BENDER ELEMENT
What complicates the
TEST ANALYSIS
SOFTWARE DEVELOPMENT
Writing for theGeotechnica this month are Karl Snelling and Dr Sean Rees, Managing Director and
Geotechnical Specialist at GDS Instruments. In this in-depth article Karl and Sean discuss the test
analysis software development for laboratories that has been developed to interpret the data from
bender element test analysis.
Bender element testing has
become increasingly common-
place in soil laboratories
since its introduction in the
late 1970s by Shirley and
Hampton (1978). The test
allows straightforward small-
strain stiffness measurements
to be made in soil specimens,
and can be performed in a wide
variety of test systems.
“To this day however
there is still no recognised
standard for interpreting
the data obtained from
bender element tests.”
To this day however there is
still no recognised standard for
interpreting the data obtained
from bender element tests.
This fact provided motivation
for GDS Instruments, who
specialise in providing soil
and rock laboratory test
systems, to help address the
main aspect of subjectivity of
the test interpretation – the
determination of the shear
wave propagation time. This
resulted in the development
of a user-friendly piece of
software to automate the Figure 1 – Illustration of the bender element test (left); GDS
propagation time analysis.
7 8
interpretation of bender
element test data?
Although the bulk density and
distance between elements can
be measured accurately in the
lab, the time taken for a shear
wave to propagate through the
soil is somewhat subjective.
FOR LABORATORIES Consider the idealised received
waveform shown in Figure 2
– which point would you say
defines the time of shear wave
Figure 2 – Idealised shear wave recorded by a receiver bender
“...if two engineers agree element.
on using the same point to
How does the bender element to pick up the shear wave consuming for labs without the test data is removed, and
define the arrival, would strong software coding skills, a number of propagation time
test work? that has propagated through they necessarily record
the specimen, with its or knowledge of which analysis estimates are provided.
Bender elements are made displacement due to the wave
the exact same time methods have previously been
from piezoelectric ceramic inducing a voltage, which is purely through visual suggested. The task presented Given the tool was developed
bimorphs, and are used in pairs then read by a data acquisition observation of the wave?” to the GDS team was therefore with the larger geotechnical
to measure the shear wave unit. Through knowing the clear: review the literature, community in mind, there
velocity in a soil specimen. This distance between the two determine the analysis were two other important
arrival? Further to this, if two methods available, and develop specifications: be simple-to-
involves inserting each element elements, and observing the engineers agree on using the
a small distance into the top time required for the shear a simple-to-use software tool use, and be flexible enough to
same point to define the arrival, that objectively finds the shear analyse data taken from any
and base of a specimen, then wave to propagate, a value of would they necessarily record
applying an excitation voltage the shear wave velocity can be wave propagation time in bender element test system,
the exact same time purely bender element tests. not just the GDS system. Both
to one element to generate obtained. From this point only through visual observation of
a shear wave in the soil, as the specimen dimensions and the wave? Development of the GDS
“Both of these
illustrated in Figure 1. soil bulk density are required
Bender Element Analysis Tool specifications were
to produce a shear stiffness These considerations are of
The other element is used estimate.
achieved by using
course not recent, with many The development process
Microsoft Excel as the
numerical methods already led GDS to create the Bender
proposed in the geotechnical Element Analysis Tool, or platform, a piece of
literature to objectively GDS BEAT for short. The tool software familiar to most
determine the propagation is unique in that it does not practicing engineers.”
time of a shear wave. Such simply settle on one specific
methods typically analyse the numerical analysis method,
of these specifications were
test data in either the time or but instead implements three: achieved by using Microsoft
frequency domain, and tend to objective
determination Excel as the platform, a piece
vary in their complexity. of Point A, B, C, and D via of software familiar to most
“However implementing software algorithm, cross- practicing engineers. The tool
such methods on a routine correlation of the generating was split into two Excel Add-
and receiving element signals, Ins, each having a specific use
basis can often be difficult
and a cross-power spectrum – the first allows the user to
and time-consuming...” calculation of the signals to load one data set into an Excel
estimate propagation time in sheet, then select the various
the frequency domain. This parameter values required to
However implementing such decision provides distinct run the analysis, whilst the
methods on a routine basis
bender elements inserted into a triaxial top-cap and pedestal advantages to the user, as the second permits multiple
can often be difficult and time- hard-work required to process GDS data files to be simply
(right).

