THE STANDARD PENETRATION TEST

AND ITS USE IN GEOTECHNICAL ENGINEERING

Introduction  

The Standard Penetration Test (SPT) is a well-established and unsophisticated method, which was developed in the United States around 1927.  It is the most widely used in-situ soil characterization test.  It provides a measure of the resistance of the soil to penetration through the blow count "N", and a disturbed but representative soil sample that can be used for classification and index tests.

SPT has since undergone refinements with respect to equipment and testing procedure. The testing procedure varies in different parts of the world.  Therefore, standardization of SPT was essential in order to facilitate the comparison of results from different investigations. The method has been standardized as ASTM D 1586 since 1958 with periodic revisions to date [1].  The reliability of the method and the accuracy of the result depend largely on the experience and care of the engineer on site.

The report presents standardization of methodology in conducting the SPT.

The report also contains various developed correlations between the Standard Penetration Test N-values and soil strength, soil compressibility, foundation bearing capacity, foundation settlement, and liquefaction potential. In each case it is made clear what corrections should be applied to the measured N-values.

Methodology

There are various methods available for advancing boreholes.  During the advancement of the boreholes, soil samples are collected at various depths for further analysis.  One of the methods used is the wash boring method shown in Figure 1 below.

Figure 1. Wash boring [2] 

The Standard Penetration Tests is standardized through ASTM D 1586.  The sampling method being used in conducting SPT is the standard split-spoon (split-barrel) sampling.

A diagram of a split-spoon sampler is shown in Figure 2.  It consists of a tool-steel driving shoe at the bottom, steel tube (that is split longitudinally into halves) in the middle, and a coupling at the top. The steel tube in the middle has inside and outside diameters of 34 mm (1⅜ in.) and 50.8 mm (2 in.), respectively.  Figure 3 shows an unassembled split-spoon sampler. [2]

Figure 2. Diagram of standard split-spoon (split-barrel) sampler. [2]
Figure 3. Split-spoon sampler, unassembled. [2]

 

When the borehole is advanced to a desired depth, the drilling tools are removed.  The split-spoon sampler is attached to the drilling rod and then lowered to the bottom of the borehole.  The sampler is driven into the soil at the bottom of the borehole by means of hammer blows.  Figure 4 shows the various schematics of hammer arrangements.  

Figure 4. Schematic of various SPT hammer arrangements. [5]

 

The hammer blows occur at the top of the drilling rod.  The hammer weighs 623 N (140 lb).  For each blow, the hammer drops a distance of 0.762 m (30 in.).  The number of blows required for driving the sampler through three 152.4 mm (6in.) intervals is recorded. The sum of the number of blows required for driving the last two 152.4 mm (6 in.) intervals is referred to as the standard penetration number, N.  It is also commonly called the blow count. [2]

When the hammer is unable to drive the sampler the full 18 inches, refusal is reached and the test is halted.  Refusal is reached when  

  1.   50 blows are required for any 150-mm increment;

  2.  100 blows are obtained; and  

  3. 10 successive blows produce no advance. [1]  

After driving is completed, the sampler is withdrawn and the shoe and coupling are removed.  The soil sample collected inside the split tube is then removed and transported to the laboratory in small glass jars.  Determination of the standard penetration number and collection of split-spoon samples are usually done at 1m intervals for the first 6 meters and subsequently 1.5 m intervals after the 6 meter depth. [2]

Figure 5. Placing representative sample into a jar. [1]
Figure 6. Sample in a jar with identification. [1]

 

Disadvantages of the SPT test are: the sensitivity to operator techniques, equipment malfunctions and poor boring practice. It has been proven that operator techniques have a strong influence on the energy transfer and resulting N-values. Lower transferred energies will result in higher N-values. (Semi) automatic trip hammers make more consistent test results. However, there are many types of trip hammers each with a different design and resulting impact velocities. [6]

To overcome the variations in transferred energies a standard N60-value is defined (ASTM D4633).  The N60-value is the N value measured in the field (NF) adjusted by calibration to a reference energy Er of 60% (E60) or [6]:

