KSCE Journal of Civil Engineering (2011) 15(6):1023-1031 Geotechnical Engineering

DOI 10.1007/s12205-011-0801-0
www.springer.com/12205

Applicability of the SPT-based Methods for Estimating Toe Bearing Capacity of

Driven PHC Piles in the Thick Deltaic Deposits
N. T. Dung*, S. G. Chung**, S. R. Kim***, and S. H. Beak****
Received March 12, 2009/Revised December 17, 2009/Accepted November 25, 2010

···································································································································································································································

Abstract

Standard Penetration Test (SPT)-based design methods for pile foundations have been extensively used in Korean practice.
However, their applicability for local application has seldom been examined, particularly for the thick deposits, such as in the
Nakdong River deltaic area. This paper examines the applicability of three common SPT-based methods to deep sandy deposits in the
delta. Routine SPT data, in which the test was completed for each of the N = 50 blows at less than 30 cm penetration in the last two
increments of 15 cm, were adopted. A special SPT was conducted to examine the general relationship between penetration and blow
counts in dense sands. Based on the special SPT and another well-documented case study, a simple linear extrapolation was
developed to estimate data equivalent to N > 50. PDA (Pile Driving Analyzer) tests were also carried out to evaluate energy
efficiency of the donut hammer used for the routine SPT. Energy correction factor (CE) was determined as 0.9. Using the corrected N-
values, the estimated SPT-based toe resistances were compared with data obtained from PDA tests, field load tests on piles, and a
CPT (Cone Penetration Test)-based method. Results indicate that the SPT-based methods generally manifest a rather low reliability.
The recommended Meyerhof’s method is deemed applicable for the preliminary design.
Keywords: toe resistance, standard penetration test N-value, PDA test, CPT, PHC pile
···································································································································································································································

1. Introduction tions. Furthermore, it is not known whether SPT-based methods,

with the use of routine practice data, could be reliably applied for
Standard Penetration Test (SPT) is one of the most common piles driven into deep sandy deposits, such as the Nakdong River
and economical field tests used during ground investigation. deltaic area.
Apart from its main applications in soil characterization, SPT N- The purpose of this paper is to examine the applicability of
value has also been extensively used for designing structural SPT-based methods in estimating toe bearing capacity of long
foundations and other earth structures, particularly, for the bear- PHC (Pretensioned spun High strength Concrete) piles driven in
ing capacity of piles (Meyerhof, 1976; Shioi and Fukui, 1982; the Nakdong River deltaic area. Initially, SPT and CPT tests were
Decourt, 1995; Robert, 1997). Thus, SPT-based methods are carried out at the Myeongji (MJ) and Shinho (SH) sites, from
included in a number of design codes or manuals (AASHTO, which several PDA (Pile Driving Analyzer) and static loading
2007; CFEM, 2006; FHWA, 1996), including in the Korean code tests on PHC piles were conducted at great length. PDA tests
(KGS, 2003). were also carried out to evaluate energy efficiency of the donut
In the last half century, a number of studies were carried out to hammer used for routine SPT, with their results consequently
evaluate the reliability of SPT, its influence factors, and other re- used to correct SPT N-values. Since the routine SPT was com-
lated criteria. Studies indicate that since the SPT N-values include pleted at each of the N = 50, even for less than 30 cm penetration
some uncertainties, they are appropriate for use in preliminary during the last two increments of 15 cm (N = 50 / <30), a special
designs (Schmertmann, 1978; Kulhawy and Mayne, 1990; Clayton, SPT was additionally conducted at the MJ site. This aimed to
1993). Few studies concentrated on the reliability of SPT-based examine the relationship between the penetration and blow
methods for estimating the bearing capacity of piles (Lee and counts, and to evaluate equivalent data for N/30. The toe bearing
Yun, 1992; Shariatmadari et al., 2006), and these studies likewise capacity of driven PHC piles was calculated using three common
indicate that the reliability is rather inconsistent and largely de- SPT-based methods, after which data were compared with those
pends on local practices, as well as on local geological condi- obtained from PDA tests, static loading tests, and the best CPT-

****Full-time Lecturer, Dept. of Civil Engineering, Dong-A University, Busan 604-714, Korea (E-mail: ntdung@dau.ac.kr)
****Member, Professor, Dept. of Civil Engineering, Dong-A University, Busan 604-714, Korea (Corresponding Author, E-mail: sgchung@dau.ac.kr)
****Member, Assistant Professor, Dept. of Civil Engineering, Dong-A University, Busan 604-714, Korea (E-mail: sungryul@dau.ac.kr)
****Member, Chief Director, GY Tech Construction Co., Ltd, Busan 612-020, Korea (E-mail: civilwhite@hanmail.net)

