INTERNATIONAL SOCIETY FOR

SOIL MECHANICS AND

GEOTECHNICAL ENGINEERING

This paper was downloaded from the Online Library of

the International Society for Soil Mechanics and
Geotechnical Engineering (ISSMGE). The library is
available here:

https://www.issmge.org/publications/online-library

This is an open-access database that archives thousands

of papers published under the Auspices of the ISSMGE and
maintained by the Innovation and Development
Committee of ISSMGE.
 Geotechnical and Geophysical Site Characterisation 5 – Lehane, Acosta-Martínez & Kelly (Eds)
© 2016 Australian Geomechanics Society, Sydney, Australia, ISBN 978-0-9946261-1-0

The SPT N-value errors examined with digital technology

Burt Look
Foundation Specialists Pty Ltd, Brisbane Queensland Australia

ABSTRACT: The Standard Penetration Test (SPT) has the simplicity of only counting the number of blows
over 3 penetration increments. The blow counts are taken as close as possible to 150mm, based on the super-
visor’s judgement. Digital measurements show the “standard” 150mm increments varied considerably. These
digital measurements were obtained using a Pile Driving Monitoring (PDM) – a device which remotely
measures set, temporary compression and velocity using optically safe infrared laser technology. This tech-
nology was applied to SPT measurements in a similar way to observe the set and energy. The 3 data points for
the N - value are measured as 30,000 data points. The energy corrections are required with the in situ N –
values to be effectively used as a design value. These energy values on the hammer and below the anvil was
different even for a specified hammer and is dependent on both soil type and rod length.

1 INTRODUCTION Thus a Pile Driving Monitoring (PDM) device

was used with the SPT to digitally measure blow
The Standard Penetration test (SPT) is one of the counts in terms of set and temporary compression,
common in situ tests used in geotechnical engineer-
similar to pile driving. At the same time energy ra-
ing to determine properties of subsurface soils. The
tio is also measured. Digital measurements provide
SPT requires 3 data points - the seating drive and 2
an improved accuracy which the current “visual”
test drives to 150mm penetration each. The N-value
measurements using chalk mark references cannot
is used to estimate the approximate shear strength
provide.
properties of the soils (Clayton, 1995). Various cor-
The PDM can also measure energy transfer.
rections should then be made to the in- situ value to However the energy transfer of the hammer at im-
allow for type of hammer, energy corrections, etc. pact is different to the energy below the anvil (Fig-
(Skempton, 1986). ure 1). The energy ratio also varies with each blow
Different types of hammers used influence the N-
(Figure 2) and is discussed further in Look et al.
value, with different energy efficiencies. The fail-
(2015). This data is apparent only with digital meas-
ings of the SPT was revealed in Look and Seidel
urements and cannot be seen by eye.
(2015). These include
1. Counting of integer values over 150mm in-
crement is inexact
2. A seating drive is a constant 150mm yet seat-
ing varies significantly (Figure 1)
3. Energy transfer which (even for given drill
rig) depends on the soil type and the variabil-
ity in each blow energy transfer from the
hammer to the anvil to the rod.
Thus the SPT is far from “standard” as the name
implies, despite a standard hammer weight and drop
height. Energy and other corrections are required to
effectively use the SPT value in design. The SPT is
analogous with pile driving, with an energy transfer
between blows, dependent on the drill string loca-
Figure 1. Energy transfer between hammer and below anvil.
tion, rod length and material stiffness.

333
 The PDA and PDM data obtained was download-
ed from the devices and processed to determine the
energy delivered to the rods for each blow and from
a hammer efficiency relationship was determined.

2.2 PDM Device

The PDM uses LED to track the movement of a re-
flector attached to the moving object, safely placed
about 10-15m from the pile and accurate to better
than 0.1mm at 10m range. There are no connections
required. Thus the device is first and foremost a
safety device to avoid operators measuring with a
Figure 2. Seating complete before the 150mm seating drive ruler below a pile driving hammer with falling parts
(but measured as 158mm). above. The device is also a quality measuring de-
vice to measure pile set and energy.
Pile driving is similar to the SPT, when a hammer
2 BACKGROUND is used to drive an object into the ground. The PDM
2.1 Applicable Standards measures set and temporary compression and the
peak velocity (energy) can also be determined.
Australian Standards (AS 1289.6.3.1-2004) for the Figure 3 shows the SPT on the hammer only,
Standard Penetration Test (SPT) does not require while Figure 4 shows measurement where 2 PDMs
energy measurements to be made, while the British were used with one on the hammer, and the other
and ASTM standards require energy measurements. below the anvil. An SPT analyser placed below the
Testing was performed in accordance with ASTM anvil is also shown in Figure 4.
D4633 – Standard Test Method for Energy Meas-
urement for Dynamic Penetrometers. This method
involves inserting an instrumented rod at the top of
the drill rod string directly below the hammer-anvil
system so that the hammer impact is transmitted
through the anvil, into the instrumented rod and then
into the drill rods. The instrumented rod contains
two strain transducers and two accelerometers such
that measurements of time varying force and accel-
eration can be made during hammer impact.
The force and acceleration signals are transmitted
by connecting the gauges to a Pile Driving Analyser
(PDA) computer which then enables recording, pro- Figure 3. PDM measurement with reflector on SPT hammer.
cessing and displaying of the data to allow determi-
nation of force and velocity versus time. This in turn
provides a measurement of the energy transferred to
the instrumented rod as a result of the impact from
the hammer. However accelerometers cannot relia-
bly determine final displacement and as such unad-
justed PDA data may lead to an overestimation of
the energy transfer.
The above is the usual approach as per the Brit-
ish or American Standards. A Pile Driving Monitor
(PDM) test was used in parallel with the PDA in or-
der to accurately determine the vertical displacement Figure 4. Two PDMs used with SPT analyser.
of the rod after each blow. The PDM measures the
vertical displacement of a target reflector which is The PDM was used both for accurate displace-
attached to the drill rod. Look and Seidel (2015), ments to be recorded for each blow and for estimat-
Look et al. (2015a and 2015b) has consistently ing energy transfer as well as using the PDA. The
shown, the digitally measured N - values are not the energy mass is not the hammer only, but the com-
same as the supervisor N-value measured by eye bined effect of hammer, anvil, rod and soil being
referenced to the chalk or paint marks on the rod. penetrated. Research is ongoing to assess that mass
more accurately.

