Seismic cone downhole procedure to measure shear wave velocity - a guideline

prepared by ISSMGE TC10: Geophysical Testing in Geotechnical Engineering

Procd sismique de downhole de cne la vitesse dondes de cisaillement de mesure - une
directive a prpar par ISSMGE TC10 : Essai gophysique dans la technologie gotechnique

A.P. Butcher
BRE, UK
R.G. Campanella
University of British Columbia, Canada
A.M. Kaynia
Norwegian Geotechnical Institute, Norway
K.R. Massarsch
Geo Engineering AB, Sweden
ABSTRACT
The International Society for Soil Mechanics and Geotechnical Engineering, Technical Committee No 10: Geophysical Testing in
Geotechnical Engineering has as part of its brief the task of drafting guidelines for geophysical techniques where no other national or
international standards or codes of practice exist. This document is the first of these guidelines and concerns the use of the Seismic
Cone to measure downhole seismic wave propagation.
RSUM
La Socit Internationale de Mcanique des Sols et de la Gotechnique, le Comit technique No. 10 : L'essai gophysique dans la
technologie gotechnique a en tant qu'lment de son dossier le charger des directives de rdaction pour des techniques gophysiques
o aucune autre norme ou recueil d'instructions nationale ou internationale n'existe. Ce document est le premier de ces directives et
concerne l'utilisation du cne sismique de mesurer la propagation sismique dondes de downhole.


1 INTRODUCTION
This document is to provide guidance to practitioners and
procurers on downhole seismic wave measurement using a
seismic cone penetrometer. The guideline has been prepared by
ISSMGE TC10: Geophysical Testing in Geotechnical
Engineering and is a supplement to the International Reference
Test Procedure (IRTP) for the electric Cone Penetration Test
(CPT) and the Cone Penetration Test with Pore pressure
(CPTU) as produced by the ISSMGE TC16. The document
therefore follows, and should be used with, the CPT IRTP
(1999).
The addition of a seismic sensor (usually a geophone but
may be an accelerometer or seismometer) inside the barrel of a
standard electric CPT is termed a Seismic Cone Penetrometer
Test (SCPT) (Robertson et al, 1986). Such a sensor allows the
measurement of the arrival of vertically propagating seismic
body waves, generated from a source on the ground surface, in
addition to the usual cone parameters that are used for detailed
stratigraphic logging.
There are two types of seismic body waves, Pressure or
Compression waves (P waves) as well as Shear waves (S
waves) and seismic sensors react to both. The P wave always
arrives first. In soils below the ground water table the P wave
typically travels 2 or more times faster than the S wave, so
separation of the two body waves is easy. Above the water
table, however, the difference is small and separation of P and S
waves may be very difficult, requiring specialized techniques.
However the most significant difference between P and S waves
is that S waves are reversible. Therefore using a source that can
produce shear waves of opposite polarity facilitates the
identification of S waves.
Since shear waves travel through the skeletal structure of the
soil at very small strains, one can apply simple elastic theory to
calculate the average elastic small strain shear modulus, over
the length interval of measurement, as the mass density times
the square of the shear wave velocity. Thus, the shear wave ve-
locity relates directly to stiffness (Massarsch, 2004) and also
may be used to estimate liquefaction susceptibility in young
uncemented sands (Youd et al, 2001).
2 DEFINITIONS
The following definitions will be used:
Accelerometer: Sensor that produces an output in response
to a seismic wave by way of a change in capacitance caused by
the relative movement of a mass and the sensor case. An accel-
erometer detects particle accelerations.
Array: group of devices at one location orientated or-
thogonally to each other.
Data recording equipment: Equipment to log the signals
from the seismometers.
Geophone: Sensor that gives an output in response to seis-
mic waves using the relative movement of a mass (magnet)
moving within a coil fixed to the sensor case. A geophone de-
tects particle velocities.
Hammer: Heavy mass to impact the Shear Beam as part of
the Source
Interval time: The difference in arrival times of seismic
waves at the receivers at two depths/distances from the Source.
The true interval is the difference in arrival times between re-
ceivers at a fixed distance apart and the pseudo interval is the
difference in arrival times to the same receiver when placed at 2
different distances from the source.
Seismometer: Device that produces a calibrated self gener-
ated output response to imposed seismic waves and gives
maximum output at its natural frequency or fundamental mode
(goes into resonance) when activated by seismic waves. A seis-
mometer can be an accelerometer, geophone or a sensor able to
detect deflections in the range 0 to 250Hz.
Seismometer natural frequency: Frequency at which the
seismometer gives its maximum output and above which the
seismometer response is constant.
Butcher, A. P., Campanella, R.G., Kaynia, A.M. and Massarsch, K. R. 2005. Seismic cone downhole procedure to measure
shear wave velocity - a guideline prepared by ISSMGE TC10: Geophysical Testing in Geotechnical Engineering. XVI
th
Internl.
Conference on Soil Mechanics and Geotechnical Engineering, Osaka, (submitted for publication), 5 p.
Shear beam: Beam that forms part of the downhole seismic
shear wave Source that is impacted by a Hammer to maximize S
waves and minimize P waves.
Source: Device that, when activated, generates polarised
shear waves that propagate into the ground. (A basic source will
include a loaded Shear Beam, Hammer and a Trigger to activate
the data recording equipment).
Trigger: Device attached to either the Shear Beam or the
Hammer to initiate the data recording equipment at the instant
the Shear Beam is struck by the Hammer.
3 METHODOLOGY
During a pause in cone penetration, a shear wave can be created
at the ground surface that will propagate into the ground on a
hemi-spherical front and a measurement made of the time taken
for the seismic wave to propagate to the seismometer in the
cone. By repeating this measurement at another depth, one can
determine, from the signal traces, the interval time and so
calculate the average shear wave velocity over the depth
interval between the seismometers. A repetition of this
procedure with cone advancement yields a vertical profile of
vertically propagating shear wave velocity. Figure 1 shows 2
alternative schematic arrangements of the SCPT, and Figure 2
shows a typical arrangement of the surface shear wave source.
4 EQUIPMENT
The general arrangement of equipment is shown in Figures 1
and 2.
Seismometer: The seismometer will typically have a natural
frequency of less than 28 Hz and must fit inside the cone barrel.
The seismometer must be mounted firmly in the cone barrel
with the active axis in the horizontal direction and the axis
alignment indicated on the outside of cone body. The cone bar-
rel at the location of the seismometer should be of a greater di-
ameter than the sections immediately below the location of the
seismometer to ensure good acoustic coupling between the cone
barrel and the surrounding soil.
Comment: Some seismic cones include 2 seismometers in an
array in the horizontal plane set with their active axes or-
thogonally. This configuration allows compensation for possi-
ble rotation of the cone drive rods, (and the cone containing the
seismometer) with the subsequent loss in response and also
gives orthogonal seismic wave traces from the same source ac-
tivation. In variable and layered ground conditions, with ambi-
ent noise or ground structures that would corrupt the received
signals, wave characteristics of the source can be used to iden-
tify the shear wave amongst the other waves.
The inclusion of a vertically orientated seismometer will al-
low the P wave element of the seismic wave to be assessed or P
wave arrival measured if a P wave source is used. In many
cases the combination of P and S wave data can enhance the