− 1023 −
 N. T. Dung, S. G. Chung, S. R. Kim, and S. H. Beak

based method for the sites (Dung et al., 2007). Some discussions 2.2 SPT-based Toe Resistance Methods
and recommendations are further presented in applications of A number of SPT-based methods were proposed to estimate
SPT-based methods for pile foundations in the deposits. the bearing capacity of driven piles. However, only three methods
commonly considered in practical design are taken into account
2. SPT N-value and Toe Resistance in this study, as shown in Table 1. As the SPT data in the clay and
upper sand layers are not available for evaluation of the shaft
2.1 SPT and its Corrected N-values resistance along the piles, only toe resistance is therefore consi-
According to ASTM D 1586 – 99 (similarly, with KSF 2307, dered for the lower sand layer.
2007), the performance of SPT is completed with the occurrence
of one of the following: (i) a total of 50 blows have been applied 3. Field Soil Tests
during any one of the three 0.15 m increments; (ii) a total of 100
blows have been applied; (iii) there is no observed advance of the 3.1 Locations and Ground Conditions
sampler during the application of 10 successive blows of the Figure 1 shows the locations of the study sites, the Myeongji
hammer; and (iv) the sampler advanced the complete 0.45 m (MJ) and Shinho (SH) residential complexes in Busan City. The
without the limiting blow counts occurring as described in the locations are situated at the coastline of the deltaic area. For land
first three conditions. The sum of the number of blows required development, a landfill of about 5m thick was constructed and
for the second and third 15 cm (6 in.) of penetration is termed as finally completed in the late 1990s (MJ site) and in mid-2000
“standard penetration resistance” or the “N-value”. (SH site). Housing construction only began a few years ago for
The normalized N-value to the standard energy ratio of 60% the MJ site.
and corrected for a number of effects is often expressed by the The ground conditions of the sites are briefly described as
following form (Skempton, 1986): follows. The fill layer is followed by loose silty sand (upper
sand), soft to medium silty clay (upper clay), loose to dense sand
N60 = NCE CB CS CR (1)
(lower sand), and sandy gravel on bed rock. One of its prominent
where CE, CB, CS, and CR are the correction factors for hammer features is a thin clayey silt layer that is mostly sandwiched in the
energy efficiency, borehole diameter, sampling method, and length lower sand layer. The thickness of the silty clay layer (upper
of drill rod, respectively. clay) varies from 16 to 20 m. Groundwater levels at the sites are
The N60-value is then corrected for overburden pressure effect approximately located 2.5 m below the ground surface.
as generally expressed by the following form:
3.2 The SPT and CPT Profiles
( N1 )60 = CN N60 (2) SPT was performed in boreholes of NX size, which were
where CN is the correction factor for effective overburden drilled using a rotary wash boring machine with drill rods of
pressure. There are several expressions for the CN in literature; Type A (41.3 mm outer diameter and 5.7 kg/m). The SPT was
however, the expression initially proposed by Liao and Whitman manually operated using a standard donut hammer-cathead
(1986) and later suggested in ASTM D 6066 - 96 is used for this system in which the rope was rolled 2.25 turns in the clockwise
study, which is as follows: direction. The common testing procedure was largely identical to
the ASTM standard D 1568-99; however, during the last two
CN = ( pa ⁄ σ 'v )
0.5
(3) increments of 15 cm, the test was completed whenever the blow
where σ 'v= vertical effective pressure at the SPT test point and pa number reached 50 even at a penetration of less than 30 cm (50 /
reference pressure = 100 kPa.

Table 1. Unit Toe Resistance from SPT-based Methods

Method Unit toe resistance Parameter
40 ( N1 )60 Db (N1)60 = normalized corrected
rt ( kPa ) = -------------------------- ≤ rl SPT number, Db = embedded
Meyerhof D
depth into bearing stratum, D
(1976) (1), (2) rl ( kPa ) = 400 ( N1 )60 for
= pile diameter, and rl = limit
Db ⁄ D ≥ 10 resistance.
Kb = 325 for sand, 205 for
Decourt
r ( kPa ) = Kb ( N60 ) sandy silt, 165 for clayey silt,
(1995)(1), (3) t
and 100 for clay.
Robert
rt ( kPa ) = Kb ( N1 )60 Kb = 190 for sand
(1997) (3)
Note: (1), (2) and (3) indicates CFEM (2006), KGS (2003), and GGEH
(2001) respectively, in which the methods are included. Fig. 1. Study Locations in the Nakdong River Delta

− 1024 − KSCE Journal of Civil Engineering

 Applicability of the SPT-based Methods for Estimating Toe Bearing Capacity of Driven PHC Piles in the Thick Deltaic Deposits