N60 x E60 = NF x Emeasured                                                                                 (1)

 

Correlations

From investigations of cohesionless soil deposits, it is recognized that measured blowcounts will be affected by the soil depth being sampled.  For identical soils, the resistance to penetration at a deep location will be greater than the resistance value developed for a shallow location because of the influences of greater overburden and confining pressure and diferent driving energy losses in the drill rods. [1]

 

According to the correlation developed by Liao and Whitman (1986) [2], the field blowcounts, NF, can be corrected for the depth effect by applying a correction factor, CN, with

 (SI units)                                                                         (2a)

       (English units)                                                                   (2b)

where  is the effective soil overburden pressure expressed in kN/m2 and U.S. ton/ft2 (1 

U.S. ton = 2,000 lb) for equations (2a) and (2b), respectively.

 

According to the correlation developed by Skempton (1986) [2],

(SI units)                                                                                 (3a)

     (English units)                                                                 (3b)  

where  is the effective soil overburden pressure expressed in kN/m2 and U.S. ton/ft2 (1

 U.S. ton = 2,000 lb) for equations (3a) and (3b), respectively.

 Therefore [1],

                                                                                        (4)

Table 1 shows approximate correlations for the standard penetration number Ncor, and relative density, Dr.

                   Table 1. Approximate relationship between Ncor Value and Dr [2]

Corrected Standard Penetration Number, Ncor

Relative density, Dr (%)

0-5

0-5

5-10

5-30

10-30

30-60

30-50

60-95

 

Cubrinovski and Ishinara (1999) [2] proposed a correlation between NF and the relative density of granular soils (Dr) in the form

                                         (5)

where  is the effective soil overburden pressure expressed in kN/m2 and D50 is the sieve

 size through which 50% of soil will pass (mm).  

 

The drained angle of friction of granular soils, f, has also been correlated to the standard peneration number.  Peck, Hanson, and Thornburn (1974) gave a correlation between Ncor and f in a graphic form, which can be approximated as (Wolff, 1989) [2]

f (deg) = 27.1 + 0.3Ncor 0.00054Ncor2                                                         (6)

 Schmertmann (1975) also provided a correlation for NF versus .  After Kulhawy and Mayne

 (1990), this correlation can be approximated as

                                                             (7)

where pa is the atmospheric pressure (same unit as ).  

Table 2. Correlation Between Soil Conditions and Standard Penetration Test N-Value [3]

Soil

Designation*

N60, Blows/30cm

Sand and Silt

Loose

0-10

 

Medium

11-30

 

Dense

31-50

 

Very dense

Over 50

 

 

 

Clay

Very soft

0-2

 

Soft

3-5

 

Medium

6-15

 

Stiff

16-25

*Based on results for drive hammers with 60 percent efficiency. Source: Acker Drill Company

 

Correlations between the blow count N from the Standard Penetration Test and the angle of internal friction have been developed.  These correlations are presented in Table 3.  Due to the generalized nature of the correlation, any application of such data to final foundation designs should be made with caution.  

Table 3. Approximate Relationship Between N and f for Cohesionless Soil [3]

N Value*

Relative Consition of Soil

Approximate Value of f

10

Loose

30

20

Medium-dense

32

30

Medium-dense to dense

35

40

Dense

38

50

Dense to very dense

40

60

Very dense

32

*Values in this table refer to soil sampling procedures where the efficiency of the drop hammer is approximately 60%.

 

Approximate relationships between values of cohesion and N have been developed.  Values are presented in Table 4.  Due to the approximate nature  of the values, the data should be applied with caution.  

Table 4. Approximate Relationship Between N and Cohesion of Clays [3]

N Value

Relative Condition of Soil

Approximate Value of Cohesion, c

Psf

KN/m2 (kPa)

2-4

Soft

250-500

12-24

4-8

Medium

500-1000

24-48

8-15

Stiff

1,000-2,000

48-96

15-30

Very Stiff

2,000-4,000

96-190

> 30

Hard

> 4,000

> 190

*Values in this table refer to soil sampling procedures where the efficiency of the drop hammer is approximately 60%.

