17th International Conference on Soil Mechanics & Geotechnical Engineering, Alexandria, Egypt, 2009

Rapid load field tests interpreted with the new guideline

Les essais de charge rapide interprte avec la nouvelle directive

P Hlscher
Deltares, the Netherlands
H.E. Brassinga
Public Works Rotterdam, the Netherlands
A.F. van Tol
Deltares , Delft University of Technolog, the Netherlands
P. Middendorp E. Revoort .
Profound, the Netherlands Funderingstechnieken Verstraeten BV, the Netherlands

ABSTRACT
In order to test the draft guideline for the interpretation of a Rapid Load Test, a field test was carried out. Two instrumented piles were
driven. Both were tested statically and rapidly. The strain in the piles and the excess pore water pressures during a rapid load test are
measured. The results are compared. The draft version of recently developed guideline is used for the interpretation of the rapid test
results. The rapid test offers good insight in the static behavior. The application of theoretical sound methods on this practical case
gives information on required improvements. The rate factor in sand seems to be defined by the dilatational behavior and the
generation of negative excess pore water pressure during a rapid load test.

RSUM
Afin d'examiner la version provisoire de la directive pour l'interprtation d'un essai de charge rapide, un essai sur le terrain a t
effectu. Deux pieux quips ont t enfoncs. Tous les deux ont t examins statiquement et rapidement. Les rsultats sont
compars. La version provisoire de la directive dveloppe rcemment est employe pour l'interprtation des rsultats d'essai rapides.
L'essai rapide offre la bonne perspicacit dans le comportement statique. L'application des mthodes thoriques sur ce cas pratique
fournit l'information sur des amliorations exiges. Le facteur de vitesse en sable semble tre dfini par le comportement dilatational
et la gnration de la pression excessive d'eau interstitielle pendant un essai de charge rapide.

Keywords : pile, bearing capacity, field test, static load test, rapid load test, regulations

1 INTRODUCTION

The Rapid Load Test seems to be a good and economical
alternative for the Static Load Test on piles. Examples of this
test method are the Statnamic (Janes et al, 1991) and Pseudo-
Static Pile Load Tester (Schellingerhout & Revoort, 1996).
However, in Europe the application of such tests is hindered by
the discussion about the interpretation of the test. This
hindrance can be overcome by proper regulation of the test. An
international project has been started with the objective of
developing a Standard for the execution of a Rapid Load Test
and Guidelines for the interpretation of the test (Hlscher,
2009).
The aim of a load test is to proof that a foundation fulfills the
specified requirements. Normally, these requirements are
specified in terms of the stiffness under working load and the
bearing capacity in ultimate limit state. Principals are interested
in these data, since they want to be sure that the installed
foundation is adequate, whereas contractors are interested in
showing that their product fulfills the contractual requirements.
In order to prove the reliability of the calculation rules in the
guideline, test results are required. These results should deliver
both the results of a static load test (SLT) as well as a rapid load
test (RLT). Since the results of a RLT needs some post
processing, the full digital measurement results must be
available. According to the standard, the load on the pile head,
the displacement and acceleration of the pile head must be
measured with high rate.
This paper describes such a test, and discusses the
application of the guideline on the test results.
2 NEW STANDARD AND GUIDELINE
2.1 Short description of research
The Dutch part of the research project exists of two parts:
The first part focuses on the interpretation of the RLT on piles
that take their bearing capacity from a deep sand layer, which is
the typical Dutch case. The theoretical studies are presented in
(Huy, 2008) and (Hlscher & van Tol, 2008).
The second part focuses on the creation and validation of
documents, which help the engineer in practical cases. These
documents are created in discussion with an international expert
group, with the objective to formulate internationally accepted
documents. The expert group met two times in Delft, discussing
the recent progress in research and the content of the document
(Hlscher & van Tol, 2008) and at the Stress Wave Conference
in Lisbon, 2008. Further discussion is done by e-mail.
The field test described in this paper, is part of the research.
The test is used to validate the theoretical model for the
influence of the pore water pressure and deliver a field test of
the quality desired for validation of the methods described in the
guideline.
2.2 Standard for test execution
The proposed Standard regulates the execution of a Rapid Load
Test. It defines the minimum requirements for proper testing
and reporting. A test executed in accordance with the Standard
can be considered as a reliable test, even though the
measurement data are still subject to interpretation. The draft of
this Standard is presented in Hlscher & van Tol (2008). The
formal document will be further proceeded by CEN TC 341,
working group 4.