dropped into the tool and batch
analysed. Screenshots of each
are displayed in Figure 3.
“It was also important to
ensure the analysed data
was presented in a clear
format, both numerically
and visually.”
It was also important to
ensure the analysed data was
presented in a clear format, both
numerically and visually. With
this in mind the tool produces
two tabs in Excel following the
analysis – one listing numerical
9 10
Figure 3 – Screenshots of the GDS BEAT Add-Ins
created for Microsoft Excel.
values of was also necessary to verify
the wave the software performed as
propagation time estimates
and analysis metrics, and the
“A triaxial specimen of
other giving visual plots of Leighton Buzzard sand
the recorded element signals was therefore prepared
relative to the estimated in a GDS Dynamic Triaxial Figure 4 – Numerical (top) and visual (base) representations of a GDS BEAT analysis.
propagation times. Presented Test System (DYNTTS)...” on-the-spot estimates for the
in Figure 4, this combination “This quickly showed how shear wave propagation time. “To do this, the raw
of reporting allows the user to useful BEAT may be in test data was sent to an
rapidly validate the analysis specified during testing. A laboratories...” While this demonstrated the academic familiar with
data, and to further process the triaxial specimen of Leighton user-friendly nature of GDS bender element analysis,
information as required. Buzzard sand was therefore were complete. This quickly BEAT, further review was and asked to provide his
prepared in a GDS Dynamic showed how useful BEAT conducted post-test to check
How well does GDS BEAT Triaxial Test System (DYNTTS), own estimates...”
may be in laboratories – how accurate the propagation
perform? with bender element tests immediately after saving the time estimates really were
conducted using a GDS Bender academic familiar with bender
bender element data, files when compared with traditional
Developing BEAT was the first Element System (BES) after element analysis, and asked to
were dropped into the tool, observation. To do this, the
step for the GDS team, but it saturation and consolidation provide his own estimates
with rapid analysis providing raw test data was sent to an by viewing the generated
 1 3
E S - 20
O U RS
NG C
O M I
UP C
01 2
2

ONLY
M a y
t
31s
0 1 3,

£225 + VAT
ch 2
-
a r
13
P
th M
SHO
7 2 0
K
ch
R
r
WO -
a
th M
N
13
O P
IPTI SHO
2 8 2 0
R K
ril
ES C WO R -
A p
L D N N
6th
SO I T I O E S I G
2
CRI P ON D
S T I
13
KD E A
UND
2 0

ROC
13
O
ch
F -
r
L T
0 a
ICA VA
3 ay 2 M
+
01
N -
5th
H 0 0
M
TEC
2
NG - £1
y t h
- I
a
O
10
E R S
M
G ING NIT O ENE S
4 t h
S T O R
2
E M A
S I TU T N A ND I N G AW
Figure 5 – Leighton Buzzard triaxial test specimen used to verify

IN T A TIO Y T EST
the performance of GDS BEAT (left); bender element signals

N TOR
R UME A
obtained from the specimen (above).

T R
INS L L ABO
across a sensible range of shear demonstration can be found by

H N ICA
wave frequencies. visiting www.gdsinstruments.

TEC
GEO
com, along with free download
“The agreement between Ultimately GDS hope their of the software for a limited
BEAT and the academic new software tool, GDS BEAT, time only.
will not only be useful for
was highly encouraging: engineers interpreting bender References
all but the cross-spectrum element data, but will also
Supported by
analysis method led to generate discussion within the Shirley D. J. and Hampton
shear wave velocities geotechnical community and L. D. (1978). Shear-wave
being calculated within a contribute in the move towards measurements in laboratory
5 m/s band...” recognised test standards. sediments. Journal of the
For all those interested, Acoustical Society of America
further details and video 63 (2), 607-613. To book your place, please contact
and received waveforms. The Equipe Training:
agreement between BEAT
and the academic was highly
encouraging: all but the cross-
info@equipetraining.co.uk
spectrum analysis method www.equipegroup.com
led to shear wave velocities 01295 670990
being calculated within a 5
m/s band, which is just 2.2 % 01295 678232
of the estimated 225 m/s shear
wave velocity, when comparing Other courses are available. Please visit
our website for more details.
11 12