334
 The accuracy of the drilling supervisor’s assess-
ment in that time frame is required, while standing at
a safe distance. A chalk mark reference is also 5mm
to 10mm wide, which accounts for a reliable reading
error range of 2 X 5mm to 2 X 10mm (7% to 13%
error / 150mm).
Look et al. (2015) show the counting variation for
54 digitally measured increments (Figure 7). The
target of 150mm had a mean and standard deviation
of 147mm and 16mm, respectively. The lowest and
highest values recorded was 109mm and 191mm.
Because two 150mm increments are used for the N-
value, compensating errors occur. The error transfer
Figure 5. PDM measurements compared with counting be- between the seating and test drive is the greater con-
tween chalk marks. cern.

Figure 6. Digital Trace of blow No. 30. Figure 7. Digital Measurements at Site 1 (54 No.)

Figures 5 and 6 show the measurement process

and compares with the supervisor’s blow count. In Figure 8. compares the visual measurements at
this case driving to refusal occurred in 42 seconds another site with similar results. This data shows
with an automatic trip hammer. The experienced counting blows is not a “standard” value as implied
supervisor measured 30 blows in the 48mm refusal by the test.
while the digital measurements show 30 blows oc- These sites have a coefficient of variation (COV)
curred within 40mm. This is a 20% counting error. of 11% and 15% for the 150mm increments. Phoon
and Kulhawy (1999) show the variability of in situ
tests with a COV of 15% to 45% for the SPT. The
2.3 Counting Accuracy counting errors alone would account for 15% COV.
Industry Standards state an SPT N – value is “factu-
al”. Yet it is only possible to count whole blows be-
tween increments. The supervisor must decide
which of the 150mm increment to assign a blow
which overlaps. The actual blows are NOT @
150mm but the nearest which is before or after the
150mm chalk mark. This “error” is then transferred
to the next 150mm increment.
In carrying out the SPT, the time is typically
- 0.2 seconds for the hammer falling
- < 0.05 second to come to a standstill (tempo-
rary compression occurs here),
- < 5 seconds before the next below is deliv-
ered Figure 8. Digital Measurements at Site 2 (30 No.).
- With some automatic hammers the time can
be less than 1 second between blows

335
2.4 SPT Correction Factors 3 FIELD RESULTS
The insitu SPT N - value provides an indicator of Using the results of site 2 (refer
relative change. However, correction factors are re-
quired for design. Typically “text book” correction Figure 8. for set variation) and the measured energy
values are used but these vary widely between refer- by the PDA analyser, the correction factors were de-
ences. These correction factors include:- termined. Figure 9 summarises the appropriate cor-
- Overburden correction factor - CN rection factors required for correcting the measured
o SPT N-value to a design N-value and considers both
- Energy correction factor (CER) to account for
o rig hammer and rod length. This assumes similar
o Hammer - C H
o energy efficiency applies for a given drill rig
o Rod Length (depth) - CR
o Sampler - Cs ; Borehole diameter - throughout the project and in the various materials.
CB ; Anvil – CA

The design value (No) 60 is then used, when also

corrected for overburden.