Shear
beam
Distance (X) from
the centre line of
shear beam to
insertion point of
seismic cone
Shear
beam
Insertion
point of
seismi c
cone
Figure 2: Typical downhole shear wave source setup with shear
beamand fixed axis swing hammers.
Figure 1a: Schematic diagramof the seismic cone test with required
dimensions, D
1
, D
2
, and X.

Axis of SCPT
D
1

D
2

CPT push
rods
X
SCPT at depth D
1

SCPT at depth D
2

Shear
beam
Assumed travel paths of
seismic waves from shear
beam to seismometers in
SCPT body at depths D
1

and D
2

L
1

L
2

Figure 1b: Schematic diagramof the dual array seismic cone test with
required dimensions, D
1
, D
2
, and X.
Shear
beam
CPT push
rods
Axis of SCPT
Receiver 1
Receiver 2
L2
L1
Assumed travel paths
of seismic waves from
shear beam to
seismometers in
SCPT body
X
Dual
array
SCPT
body
D1
D2
identification of stratigraphic boundaries.
Shear Beam: The beam can be metal or wood encased at the
ends and bottom with minimum 25 mm thick steel. The strike
plates or anvils at the ends are welded to the bottom plate and
the bottom plate should have cleats welded to it, to penetrate the
ground and prevent sliding when struck. The shear beam is
placed on the ground and loaded by the levelling jacks of the
cone pushing equipment or the axle load from vehicle wheels.
The ground should be prepared to give good continuous contact
along the whole length of the beam to ensure good acoustic
coupling between the beam and the ground. The Shear Beam
should not move when struck by the hammers otherwise energy
is dissipated and does not travel into ground and does not pro-
duce repeatable seismic shear waves. The anvils, on the ends of
the Shear Beam, when struck in the direction of the long axis of
the Shear Beam, will produce shear waves of opposite polarity.
Comment: The beam can be continuous (approximately 2.4
m long) i.e. greater than the width of a vehicle or equipment
used to load the beam and 150 mm wide or alternatively can be
two shorter beams placed and loaded so that the anvils oppose
and can be struck by the hammers to produce shear waves of
opposite polarity. Care must be taken to position the beams and
strike direction to maximise S waves and minimise the produc-
tion of P waves.
Heavy hammer(s): Heavy hammer(s) with head mass of be-
tween 5 to 15 kg to strike the plate or anvil on the end of the
shear beam in a direction parallel to the long axis of the shear
beam and the active axis of seismometer.
Comment: Two fixed axis hammers, one to strike each end of
the beam in the specified directions, will significantly speed up
the operation and give controllable and consistent source out-
put. A typical setup is shown in Figure 2.
Data recording equipment: The recording equipment can be
a digital oscilloscope, a P.C. with installed A/D board and oscil-
loscope software or a commercial data acquisition system such
as a seismograph. The data recording equipment must be able
to record at 50 s (microsecond) per point interval, or faster, to
ensure clear uncorrupted signals and to start the logging of the
seismometer outputs using an automatic trigger. An analogue
anti aliasing filter should be used to avoid corruption of signal
frequencies above the device limits. Commercial data recording
equipment usually include amplifiers and signal filters to help
enhance recorded signals. The effect of these processes on the
recorded signals must be considered before their use. For exam-
ple filtering can cause phase shift of signals and amplification is
usually limited to a frequency range. In either case the signals
may not be directly comparable.
Comment: Experience has shown that there is a significant
advantage to record the unprocessed data and then the effect of
filtering and processing can be assessed during post processing.
Most modern acquisition equipment allows the viewing of fil-
tered signals during acquisition (to assess quality and repeat-
ability) but saves the data un-filtered. Most modern acquisition
equipment allows signal stacking to improve signal to noise ra-
tio.
Trigger: The trigger can be fixed to the hammer head or the
beam. The trigger is required to be very fast (less than 10 mi-
crosecond reaction time) and repeatable. When the hammer hits
the shear beam, the electrical reaction of the trigger activates the
trigger circuit that outputs to the signal recording equipment. A
typical trigger circuit is given in Campanella & Stewart (1992).
A seismic trigger mounted on the beam may be used if it is fast