<30). Due to the thick clay layer known as “Busan clay” (Chung routine SPT, PDA test was additionally carried out at a neighbor-
et al., 2002), SPT was performed only in the lower sand layer (at ing site in the delta area where the same equipment and operators
least below 30 m), which was considered as the bearing stratum were employed. Calibration test procedures conformed to the
of piled foundations. ASTM D 4633-05. Table 2 provides the summarized results
At the same time, Cone Penetration Test (CPT) was also per- from the test.
formed at the sites using a 20-ton capacity CPT equipment. The
test was carried out using an electrical cone of 60o apex angle 4. Correction of the Measured N-values
with base cross-sectional and sleeve friction areas of 15 cm2 and
225 cm2, respectively. A porous element was mounted immedi- 4.1 Correction Factors for Energy Efficiency and Other
ately behind the cone shoulder to measure induced pore water Effects
pressure (u2). Average penetration rate was 20 mm/s, as recom- In this study, the correction factors CB, CS, and CR in Eq. (1)
mended by ISSMFE (1989). A total of 16 SPT - CPTU adjacent were taken as unity (Skempton, 1986) since the test was per-
tests were conducted at the two sites, in which the distance formed in relation to the borehole diameter of NX size, standard
between SPT and CPTU was less than 5 m at each test location. sampler, and drill rod larger than 30 m.
Fig. 2 shows the typical five SPT-N and CPTU-qc profiles at the Table 2 on the calibrated results for three types of hammers
sites, in which the first two profiles (MA1P-5, MC2-2) belong to obtained from the PDA test shows that the average Energy
the MJ site and the others (SO2-1, SO3-2, and SO5-3) to the SH Transfer Ratio (ETR) obtained from the tested donut hammer
site. varies from 53.9 to 55.9%. These ratios closely agree with those
obtained from other Korean studies (Lee et al., 1992; Park et al.,
3.3 Special SPT and PDA Tests 1994; Lee et al., 2005), which suggest that average energy
Apart from the main field soil tests, a special SPT was design- transfer ratio for donut hammers used in Korean application can
ed to examine the general relationship between penetration and be averagely taken as 49.5~53.3%. Based on the measured
blow counts until full penetration of 45 cm was achieved; it was results, an average ratio of ETR = 54% is obtained for this study.
carried out at the MD1P-2 location of the MJ site (Fig. 1). Drill Thus, CE = ETR/60 = 0.90 and N60 = 0.9 N.
rod penetration was continuously measured for each blow. For
this test, an automatic hammer system (hammer mass = 63.5 kg, 4.2 Extrapolation Technique for Terminated N-values
fall height = 76 cm) was used together with the drill rods of Type Results from the special SPT performed at MD1P-2 are shown
A connected to the standard sampler and for the casings of NX in Fig. 3 adjacent to the CPTU-qc profile. The test was carried out
size. Results are shown in Figs. 3 and 4, and consequently only in the denser upper part of the lower sand layer at intervals
analyzed. of 1 m. Fig. 4 shows two typical profiles of penetration-blow
For the calibration of the hammer energy efficiency in the counts where SPT N-values are larger than 50 at 33.5 and 35.5 m

Fig. 2. Typical SPT-N and CPTU-qc Profiles at the MJ and SH Sites: (a) MA1P-5, (b) MC2-2, (c) SO2-1, (d) SO3-2, (e) SO5-3

Vol. 15, No. 6 / July 2011 − 1025 −

 N. T. Dung, S. G. Chung, S. R. Kim, and S. H. Beak

Table 2. Calibrated Results from the SPT Equipment Used

Hammer type Safety Donut Automatic
Depth GL-13.5m GL-30.0m GL-9.0m GL-4.5m GL-1.0m
Boring No. BH3-68 BH3-76 BH3-102 BH3-106 BH3-114
Soil type Weathered soil Weathered rock Weathered soil Sandy gravel Sandy gravel

EMX Range 0.022~0.033 0.021~0.034 0.021~0.032 0.021~0.032 0.020~0.034

(tf·m) Avg. 0.027 0.026 0.027 0.026 0.027

ETR Range 45.6~68.4 43.5~70.5 43.5~66.3 43.5~66.3 41.4~70.6

(%) Avg. 55.9 53.9 55.9 53.9 55.9
Note: EMX (measured energy) = ∫ F ( t )V ( t ) dt ; ETR (Energy transfer ratio) = EMX/PE, where PE = potential energy of the hammer (PE = WH =
0.04826 tfm).