 

The unconfined compression strength (qu) of clay soils can also be approximately correlated to the standard penetration number N.  Table 5 gives the approximate relationship among the standard penetration number at a given depth, the consistency, nd the unconfined compression strength of clayey soils.  

Table 5. Approximate Correlation of SPT N-Value and Consistency of Clay [2]

N Value

Consistency

Unconfined Compression Strength

Psf

KN/m2 (kPa)

0

 

0

0

 

Very soft

 

 

2

 

500

25

 

Soft

 

 

4

 

1,000

50

 

Medium Stiff

 

 

8

 

2,000

100

 

Stiff

 

 

16

 

4,000

200

 

Very Stiff

 

 

32

 

8,000

400

> 32

Hard

>8,000

>400

 

It is important to point out that the correlation between N and qu given in Table 5 is only approximate.  The sensitivity St of clay soil also plays an important role in the actual N value obtained in the field.  In any case, for clays of given geology, a reasonable correlation between N and qu can be obtained, as shown in Figure 7.   

    Figure 7. Correlation between N and qu/pa (based on Djoenaidi, 1985) [2]

 

In this figure, the notation pa is the atmospheric pressure (in the same unit as qu).  For the data shown in Figure 7, the reported regression is given by (Kulhawy and Mayne, 1990).

                                                                                     (8)

 Sand deposits may experience liquefaction during an earthquake or other occurrence  that creates seismic shock; liquefaction results in a temporary loss of soils shearing strength.  Factors such as the ground acceleration created  by shock waves, soil density, and depth are known to have an influence.  The susceptibility to liquefaction has been related to the  corrected N value (the N value that includes depth and energy ratio corrections).  Sites where the N values are less than Ncritical (as calculated from the following expression) are indicating susceptibility to liquefaction [3]:

             (9)

 In this expression, use equal to 6 for an earthquake ntensty of VII, 10 for an earthquake

 intensity of VIII, and 16 for an earthquake intensity of IX.  The percent clay refers to the material finer than 0.002 mm.  The terms  and  refer to the depth in meters to the

 sand layer under study and to the groundwater table, respectively [3].

 

Generally, corrected N values less than 20 are considered to indicate that the site has a high potential for earthquake damage.  Sites where the considered N values are between 20 and 30 are classified as having the potential for an intermediate degree damage.  Sites where N values are greater tha 30 are considered locations where no significant damage from earthquake or seismic shock is expected [3].

Empirical relationships between the maximum shear modulus, Gmax, for sandy soils and N-value also exist.  Ohta and Goto (1976) and Seed et al. (1986) gave the relationship [4]

                                                           (10)

where Gmax and are in lb/ft2.

Imai and Tonouchi (1982) gave [4]

                                                                                          (11)

where Gmax is in kips/ft2.

 

Conclusion

In general, more than one empirical correlation is available to derive each particular soil property, and the accuracy of estimations is highly dependent on selection of the appropriate correlations for the respective material. Use of available laboratory soil test results, as well as any other available collateral information to select the right correlation formula for each soil material can increase significantly the confidence in the estimated parameters.

 

The standard penetration number is a useful guideline in soil exploration and assessment of subsoil condition, provided that the results are interpreted correctly.  Note that all equations and correlations relating to the standard penetration numbers are approximate.  Because soil is not homogeneous, a wide variation in the NF value may be obtained in the field.  For soil deposits that contain large boulders and gravel, the standard penetration numbers maybe erratic.

 

References

 

[1]        Bowles, Joseph E., Foundation Analysis and Design, 4th edition, 1988.

[2]        Das, Braja M., Principles of Geotechnical Engineering, 5th edition, 2002.

[3]        McCarthy, David F., Essentials of Soil Mechanics and Foundation, 5th edition, 1998.

[4]        Kramer, Steven L., Geotechnical Earthquake Engineering, 1996.

[5]        Schematic of various SPT hammer arrangements.htm

[6]        SPTC-Standard Penetration Test Control.htm
Hosted by www.Geocities.ws

1