enough, repeatable and delay time is checked and known or a
contact trigger that works the instant contact is made between
the hammer and the anvil.
Comment: The use of 2 arrays of seismometers set in the
cone barrel a fixed distance apart, say 0.5m or 1.0m, (termed a
dual array seismic cone, see Figure 1b) would enable the travel
time of the shear wave to be measured between the seismome-
ters from the same source activation thereby avoiding possible
errors from selection of signal from different source activation,
the speed of the trigger, and the accuracy of distance from the
source to the receivers from successive pushes of the drive rods
to each depth. In this case the seismometers must have identical
response characteristics (natural frequency, calibration and
damping). However if signals are to be stacked, that is the sig-
nals from successive source activations added together to im-
prove signal to noise ratio, the trigger time must be repeatable.
5 TEST PROCEDURES
At the start of the SCPT, the body of the cone should be rotated
until the axis of a seismometer is parallel to the long axis of the
shear beam.
a) The cone is pushed into the ground, monitoring the incli-
nation of the cone barrel during the push.
Comment: It is important to know the exact location of the
receivers in all three axes and the inclinometer in the cone bar-
rel will give the horizontal component and the depth measuring
system of the CPT the vertical component.
b) The penetration of the cone is stopped and the depth to the
seismometer/s is recorded. The horizontal offset distance, X,
from cone to centre of the shear beam should also be recorded
(see Figure 1)
Comment: Typically this procedure is carried out at depths
greater than about 2-3m in order to minimize the interference of
surface wave effects. If the seismic cone includes a fully opera-
tive electric cone then it will be advanced at 2 cm/s and stopped
typically at a rod break at 1m intervals or for pore water pres-
sure dissipation tests. If acceptable such stoppages can also be
used for downhole seismic wave measurements. Alternatively
the seismic cone can be pushed to a predetermined depth at
which the shear wave velocities are required and the measure-
ments made. To avoid the possible effects of time between stop-
ping, pushing and making measurements it is advisable to keep
this time interval consistent. The horizontal distance, X, be-
tween the entry point of the seismic cone and the source should
be kept at around 1m. Greater distances will require the effects
of curved travel paths, that particularly affect single array
SCPTs, to be addressed. It is advisable at the first depth of
measurement to monitor the output of the receivers without ac-
tivating the source to determine the ambient seismic noise in the
ground and thereby enable the filtering, as far as possible, the
ambient noise. Experience has shown that ambient noise can be
reduced by retracting the cone pushing system, so that the drive
rods are unloaded and there is no contact between the shear
beam system and the cone drive rods through the cone drive ve-
hicle, and the cone driving equipment motors are not running.
c) The shear beam is struck by the hammer and the trigger
activates the recording equipment that then displays the time
based signal trace received by the seismometer.
Comment: For quality assurance, it is recommended to reset
the trigger and repeat the procedure until a consistent and re-
producible trace is obtained. The voltage-time traces should lie
one over the other. If they do not, continue repeating until
measured responses are identical. In the case of the dual array
SCPT the traces from both the seismometers can be displayed
together giving a rapid assessment of the shear wave propaga-
tion time. If the seismic wave velocity appears too high then
there may be a connection between the cone drive system and
the seismic cone so allowing the seismic waves to travel
through the cone drive rods instead of the ground.
d) The trigger is reset and the shear beam is then struck by
the hammer on the opposite end on the other side of vehicle
(causing initial particle motion in the opposite direction and a
shear wave of opposite polarity) and procedure in step c) is
again completed.
e) Show the traces from step c) and d) together and identify
the shear wave (usually clearly seen with traces from the oppo-
site polarity shear waves as a mirror image in time) and pick an
arrival time. An example of a pair of signals is shown in Figure
3.