Fig. 4. Two Penetration-blow Count Profiles at MD1P-2

Fig. 3. SPT & CPT Profiles at MD1P-2

depth. Data indicate that after the seating drive zone (the first 15
cm), the relationships seem to be largely linear.
To verify the above result, a well-documented experimental
data is provided. Daniel (2000) experimentally performed a
comprehensive study on correlation between Large Penetration
Test (LPT) and SPT at Seward Site, Alaska, USA, where the
deposit consists of silty and sandy gravels in the flood plain. For
these tests, a safety hammer of 63.5 kg (fall height = 76 cm) was
used, and the blow counts were recorded at intervals of 1 in (2.54
cm). Fig. 5 shows three typical profiles of penetration-blow
counts, which indicate a similar trend to our study results (Fig.
4).
Fig. 5. Three Penetration-blow Count Profiles at Seward Site (Data
As shown in Figs. 4 and 5, the entire curves of penetration from Daniel, 2000)
versus blow counts generally exhibit a hyperbolic relationship
similar to the load-displacement curve observed from the plate
load test or pile static loading test. However, it is interesting to NLinear = 50 × 0.3 ⁄ d50 (4)
note that after passing the seating drive zone (the first 15 cm),
penetration versus blow count curves become approximately where NLinear = equivalent blow count per 0.3 m penetration, d50
linear. For N-values counted after the seating drive zone, a (m) = measured penetration when the test is terminated at N = 50.
simple linear extrapolation equation can thus be developed: The mechanism of driving the SPT sampler into dense sandy

− 1026 − KSCE Journal of Civil Engineering

 Applicability of the SPT-based Methods for Estimating Toe Bearing Capacity of Driven PHC Piles in the Thick Deltaic Deposits

deposits is similar with driving open-ended piles. Since the equivalent N-values larger than 50.
sampler is rather small and short (length = 9 times of diameter),
its total shaft resistance is significantly smaller than the total base 4.3 Linear Extrapolated N-values
resistance. In addition, due to cyclic loading, soil elements Figure 2 shows five typical soil profiles at the sites where the
around the sampler walls become fatigued and loosened (White, PDA and SLT were elaborately executed. In this analysis, the last
2005; White and Lehane, 2004); thus, shaft resistance decreases four locations (i.e., Figs. 2b~2e) where both PDA and CPT
with increasing number of blows. In dense sand, the hollow profiles were fully made available are taken into consideration.
sampler is quickly plugged during penetration in the seating drive Table 3 shows the measured N-values, at which the test was
zone; consequently, base resistance steadily increases. Similarly, completed at 50 blows with d50 < 30 cm, and consequently, with
total resistance increases steadily after the seating drive zone, the linearly extrapolated values (NLinear) based on Eq. (4). This
showing a slight nonlinear curve typical of those illustrated in indicates that the extrapolated N-values vary from 54 to 94.
Figs. 4 and 5.
From the recorded data, the penetrations at N = 50 blows (d50) 5. Toe Resistances
were obtained and the linear extrapolated values (NLinear) were
calculated using Eq. (4). The extrapolated values (solid symbols) 5.1 Toe Resistances based on CPT and Other Methods
are plotted together with the measured values, as shown in Figs. To examine applicability of SPT-based methods, toe resistances
4 and 5, which show that the extrapolated values are slightly obtained from CPT, PDA, and Static Loading Test (SLT), which
smaller than those measured at full penetration of 45 cm. The are known as the more reliable methods, were compared with
discrepancy between the measured and extrapolated values each other.
becomes larger in dense gravel sands; however, SPT for such
deposits is noted as less reliable. As a consequence, it can be 5.1.1 CPT-based Toe Resistances
stated that the linear extrapolated technique can conservatively It is known that the PDA-based method (CAPWAP analysis)
be used for practical design. It is applied herein to evaluate could be the most reliable field test used to determine pile
bearing capacity if SLT is not performed on piles. Dung et al.
Table 3. Linearly Extrapolated Values for N> 50 (2007) reported a comparison of shaft and toe resistances
Depth N/d50 NLinear obtained from various CPT-based and PDA-based methods at the
Borehole
(m) (blow/cm) (blow) sites. Results indicate that among 10 common CPT-based methods,
34.5 50/22 68 toe resistance obtained from the Aoki and De Alencar (1975)
36.0 50/17 88 method provided the best agreement for those obtained from the
PDA test. A further study carried out by Hung et al. (2008) also
MC2-2 37.5 50/21 71
drew the same conclusion based on a similar analysis using a
49.5 50/19 79 total of 172 PDA and CPTU test data (without adjacent SPT data
51.0 50/22 68 available). Therefore, the Aoki and De Alencar (1975) method
59.5 50/24 63 (hereafter denoted as “CPT-based”) are applied herein, the result
of which is compared with those from the SPT-based methods.
SO2-1 61.0 50/28 54
62.5 50/18 83
5.1.2 Toe Resistances from PDA and SLT
59.5 50/23 65 PDA test was performed for four closed-ended PHC piles of
SO3-2
61.0 50/22 68 600 mm in diameter. One pile was driven at MC2-2 location
42.0 50/25 60 (named here as pile MC2-2) and three others at the locations of
SO2-1, SO3-2, and SO5-3, and with similar names assigned. The
42.5 50/24 63
piles were successfully driven using a hydraulic impact hammer
43.0 50/23 65 with the maximum potential energy of 24 tf · m. PDA test was
43.5 50/19 79 systematically performed through the driving process from the
44.0 50/23 65 first until the final stroke. Piles were then re-struck 39 days, 42
days, 56 days and 59 days after the End of Initial Driving
SO5-3 44.5 50/23 65
(EOID). Further details on the PDA results were reported by
45.0 50/17 88 Dung et al. (2007).
45.5 50/16 94 Piles MC2-2 and SO5-3 were well-instrumented with strain
46.0 50/19 79 gauges to monitor development of residual loads with time. An
46.5 50/18 83 O-cell was installed at the bottom of the pile SO5-3. Static
Loading Test (SLT) and O-cell test were then applied to the piles
47.0 50/18 83
MC2-2 and SO5-3 at 164 days and 219 days after driving,
Note: the NLinear values were rounded to be integers. respectively. Further details of the two piles can be inferred from