2.3 Guideline for interpretation

The guideline discusses the aspects that should be considered

while interpreting a RLT. It assumes that the test is carried out
in accordance with the requirements of the standard.
Three method of interpretation are described:
The Sheffield method (Brown & Hyde, 2006)
The unloading point method (Middendorp, 1992)
The variable damping method (Matsumoto et al, 1994)
The Sheffield method requires some laboratory testing for the Figure 2 The Statnamic apparatus (RLT) and the counter weight (SLT)
rate dependency of the soil, which is very important for piles in
clay. The other two methods can be applied without laboratory 3.3 Results of static load test
testing.
Figure 3 shows the load displacement curves for the head of
the test piles from SLT. From these curves it was concluded the
3 FIELD TEST AND STATIC LOAD TESTS ultimate bearing capacity was 1150 kN for pile 1 and 1120 kN
for pile 2.
0 200 400 600 800 1000 1200
3.1 Site description
0
load (kN)
The test location is situated in Waddinxveen, in the western part
of the Netherlands. Figure 1 shows the result of a CPT test at 20
the location of the test site. The surface level lies at NAP -5.07 pile 1
m (NAP = Dutch reference level). Below surface, holocene peat pile 2 pile 1
and clay layers are found until NAP -12 m. Under these 40
holocene layers, the Pleistocene sand layer is found. This layer
consists mainly of medium fine sand. The freatic ground water pile 2
level and the piezometric head in the Pleistocene sand layer are 60
found at about NAP -6 m.
pile head
0 5 10 15 20 displacement (mm)
-5 80

cone resistance (MPa)

Figure 3 Load displacement diagrams of SLT
-10

3.4 Results of strain measurements

-15
During SLT the strains are measured at 3 levels (see Section
3.2).
-20 Figure 4 shows the forces in pile 2 measured by the strain
gauges. The load measured during SLT and the force estimated
depth (m NAP) from the strain gauges correspond well. The shaft carries about
-25 15% of the total load.

Figure 1 Cone resistance at site

1200

1000 pile head

3.2 Test execution
800
Force from strain [kN]

top sand layer

Two precast concrete piles were driven to a level of NAP -15.35
600
m. The dimensions of the piles were 350 * 350 mm2. pile toe
Both piles were instrumented with two strain gauges at three 400

levels: 0.5 m below the head of the piles, at the transition

200
between Holocene and Pleistocene layers (NAP -13.95 m) and
at 0.5 m above the toe of the piles. At the toe of both piles a 0
BAT sensor was installed to measure the pore pressures during
the Statnamic tests. In addition, a piezocone was installed in the -200
0 120 240 360 480 600 720
Pleistocene sand layer near pile 2 at NAP -15.35 m. Time [min]
Both piles were tested statically and in a Statnamic test. Test Figure 4 Forces from strain measurements during SLT pile 2
pile 1 was first tested by means of the Statnamic method,
followed by the static test, while for test pile 2 this sequence The strain gauges in the head of pile 1 did not function well.
was reversed. The distance between the piles was about 6 m. Therefore, Figure 5 shows the results for the toe only. For pile
The static load tests were executed according to NEN 6745. The 1, the friction along the shaft in the sand is higher than the
predicted ultimate bearing capacity was about 1350 kN. The friction along the shaft for pile 2. The value measured for pile 1
piles were loaded by means of a hydraulic jack using water is in accordance with our expectation.
tanks as ballast. The load step varied between 140 kN for test
pile 1 and 175 kN for test pile 2.
 the pile head corrected for the inertia force (pile mass *
1000
acceleration) at the unloading point. Pile 1 just reached almost
failure (displacement is 10% of equivalent pile diameter).
800
According to the guideline, the measured soil resistance
Force from strain [kN]

sandlayer (corrected for intertia, see Figure 7) must be corrected with a

600 factor (R) for rate effects. Based on empirical data the draft
pile toe guideline suggests for sand a mean value of R = 0.94 (with
400 coefficient of variation 0.15), Hlscher & van Tol, 2008). The
derived bearing capacity of this pile is 0.94*1232 = 1158 kN.
200 Pile 2 did not reach failure. It can be concluded that the bearing
capacity will be higher than 1137 (0.94*1210).
0
Based on the RLT, pile 2 has a higher capacity than pile 1.
0 120 240 360 480 600 720 840 This is opposite the observation during the SLT.
Time [min] Regarding the stiffness, the measurements from pile 1 have
Figure 5 Forces from strain measurements during SLT pile 1 some strange points in the beginning. Ignoring these points, the
measured stiffness is presented in Table 1