t
2
-t
1
= 5.53ms
Figure 3: An example of oppositely polarised shear wave traces with
clear crossover of traces showing the interval time T
2
T
1
.
T
2
-T
1
=5.53ms
With reversed image traces, the first major cross-over can be
taken as the reference arrival, or one trace may be used and an
arrival pick made visually by an experienced operator. If the
wave arrival point is not clear then a significant point that oc-
curs on both traces can be used provided it occurs shortly after
the likely wave arrival, later selections are likely to be affected
by signal attenuation and dispersion. Alternately, a cross-
correlation procedure may be used to find the interval travel
time using the wave traces from strikes on the same side at suc-
cessive depths (Campanella & Stewart, 1992). This technique
is more complex, but eliminates the arbitrary visual pick of arri-
val time and is necessary if symmetry of reverse wave traces is
lacking. If a dual array seismic cone is used then the wave
traces from each seismometer can be compared to get the travel
time between seismometers. Figure 4 shows an example of
pseudo interval traces between 4 and 15m depth.
Comment: As depth increases the signal to noise ratio de-
creases. At large depths it may be necessary to increase sig-
nal/noise (depending on the amplification, resolution and accu-
racy of the data recording equipment). This can be achieved by
using multiple source activation events (from 4 to 10) and add-
ing (or stacking) the measured signals, this will reduce most of
the random noise and increase signal/noise ratio.
The average downhole shear wave velocity is calculated for
the depth interval the cone has been driven between measure-
ments or the fixed distance between the two seismometer sets in
a dual array seismic cone.
The average shear wave velocity for the given depth interval
in units of m/s and assuming straight ray paths (see Figure 1) is
given by Equation (1):

Time (milliseconds)
where:
0 40 80 120 160
1 2
1 2
S
T - T
L - L
= V
(1)
15
10
5
0
D
e
p
t
h

(
m
)
L
2
= calculated length, m of the straight travel path distance
from source to receiver at greater depth (use horizontal
offset, X, and vertical depth D
2
).
L
1
= calculated length, m of the straight travel path distance
from source to receiver at shallower depth (use horizon-
tal offset, X, and vertical depth D
1
).
T
2
= shear wave travel time from source to receiver at greater
depth (along wave path L
2
).
T
1
= shear wave travel time from source to receiver at shal-
lower depth (along wave path L
1
).
T
2
-T
1
=interval travel time.
6 REPORTING OF RESULTS AND INTERPRETATION
PROCEDURES
The following information shall be reported:

For each site:
a) Length of shear beam (lengths if two beams are used)
and material and composition including anvils.
b) Mass of swing hammers.
c) Fixed or free pivot point of swing hammers
d) Trigger type and location. (for single seismometer
seismic cones a typical trigger delay time)
e) Distance (X) of shear beam from insertion point of
SCPT, and distance of impact points from the inser-
tion point of the SCPT.
Figure 4: Example of pseudo interval traces of shear waves
at depths 4mto 15m.
f) Type of receivers, their specifications, serial numbers
and name of manufacturer and last dated response
calibration.
g) Type, serial number and specification of data re-
cording equipment and name of manufacturer.