Vol. 15, No. 6 / July 2011 − 1027 −

 N. T. Dung, S. G. Chung, S. R. Kim, and S. H. Beak

the reports of Kim et al. (2006), Dung (2008), and Fellenius et al. 5.2 SPT-based Toe Resistances
(2009). An additional instrumented pile was later installed at The toe bearing capacity of driven closed-ended PHC piles of
location MA1P-5 in the MJ site, which also included an O-cell at 600 mm diameter was estimated using the corrected N-values
the bottom of the pile. Further details of pile MA1P-5 can be (i.e., N60, (N1)60). SPT-based toe resistances from the three
inferred from Kim et al. (2008). Table 4 shows the toe and shaft methods were then plotted together with the values obtained
resistances obtained from the PDA, SLT, and O-Cell tests on the from the PDA-based, CPT-based, and SLT methods, as shown in
three instrumented piles. Fig. 6. It appears that the CPT-based resistances match rather
It should be noted that PDA-based toe resistances illustrated in well with the PDA-based values, except for some points at SO3-
Table 4 were taken from the EOID. It is because the hammer 2. The discrepancy was probably due to the heterogeneity of soil
energy was insufficient to fully mobilize the shaft resistance of in the horizontal direction at the site, which made the PDA-based
the long PHC piles that were subjected to large residual loads. resistances smaller in the middle part. The relatively good agree-
Toe resistances obtained from the re-strikes were therefore ment between CPT-based, PDA-based, and SLT results, as
smaller than with the EOID. As shown in Table 4, PDA-based shown in Table 4 and Fig. 6, further confirms previous studies
toe resistances matched relatively well with those obtained from (Dung et al., 2007; Hung et al., 2008); that is, both PDA-based
the SLT and O-Cell tests. It can also be inferred that if hammer and CPT-based design approaches are appropriate for the design
energy was sufficient to fully mobilize the piles during the re- of the driven PHC piles in the thick delta. The methods of
strikes, PDA-based toe resistances would have been closer to the Decourt (1995) and Robert (1997), which were based on SPT,
data on the SLT and O-Cell tests. This is in good agreement with appear to have produced larger and smaller toe resistances,
existing conclusions supporting PDA-based resistances, which respectively, in comparison with the PDA-based and CPT-based
can be reliably used for practical applications (Likins et al., 1996; values. The Meyerhof (1976) method appears to be the best
Likins and Rausche, 2004). among the SPT-based methods, giving close resistances not only
to the PDA-based, but also to the CPT-based values.

Table 4. Toe and Shaft Resistances Obtained from the Instrumented Piles
PDA-based resistance SLT-based resistance
Pile Length (m) Applied test
Rt (kN) Rs (kN) Rt (kN) Rs (kN)
MA1P-5 35.0* PDA, O-Cell 2520 930 2480 2830
MC2-2 35.0 PDA, SLT 2800 1430 3250 3680
SO5-3 56.6 PDA, O-Cell, SLT 3300 5570 3920 8640
*Including 3m excavation from the original ground surface.

Fig. 6. Toe Resistances from Different Methods: (a) MC2-2, (b) SO2-1, (c) SO3-2, (d) SO5-3

− 1028 − KSCE Journal of Civil Engineering

 Applicability of the SPT-based Methods for Estimating Toe Bearing Capacity of Driven PHC Piles in the Thick Deltaic Deposits