4 RESULTS OF RAPID LOAD TESTS, COMPARISON Table 1 Stiffness (in kN/mm) at working load values (in kN)
test SLT SLT SLT RLT RLT
load [kN] 400 800 800 400 800
4.1 Measurements during Rapid Load Test case initial 1 hour
pile 1 244 143 120 -- 103
The RLT is carried out by using a 4 MN Statnamic apparatus of pile 2 256 120 97 107 113
Profound (Profound, 2009). According to the draft standard, the
load, displacement and acceleration are measured at the pile The stiffnesss at 800 kN load are in good agreement. For
head. Additionally, the strains in the piles are measured at 3 stiffness a correction for rate effect in sand is not required, since
levels (see Section 3.4) and the pore water pressure under the at the start of the load test, the pore water pressure plays a minor
pile toe and in the soil close to the pile. role in the behaviour of the sand around the toe.
Figure 6 shows the results of the RLT on pile 2, the fourth
(last) load step. The Figures shows the load on the pile Soil resistance in unloading point [kN]
(measured and corrected for inertia of the pile), the
Displacement in unloading point [mm]

0 500 1000 1500

displacement, the acceleration and the rapid load-displacement 0
diagram. For these piles, the inertia correction is very small.

-10
pile 1
pile 2
.

-20

-30

-40
Figure 7 Unloading points for both piles (with correction for inertia,
without correction factor for rate effects)