For each location
h) Date and time of test
i) Identification of test
j) Altitude and location of insertion point of SCPT

For each depth:
k) Depth of receiver(s) from ground level.
l) Direction of swing hammer action.
m) Rate of sampling and sample length for each record.
n) Name of files where raw and processed data are re-
corded including media and location of storage.
o) Type and specification of real time processing in-
cluded in the recorded data.
p) Type and specification of post measurement process-
ing included in the presented data.
q) Calculated propagation times of the shear waves and
the depth range over which the measurement was
taken.
r) Calculation of the Shear Wave velocities and the
depth range over which the velocity was calculated.

The data files in n) should be stored for future access or for
further processing until the end of the project or as specified by
the client.
7 ACKNOWLEDGEMENTS
Drafts of this document were discussed at the TC 10 Members
Meetings in Prague (2003) and Porto (2004). Valuable com-
ments and suggestions for improvements were made by mem-
bers of TC 10 as well as members of TC 16 Ground Properties
from In-situ Testing and TC 1 Offshore and Nearshore Geo-
technical Engineering. Their contributions are acknowledged
with gratitude.
8 REFERENCES AND FURTHER READING
Butcher, A.P. and Powell, J .J .M., 1995. Practical considerations
for field geophysical techniques used to assess ground stiff-
ness. Proc. Int. Conf. on Advances in Site Investigation
Practice, ICE London, March 1995. Thomas Telford, pp
701-714 .
Campanella, R.G. and Stewart, W.P. 1992. Seismic Cone
Analysis using digital signal processing for dynamic site
characterization, Canadian Geotechnical Journal, Vol. 29,
No. 3, J une 1992, pp.477-486.
IRTP, 1999:ISSMGE Technical Committee TC16 Ground
Property Characterisation from In-situ Testing, 1999.
:International Reference Test Procedure (IRTP) for the Cone
Penetration Test (CPT) and the Cone Penetration Test with
pore pressure (CPTU). Proc. XIIth ECSMGE Amsterdam.
Balkema. pp 2195-2222.
Massarsch, K. R. 2004. Deformation properties of fine-grained
soils from seismic tests. Keynote lecture, International Con-
ference on Site Characterization, ISC2, 19 22 Sept. 2004,
Porto, 133-146.
Robertson, P.K., Campanella, R.G., Gillespie, D. and Rice, A.
1986. "Seismic CPT to Measure In-Situ Shear Wave Veloc-
ity", ASCE, Journal of Geotechnical Engineering, Vol. 112,
No. 8, August 1986, pp. 791-804.
Youd, T.L., Idriss, I. M., Andrus, R. D., Arango, I., Castro, G.,
Christian, J .T., Dobry, R., Liam Finn, W.D., Harder J r.L.F.,
Hynes, M.E., Ishihara, K., Koester, J .P., Liao, S.S.C.,
Marcuson III, W.F., Martin, G.R., Mitchell, J .K., Moriwaki,
Y., Power, M.S., Robertson, P.K., Seed, R.B., and Stokoe II,
K.H., 2001. Liquefaction Resistance of Soils: Summary Re-
port from the 1996 NCEER and 1998 NCEER/NSF Work-
shops on Evaluation of Liquefaction Resistance of Soils J.
Geotech. and Geoenvir. Engrg., Volume 127, Issue 10, pp.
817-833 (October 2001).


APPENDICES

Appendix A: Maintenance, Checks and Calibrations,

This appendix contains informative guidance on maintenance,
checks and calibrations for the SCPT but excludes those parts
that are common to the CPT and are included in the CPT IRTP
(1999).
A.1: Seismometers.
The seismometers should be checked to ensure they comply
to the manufacturers specification in response to seismic waves
in regard to frequency, phase and damping before each profile.
Where arrays of seismometers are used, such as for true in-
terval time measurements, each seismometer must have an iden-
tical response, in laboratory test conditions, to seismic waves in
regard to frequency, phase and damping.
A.2: Source and Triggers
Where single seismometer seismic cones are used the source
activation and trigger time delay will have to be quantified. The
trigger delay time needs to be repeatable and not vary by more
than 1%.