As shown in Fig. 6, while the Meyerhof (1976) method achieved

toe resistances comparable to those obtained from either PDA-
based and CPT-based methods, even at depths as deep as 60m
(Db/D ≥ 10), the method provided rather smaller results than
those from the two methods in the upper part of the lower sand
layer (i.e., in the embedded depths of Db/D < 10 into the bearing
stratum). Thus, it seems that smaller toe resistances have resulted
directly from the influence of the embedded depth ratio defined
for the unit toe resistance (Table 1). For example, in many cases
where we were unable to drive the piles into a very hard stratum,
toe resistance was negligible because Db ≈ 0. In reality, however,
piles could contribute a significant amount of toe resistance. To
avoid inconsistencies, limited unit resistance was applied to the
whole bearing stratum (the lower sand layer), and thus, unit toe
resistance can be simply given as follows (denoted as the
modified Meyerhof method):
rt (kPa) = 400(N1)60 (5)
Fig. 8. Rt-SPT versus Rt-PDA
Figure 7 shows a comparison between the toe resistances
obtained from the PDA- and CPT-based methods, the original
and modified Meyerhof (1976) methods, for the depths up to the linearly interpolated from the two closest Rt-PDA values. Both
embedded depth (Db) of 10 times pile diameter (D) into the comparisons generally indicate that the SPT-based values tend to
lower sand layer. It appears that the modified toe resistances be poorly correlated with the CPT and PDA-based methods.
were closer to the CPT-based values than with the original Particularly, it seems that from Figs. 6, 8, and 9, the Decourt
values. (1995) method strongly overestimates the toe resistances as
compared with the PDA- and CPT-based values. Meanwhile, the
5.3 Reliability of the SPT-based Methods Robert (1997) method underestimates the values, especially in
To examine the reliability of the SPT-based methods, SPT- dense sands. The Meyerhof (1976) method seems to be better, as
based toe resistances (Rt-SPT) versus PDA-based (Rt-PDA) and CPT- compared to the other two methods.
based (Rt-CPT) values were plotted; they are shown in Figs. 8 and To be quantitatively clear, Table 5 is given to show the geomet-
9, respectively. When the tested depths were different between ric mean values of Rt-SPT/Rt-PDA and Rt-SPT/Rt-CPT, and the coeffici-
the PDA and SPT, the Rt-PDA value at the level of the SPT was ents of determination [R2 = 1 - (the sum of square errors)/(the

Fig. 7. Toe Resistances from the Original and Modified Meyerhof (1976) Methods: (a) MC2-2, (b) SO2-1, (c) SO3-2, (d) SO5-3

Vol. 15, No. 6 / July 2011 − 1029 −

 N. T. Dung, S. G. Chung, S. R. Kim, and S. H. Beak

cal design manuals. The Meyerhof (1976) method is applicable

to preliminarily determine the toe bearing capacity of driven piles,
provided that SPT N-values are reliably obtained. In addition,
practicing engineers may use the limited unit toe resistance, Eq.
(5), for sand layers in preliminary designs.

6. Conclusions

This study evaluated whether three common SPT-based methods

for toe bearing capacity of piles are applicable for deep sandy
deposits in the Nakdong River estuary area in Busan City, Korea.
To achieve this, routine SPT were carried out in two sites. The
test was terminated whenever the blow number reached 50 in the
last two 0.15 m increments. A special SPT was performed to
develop a method for extrapolating terminated N-values. In addi-
tion, PDA tests were also conducted to determine energy correc-
tion factors for the routine SPT. The corrected N-values (i.e. for
N60 or (N1)60) were used to estimate the toe bearing capacity of
Fig. 9. Rt-SPT versus Rt-CPT
driven PHC piles. Results were compared with those obtained
from the PDA, SLT, O-cell tests, and a CPT-based method. The
total sum of squares)]. It should be noted that the coefficient of conclusions drawn from this study are as follows:
determination was determined based on the equal line (Rt-PDA = 1. From the special SPT and another well-documented case
Rt-SPT; Rt-CPT = Rt-SPT), where Rt-PDA and Rt-CPT are considered as study, an approximate linear relationship between penetration
basic values for the comparison. It appears that the reliability of and blow counts was found for the last two 15 cm increments,
SPT-based toe resistances is very low since most of the coeffici- which was used for extrapolating the terminated N-values of
ents were negative. It is probable that the high overestimation 50. Linear extrapolated values (NLinear) appear to be slightly
have resulted from the Decourt (1995) method, which was smaller than those obtained from the measured data. It was
mainly due to the use of the normalized SPT number, N60, rather also found that energy efficiency of the donut hammer used at
than the normalized corrected one, (N1)60. That is, N60 value does the sites was on average 54% of the theoretical free-fall
not include the factor CN which becomes as small as 0.5 in the energy. Thus, energy correction factor was determined as CE
deep deposits as the cases presented herein. However, the = 0.90.
underestimation resulting from the Robert (1997) method may 2. In general, the toe bearing capacity estimated from the SPT-
have come directly from the use of the rather small ratio of rt / based methods has low reliability in comparison with those
(N1)60 = 190. This fact was also highlighted by Poulos (2001). obtained from CPT-based methods, PDA, SLT, and O-cell
Therefore, it is not recommended that these two methods be used tests. The Decourt (1995) and Robert (1997) methods seem to
for thick and inhomogeneous deposits, such as the Nakdong highly overestimate and underestimate the toe bearing cap-
River deltaic area. acity, respectively; unlike the Meyerhof (1976) method, which
As opposed to the Decourt (1995) and Robert (1997) methods, particularly provided closer values to those obtained from the
the Meyerhof (1976) method seems to produce relatively reason- CPT-based, PDA-based methods, and field load tests.
able toe resistances with mean values of Rt-SPT/Rt-PDA and Rt-SPT/Rt- 3. A modified Meyerhof method was proposed based on other
CPT are 1.04 and 0.87, respectively. However, the coefficient of reliable methods that were adopted herein; that is, the limited
determination remains very low (i.e., R2 = 0.0044 and 0.4434, unit resistance (rl) is considered for the whole bearing stratum
respectively), which indicates that even the Meyerhof method instead of using embedded depth ratio to up to Db /D = 10 in
has low reliability compared to the CPT- and PDA-based methods. the Meyerhof (1976) method. It is recommended that the
Nevertheless, this result supports data mentioned in many practi- modified Meyerhof method be used for the toe bearing
capacity of the driven PHC piles for preliminary design,
Table 5. Mean Values of toe Resistance Ratios and Correlation particularly, in the thick deltaic deposits.
Coefficients
Rt-SPT/Rt-PDA Rt-SPT/Rt-CPT Acknowledgements
Method 2 2
Mean R Mean R
This work was supported by the Korea Science and Engineer-
Meyerhof (1976) 1.04 0.0044 0.87 0.4434
ing Foundation (KOSEF) NRL Program grant funded by the
Decourt (1995) 1.58 -0.1887 1.42 -0.0004 Korea government (MEST) (No. R0A-2008-000-20076-0), and
Robert (1997) 0.54 -0.9316 0.46 -0.4829 partly by Dong-A University, Busan, Korea.