(No) 60 = CN CER N. (1)

Aggour and Radding (2001) summarise the SPT

correction factors applied by the various researchers
and conclude that the most significant factor affect-
ing the measured N values is the amount of energy
delivered to the drill rods. Yet Skempton (1986) us-
es an anvil correction factor of 0.6 to 0.8, while this Figure 9. SPT Energy Efficiency Measurements with correc-
is not included in the correction factors by others. tion factors - Site 2.
This significant factor warrants further research.
Australian Standards do not currently specify en- This was for a large site investigation where 3
ergy requirements in the SPT procedure, thus energy drilling rigs were operational. The energy associat-
is rarely measured. Any correction (if applied) is ed with the hammer and rod length have been meas-
based on the international literature. There is a wide ured for only 2 of the 3 rigs used at this site. These
range of energy correction factors described in the 2 factors (energy and rod length) are known to be
technical literature (Table 1). The SPT energy cor- the main variables for energy correction factors. The
rection for a trip hammer is therefore 80% to 172% energy ratio is then normalized to 60% of total ener-
depending on one’s favourite reference. Thus only gy.
actual energy measurements would provide confi- As rig No. 3 was not measured for its energy,
dence of an appropriate correction factor. then the energy correction factor is unknown and a
value of 1.0 is shown in Figure 10. With a correc-
Table 1: Energy Correction Factors for trip or Automatic
Hammers tion factor of 1.5 for rig 1, and using equations 2 and
3, then the design strength and modulus would pro-
Reference Energy ratio portionally increase by 50%.
Bowles (1996) 1.14 – 1.72 (N70 used) For this site, there was 1120 SPTs carried out.
Skempton (1986) None Listed With one rig having no correction applied, another
Robertson and Wide (1997) 0.8 – 1.5 with a negligible correction and the third with a sig-
Seed (1984) 1.67 nificant change, then 18% of the N- values required
a classification adjustment. No classification correc-
While correction factors are emphasised for lique- tion applies to an N-Value > 50, although the value
faction studies, many design correlations for itself changed.
strength and settlements also rely on an energy cor-
rection (Equations 2 and 3, respectively from
Stroud, 1989)
Modulus (E′) ~ N
Undrained Cohesion (C u) ~ 5 N60 (kPa) (2)
Modulus (E′) ~ N60 (MPa) for a factor
of safety of 3 (3)

336
4 CONCLUSIONS Measured energy losses are essential in using the
N-value as a design value. The errors are amplified
Integer values are measured for each 150mm incre-
when multiple drill rigs are used on the same pro-
ment in the SPT. PDM measurements show that in-
ject.
teger values required by the test means that blow
This case study shows that the same SPT value
counts are taken as close as possible to 150mm
obtained with the 2 drill rigs on this one site can
which may be more or less based on the supervisor’s
have its design value (N60) vary by a factor of
judgement. The “standard” 150mm increments has a
1.5/1.1 ~1.4. Even more concerning was that the
counting measurement coefficient of variation of
energy of the 3rd drilling rig was not measured. And
15%.
yet the variation of 140% is superior to using a value
The PDM provides both a true count of the SPT
from a reference which can vary from 80% to 172%
set values and the energy for the various elements.
for the hammer being used.
The digital measurements expose the inaccuracy for
Perhaps it is time for the ubiquitous SPT (1940s
N - values measured visually.
procedure) to enter the 21st Century digital age.

Figure 2. Summary of Procedure and Correction Factors for each rig at Site 2

337
5 REFERENCES

Aggour, M. & Radding, W.R. 2001. Standard Penetration Test

Correction. Maryland State Highway Administration,
Report No. MD02-007B48.
Australian Standards. 2004. Soil Strength and Consolidatyion
Tests – determination of the penetration resistance of a soil
– Standrad PenetrationTest (SPT). AS1289.6.3.1 Methods
of tesing soils for engineering purposes.
ASTM D4633-10. 2010. Standard Test Method for Energy
Measurement for Dynamic Penetrometers. American Socie-
ty for Testing and Methods.
British Standards. 2012. Geotechnical investigation and testing
- Field testing - Part 3: Standard Penetration Tests. BS
EN ISO 22476-3:2005+A1: 2011.
Clayton C.R.I. 1995. The Standard Penetration Test (SPT):
Methods and Use. CIRIA Report 143
Look, B. & Seidel, J. 2015. SPT and PDM – Macro vs micro
measurements. 12th Australia New Zealand Conference in
Geomechanics, Wellington, New Zealand, P015.
Look, B.G., Seidel, J.P., Sivakumar, S.T & Welikala, D.L.C.
2015. Real time monitoring of SPT using a PDM device –
The Failings of our Standard Test revealed. International
Conference on Geotechnical Engineering, Colombo, Sri
Lanka. 435 – 438.
Look, B.G., Seidel, J.P., Sivakumar S.T & Welikala, D.L.C.
2015. Standard Penetration Test measurement exposed us-
ing a Digital PDM device. International Conference on
Geotechnical Engineering, Colombo, Sri Lanka. 451 – 454.
Phoon, K. & Kulhawy, F.H. 1999. Characterization of ge-
otechnical variability. Canadian Geotechnical Journal. 36:
612 – 624.
Skempton, A.W. 1986. Standard Penetration Test Procedures
and the Effects in Sands of Overburden Pressure, Relative
Density, Particle Size, Ageing and Over-consolidation. Ge-
otechnique, 36 (Sept): 425 – 47.
Stroud, M.A. 1989. The Standard Penetration Test – its appli-
cation and interpretation. Proceedings ICE Conference on
Penetration Testing in the UK, London: Thomas Telford.

338