A RLT with more load steps gives a reasonable overview of

Figure 6 Results of RLT pile 2 load step 4 (see text for legend).
4.2 Bearing capacity based on guideline
The measurements will be interpreted according the draft
guidelines.
The Sheffield method is not studied in detail, since this
method is especially meant for piles in clay and the required
data (from laboratory experiments) are not available. The rate
effects in sand are small, so the Sheffield method leads to small
corrections.
The applicability of the variable damping method is hindered
by the strong dependency of the method on the initial part of the
measurement. This part of the interpretation method must
therefore be improved in the draft guideline.
The unloading point method focuses on the bearing capacity,
the soil force at the unloading point. Figure 7 shows the
calculated unloading points for both piles at all load steps. The
vertical axis shows the displacement at the unloading point (the
moment of zero velocity). The horizontal axis shows the load on
the full load displacement curve. However, at higher rates the
rate effect must be included. At this moment, no reliable method
is described in the draft guideline. A method with a smooth
transition between uncorrected values at low velocity and more
correction at higher velocity (such as the Sheffield method for
clay) is required.
4.3 Strain measurements
Figure 8 shows the forces in pile 2 during the RLT load step 4.
No influence of wave propagation is observed. These results can
be compared with the results of the SLT in Figure 4. The
distribution of the shaft and toe resistance differs, during the
RLT the shaft carries more load. This might be due to the rate
effect in the soft upper layer. It is unknown why the forces from
the strain measurements are 25% higher than the values
measured at the pile head. Common explanations (such as rate
effect in the concrete or eccentricity of the load) do not lead to a
higher value from the strain gauges.
4.4 Pore water pressure measurements
The theoretical research showed that excess pore water pressure
plays a role in the interpretation of RLT on piles in sand (Huy,
2008). In earlier research, pore water pressures are measured in
the sand close to the pile toe (Hlscher, 1995) (Maeda, et al
1998). The field test should deliver information on this
influence.
Both piles had a pore water transducer at the toe. However,
during the first RLT (on pile 1), it was observed that the pore
water transducer was not fast enough to follow the real
pressures. Therefore, these measurements are disregarded here.
An additional pore water transducer had been installed near the
pile toe of pile 2 before the RLT. The distance was about
0.70 m from the pile toe.
pile 2, loadstep 4
1600
Mean force [kN]
1200
800
400
0 Hlscher, P., van Tol, A.F. eds. 2008 Rapid Load Testing on piles, CRC
0.1 0.2
Time [s]
Figure 8 Forces in the pile from strain measurement (pile 2 load step 4)
Figure 9 shows the results of the dynamic pore water
transducer and the force in the pile toe during the fourth load
step. Due to the distance between the transducer and the pile
toe, the change in excess pore water pressure lags behind the
force in the pile toe. However, the pore water pressure decreases
before the force reaches its maximum value. During unloading
of the pile toe, the pore water pressure decreases strongly. The
hydrostatic water pressure is 95 kPa. The minimum value of the
excess pore water pressure is -76 kPa.
1000
force in pile toe
800
Force in pile toe [kN]
pore water pressure in soil
600
400
200
0 -100
0.1 0.2 0.3 0.4
Time [s]
Figure 9 Force and excess pore water pressure in the sand during RLT
(Pile 2, load step 4)
The decrease of excess pore water pressures during the
loading phase of the rapid load test shows that dilatancy of the
sand around the pile toe plays a role. This phenomenon is
observed in centrifuge testing (Huy, 2008). This effect explains
the rate factor that is observed in sand.
5 CONCLUSIONS
Two instrumented driven piles are tested by both SLT and RLT.
The measurements lead to the following conclusions:
The SLTs show a bearing capacity of 1120 kN respectively
1150 kN. The RLTs show a derived bearing capacity of
> 1137 kN respectively 1158 kN.
The stiffness measured by RLT is practically useful.
The distribution between shaft and toe in a RLT differs
from a SLT, presumably due to rate effects in the soft soil.
Excess pore water pressures are measured close to the pile
toe.
The upm-method in the draft guideline gives information on
stiffness and bearing capacity, for derivation of the full
derived load-displacement diagram additional development
is required
The variable damping method must be developed further..
ACKNOWLEDGEMENTS
The Delft Cluster and all companies involved are kindly
acknowledged for their contribution to carry out these tests.
REFERENCES
pile head
sand layer Brown, M.J. & Hyde, A.F.L. 2008. Rate effects from pile shaft
pile toe resistance measurements. Canadian Geotechnical Journal, 45(3):
425-431
Hlscher, P. 2009, project-information: www.rapidloadtesting.eu,
visited February 2009
Hlscher, P. 1995. Dynamical response of saturated and dry soils. PhD
thesis, Delft University of Technology, Delft, the Netherlands
0.3 0.4 Press, ISBN 978-0-415-48297-4, Leiden, the Netherlands
Huy, N.Q., 2008. Rapid Load Testing of piles in sand, effects of loading
rate and excess pore pressure, PhD-thesis Delft University of
Technology, the Netherlands
Janes, M.C., Bermingham, P.D. & Horvath, R.C. 1991. An innovative
dynamic test method for piles, Proc. 2nd Int. Conf. Recent advances
in geotechnical earthquake engineering and soil dynamics, St.
Louis, Missouri
Schellingerhout, A.J.G. & Revoort, E., 1996. Pseudo static pile load
tester, Proc. 5th Int. Conf. Application of stress-wave theory to
piles, Florida
Maeda, Y., Muroi, T., Nakazono, N., Takeuchi, H., and Yamamoto, Y.
1998. Applicability of Unloading-point-method and signal
matching analysis on the Statnamic test for cast-in-place pile.
Proceedings of the 2nd International Statnamic Seminar, pp. 99-108
150
Matsumoto, T., Tsuzuki, M. & Michi, Y. 1994. Comparative study of
static loading test and Statnamic on a steel pipe pile driven in a soft
Pore water pressures [kPa]
100
rock. Proc. 5th Int. conf. and Exhibition on Piling and Deep
Foundations, Bruges, Belgium
50
Middendorp, P.; Bermingham, P.; Kuiper, B. 1992. Statnamic load
testing of foundation piles, "Proc. 4th Int. Conf. Appl. Stress-Wave
0 Theory to Piles, The Hague, Sept. 1992", Rotterdam, Balkema,
1992, pp. 581-588
-50 Profound, 2009. see http://www.profound.nl/statnamic.html, visited
February 2009
0.5