− 1030 − KSCE Journal of Civil Engineering

 Applicability of the SPT-based Methods for Estimating Toe Bearing Capacity of Driven PHC Piles in the Thick Deltaic Deposits

References Kim, S. R., Chung, S. G., and Lee, B. Y. (2008). “Analysis of a bi-
directional load test result on long PHC piles in consideration of
AASHTO (2007). Standard specifications for highway bridge, 18th residual load.” Journal of the Korean Geotechnical Society, Vol. 24,
Edition, Washington, D.C. No. 6, pp. 85-93.
Aoki, N. and de Alencar, D. (1975). “An approximate method to KSF 2307 (2007). Test method for standard penetration, Korean
estimate the bearing capacity of piles.” Proceeding, the 5th Pan- Industrial Standards.
American Conference of Soil Mechanics and Foundation Kulhawy, F. H. and Mayne, P. H. (1990). Manual on estimating soil pro-
Engineering, Buenos Aires, Vol. 1, pp.367-376. perties for foundation design, Electric Power Research Institute, EPRI.
ASTM D6066 (1996). Standard practice for determining the normaliz- Lee, M. H., Kang, W. T., Lee, W. J., and Kim, Y. J. (1992), “Evaluation
ed penetration resistance of sand for evaluation of liquefaction of the falling velocity of SPT hammer via actual measurement”,
potential, American Society for Testing and Materials, ASTM. Geotechnical Engineering, Korean Geotechnical Socieity, Vol. 8,
ASTM D1586 (1999). Standard test method for penetration test and No. 1, pp. 59-66.
split-barrel sampling of soils, American Society for Testing and Lee, C. H. and Lee, W. J. (2005), “Ratio of hammer energy and dynamic
Materials, ASTM. efficiency of standard penetraion test.” Journal of Korean Geotech-
ASTM D4633 (2005). Standard test method for energy measurement for nical Society, Vol. 12, No. 9, pp. 5-12.
dynamic penetrometers, American Society for Testing and Materials, Lee, M. H. and Yun, S. J. (1992). “Comparison of the methods used in
ASTM. determining the pile design load.” Proceedings of the KGS (Korean
Chung, S. G., Giao, P. H., Kim, G. J., and Leroueil, S. (2002). “Geotech- Geotechnical Society)Geotechnical Engineering Conference, pp.
nical characteristics of Pusan clays.” Canadian Geotechnical 69-102.
Journal, Vol. 39, No. 5, pp. 1050-1060. Liao, S. S. C. and Withman, R. V. (1986). “Overburden correction
CFEM (2006). Canadian Foundation Engineering Manual, 4th edition, factors for SPT in sand.” Journal of Geotechnical Engineering,
Canadian Geotechnical Society, 485p. ASCE, Vol. 112, No. 3, pp. 373-377.
Clayton, C. R. I. (1993). The Standard Penetration Test (SPT): Methods Likins, G. E. and Rausche, F. (2004). “Correlation of CAPWAP with
and use, Report Prepared under Contract to CIRIA by the University static load test.” Proceeding of the 7th International on Application
of Surrey, p. 130. of Stresswave Theory to Piles, Petaling Jaya Malaysia, pp. 153-165.
Daniel, C. R. (2000). Split spoon penetration testing in gravels, Master Likins, G. E., Rausche, F., Thendean, G., and Svinkin, M. (1996).
Thesis, The University of British Columbia, 187p. “CAPWAP correlation studies.” Proceeding of the 5th International
Decourt, L. (1995). “Prediction of load-settlement relationships for on Application of Stresswave Theory to Piles, University of Florida,
foundations on the basis of the SPT-T.” Ciclo de Conferencias Inter. Orlando Florida, USA. pp. 447-464.
“Leonardo Zeevaert”, UNAM, Mexico, pp. 85-104. Meyerhof, G. G. (1976). “Bearing capacity and settlement of pile
