

3565

35

30.93

12

23

13

1015

30

32.44

49

11

13

1080

32

30.13

12

69

50

3420

35

32.21

49

23

17

1340

30

31.83

Note: Data from Commonwealth Associates Inc., Jackson, MI. Logs SB8 and SB22; Drilling by Braun

Intertec Corporation, in 2006, Circle Pines, MN, 75% efficiency assumed for automatic hammer. Logs

12, 23, 32, 44, 49; Drilling by American Engineering Testing Inc. in 2005, Farmington, MN, hammer

efficiency 60-65%.

Table 2. Data Used for Verification of Eq. 16

Borehole

No.

N60

Depth

(ft)

cu -lab

(psf)

cu -Eq.16

(psf)

Borehole

No.

N60

Depth

(ft)

cu -lab

(psf)

cu -Eq.16

(psf)

1

11

9

750

915

49

7

3

750

582

1

6

13

500

499

49

12

43

2000

998

2

7

2

500

582

49

10

48

1000

832

2

6

4.5

500

499

49

16

53

1500

1331

12

7

2.5

750

582

4066

10

4.5

750

832

12

12

26

1125

998

4066

17

7

1250

1414

32

9

4

1125

749

4066

31

10

2000

2579

44

16

4

1000

1331

4066

35

35

3000

2911

Note: Data from Commonwealth Associates Inc., Jackson, MI. Drilling by American Engineering

Testing Inc., logs 1 and 2 in 2001, Empire, MN, other logs in 2005, Farmington, MN, hammer

efficiency 60-65%.

Table 3. Statistical Comparison of Methods

Method

Average error (%)

Standard deviation of error (%)

Proposed (Eq. 14)

1.94

6.50

Wolff (1989)

-5.30

9.83

Kulhawy and Mayne (1990)

-37.62

13.43

Hatanaka and Uchida (1996)

-31.02

12.85

Proposed (Eq. 16)

2.90

23.12

Terzaghi and Peck (1967)

-40.08

33.35

It should be emphasized that the validity of a correlation depends highly on the

quality of data used. A close inspection of Figures 1 and 2 reveals that the laboratory

values tend to follow rounded off number pattern. Most of the friction angles are

either 300, 320, or 350 and the undrained cohesion values are either 500, 1000, or

2000 psf. It is unclear if it was a coincidence or a biased interpretation. Personal

ASCE Geotechnical Special Publication No. 179, ISBN 978-0-7844-0972-5

Page

7

communications with the drilling companies revealed that they have conducted some

direct shear tests and unconfined compressive strength tests. However, details of the

laboratory testing were not available with the borehole records.

25

30

35

40

45

50

25 30 35 40 45 50

Friction Angle_measured

Friction Angle_calculated

Proposed

Wolff (1989)

Kulhawy and Mayne (1990)

Hatanaka & Uchida (1996)

FIG. 1. Comparison of calculated friction angle with the measured.

0

500

1000

1500

2000

2500

3000

3500

4000

4500

0500 1000 1500 2000 2500 3000 3500 4000 4500

Cu_measured (psf)

Cu_calculated (psf)

Proposed

Terzaghi & Peck (1967)

FIG. 2. Comparison of calculated undrained cohesion with the measured.

ASCE Geotechnical Special Publication No. 179, ISBN 978-0-7844-0972-5

Page

8

CONCLUSIONS

The equations proposed to estimate shear strength properties in this manuscript

use a principle based on energy balance. SPT was treated analogous to driving a

miniature pipe pile. The energy input to the soil was used to correlate the SPT blow

count to the skin resistance which is a function of shear strength properties of the soil

at the depth of testing. Logical reasoning behind the proposed method makes it a

stronger prediction technique. A statistical analysis revealed that the proposed

method does a better estimation than the existing equations in predicting

from SPT

data. Undrained cohesion prediction for the set of data analyzed was not as strong as

the

prediction. However, the proposed prediction method suggests that the N60

should be directly proportional to undrained cohesion supporting Terzaghi and Peck

(1967).

REFERENCES

Brown, T.S. (2007). “Estimating shear strength properties of soils using SPT results,”

Graduate Project Report, Department of Civil Engineering, Lawrence

Technological University, Southfield MI.

Gibbs, H.J. and Holtz, W.G. (1957). "Research on determining the density of sand by

spoon penetration test," Proc. 4th ICSMFE, Vol. 1, pp. 35-39.

Hara, A., Ohta, T., Niwa, M., Tanaka, S., and Banno, T., (1974). “Shear Modulus and

Shear Strength of Cohesive Soils,” Soils and Foundations, Vol.14, No.3, pp.1-12.

Hatanka, M. and Uchida, A. (1996). “Empirical correlation between penetration

resistance and internal friction angle of sandy soils,” Soils and Foundations, Vol.

36, No. 4, pp. 1-9.

Japan Road Association (1990). Specifications for highway bridges, Part IV.

Kulhawy, F.H. and Mayne, P.W. (1990). Manual on estimating soil properties for

foundation design, Electric Power Research Institute, Palo Alto, CA.

Liao, S.S.C. and Whitman, R.V. (1986). “Overburden correction factors for SPT in

sand,” J. of Geotechnical Engineering, ASCE, Vol. 112, No. 3, pp. 373-377.

Meyerhof, G.G. (1956). “Penetration tests and bearing capacity of cohesionless

soils,” J. of Soil Mech. and Foundations Div., ASCE, Vol.82, No.SM1, pp.1-19.

Peck, R.B., Hanson, W.E., and Thornburn, T.H. (1953). Foundation Engineering,

John Wiley and Sons, pp. 222.

Peck, R.B., Hanson, W.E., and Thornburn, T.H.,(1974). Foundation Engineering, 2nd

ed., John Wiley and Sons, New York, NY.

Reese, L.C., Isenhower, W.M., and Wand, S.T. (2006). Analysis and Designing of

Shallow and Deep Foundations, John Wiley and Sons, pp.574.

Schmertmann, J.H. (1975). “Measurement of In-Situ Shear Strength", Proc., ASCE

Specialty Conference on In-Situ Measurement of Soil Properties, Vol. 2, Raleigh,

SC, pp. 57-138.

Terzaghi, K. and Peck, R.B. (1976). Soil Mechanics in Engineering Practice, 2nd ed.,

John Wiley and Sons, New York, pp. 729.

Wolff, T.F. (1989). “Pile capacity prediction using parameter functions,” ASCE

Geotechnical Special Publication No. 23, pp. 96-107.