Dung, N. T. (2008). Evaluation of design parameters and performance foundations.” Journal of Geotechnical Engineering, The Eleventh
of driven pile foundations in the thick delta deposits, PhD Thesis, Terzaghi Lecture, ASCE, Vol. 102, GT3, pp. 195-228.
Dong-A University, p. 236. Park, Y. W., Lee, H. C., and Park, J. H. (1994). “A comparison of SPT
Dung, N. T., Chung, S. G., and Kim, S. R. (2007). “Comparative study hammer energy.” Proceeding of the Korean Society of Civil
between design methods and pile load tests for bearing capacity of Engineers Conference, pp. 821-824.
driven PHC piles in the Nakdong River delta.” Journal of the Poulos, H. G. (2001). “Pile foundations.” Chapter 10, Geotechnical and
Korean Geotechnical Society, Vol. 23, No. 3, pp. 61-75. Geoenvironmental Engineering Handbook, Edited by R.K. Rowe,
Fellenius, B. H., Kim, S. R., and Chung, S. G. (2009). “Long-term moni- Kluwer Academic Publisher, p. 1088.
toring of strain in instrumented piles.” J. of Geotechnical and Geo- Robert, Y. (1997). “A few comments on pile design.” Canadian Geo-
environmental Engineering, ASCE, Vol. 135, No. 11, pp. 1583-1595. technical Journal, Vol. 34, No. 4, pp. 560-567.
FHWA (1996). Design and construction of driven pile foundations, US Schmertmann, J. H. (1978). “Use of SPT to measure soil properties? –
Department of Transportation, Federal Highway Administration, Yes, But..!” Dynamic Geotechnical Testing, (STP 654), ASTM, pp.
Report No. FHWA-HI-97-013. 341-355.
GGEH (2001). Geotechnical and geoenvironmental engineering hand- Shariatmadari, N., Eslami A., and Karimpour, M. (2006). “A new
book, Edited by R.K. Rowe, Kluwer Academic Publisher, p. 1088. approach to estimate the base bearing capacity of driven piles based
Hung, L. C., Kim, S. R., and Chung, S. G. (2008). “Applicability of on SPT data.” Proceedings of the 31st Annual Conference on Deep
CPT-based toe bearing capacity of driven PHC piles.” Proceeding of Foundations, DFI, Washington, D.C.
the KGS (Korean Geotechnical Society) Geotechnical Engineering Shioi, Y. and Fukui, J. (1982). “Application of N-value to design of
Conference, Kwangju, pp. 792-798. foundations in Japan.” Proceeding of the Second European Sym-
ISSMFE (1989). “International reference test procedure for cone posium on Penetration Testing, Amsterdam, Vol. 1, pp. 159-164.
penetration test (CPT).” Report of the ISSMFE Technical Committee Skempton, A.W. (1986). “Standard penetration test procedures and the
on Penetration Testing of Soils – TC16, with Reference to Test effects in sands of overburden pressure, relative density, particle
Procedures, Swedish Geotechnical Institute, Linkoping, Information, size, aging and overconsolidation.” Geotechnique, Vol. 36, No. 3,
7, pp. 6-16. pp. 425-447.
KGS (2003). Design criteria of structure foundations, Korean White, D. J. (2005). “A general framework for shaft resistance on dis-
Geotechnical Society. placement piles in sand.” Proceeding of the 1st International
Kim, S. R., Chung, S. G., and Dung, N. T. (2006). “Determination of Symposium on Frontiers in Offshore Geotechnics (ISFOG-2005),
true resistance from load transfer test performed on a PHC pile.” Gourvenec and Cassidy (Eds.), Taylor & Francis Group, pp. 697-703.
Journal of the Korean Geotechnical Society, Vol. 22, No. 11, pp. White, D. J. and Lehane, B. M. (2004). “Friction fatigue on displacement
113-122. piles in sand.” Geotechnique, Vol. 54, No. 10, pp. 645-658.

Vol. 15, No. 6 / July 2011 − 1031 −