Recent Advances in Mechanical Engineering and Automatic Control

Monitoring of Mechanical Properties of Gypsum Using

the Impulse Excitation Method
RICHARD ŤOUPEK, TOMÁŠ PLACHÝ, MICHAL POLÁK, PAVEL TESÁREK
Department of Mechanics, Faculty of Civil Engineering
Czech Technical University in Prague
Thákurova 7, Praha 6, 166 29
CZECH REPUBLIC
plachy@fsv.cvut.cz http://mech.fsv.cvut.cz/

Abstract: - The paper presents results of gypsum mechanical properties monitoring using non-destructive
impulse excitation method. The advantage of the method is that the same specimen can be tested in different
time instants repeatedly without any damage. The first 14 days are the most important for material properties
evolution. After that time, the mechanical properties are not changing significantly. The article describes the
development of mechanical properties (Dynamic Young’s Modulus, Dynamic Shear Modulus, Poisson’s ratio)
of grey gypsum during its hardening.

Key-Words: - gypsum; impulse excitation method; Young’s modulus; shear modulus

1 Introduction [5]. Three gypsum specimens were made for the

The gypsum has been used as a building material for
several thousands of years. At present time, it is
used in plasterboards. They are used in new
constructions and also during reconstructions. From
the view of experts and also general public, the
gypsum is a material, which is used in the building
industry for its excellent usage properties (fast
hardening, good mechanical and thermal properties
and good workability) but only in interiors.
The little knowledge of this material does not
allow extending its use in exteriors. From the
chemical view, it is calcium sulfate dihydrate. The
gypsum binder can contain also other modifications
of the system calcium sulfate – water [1]. Therefore,
it is interesting to investigate this material also on
microlevel and to determine the content of the
gypsum phases in the investigated gypsum using the
nanoindentation technics [2]. The problem with
gypsum usage starts to be complicated when the
additives are added to the gypsum mixture to
improve the chosen properties [3], [4].
The time dependent changes of the silicate based
materials has been tested for several years at the
workplace of authors. From the view of the
mechanical properties evolution, the behaviour of
the gypsum is specific and different from the other
porous building materials based on cement or lime
binders. Especially, it is very fast creation of a solid
structure (removing of the gypsum from the mould
30 minutes after mixing gypsum binder with water)
and also non-traditional decrease of the mechanical
properties in the first week after making the gypsum
purpose to monitor and identify the gypsum
mechanical properties evolution in more detail.
2 Material and Specimens
The grey gypsum produced by Gypstrend co. Ltd.
was used for making the specimens. It is composed
of two different dehydrates. These two components
are natural gypsum and gypsum from chemical
industry mixed in one-to-one ratio. Water/gypsum
ratio equal to 0.71 was used for making the
specimens and it corresponds to the gypsum paste
with normal consistence.
For the purpose of this test, three identical
gypsum specimens of dimensions 40× 40 × 160 mm
were made in a stainless mould with three sections.
These specimens were put off the mould after
twenty minutes and the experimental measurement
of the Dynamic Young’s Modulus, Dynamic Shear
Modulus and Poisson’s ratio started immediately.
3 Impulse Excitation Method
The Impulse Excitation Method (IEM) was chosen
for determination of dynamic Young’s modulus,
shear modulus and Poisson’s ratio because of its
non-destructive character. This method is quick, the
measurement line is small enough and mobile. The
mechanical characteristics were determined based
on the equation for vibration of the isotropic beam
with continuously distributed mass, based on

ISBN: 978-1-61804-142-5 163

 Recent Advances in Mechanical Engineering and Automatic Control

measured resonant frequencies of longitudinal, where: Edl is the Dynamic Young´s Modulus [Pa], fl
flexural and torsional vibration of the specimens, is the fundamental longitudinal frequency [Hz], b is
dimensions of the beam and its mass. the width of the specimen [m], t is the height of the
The measurement line (Fig. 1) consisted of the specimen [m], l is the length of the specimen [m], m
acceleration transducer Brüel&Kjær of Type 4519- is the mass of the specimen [kg].
003, the impact hammer Brüel&Kjær of Type 8206,
the vibration analyzer Brüel&Kjær Front-end 3560-
B-120 and program PULSE 13.5. The vibration was
induced by the strike of the impact hammer. The
waveforms of the excitation force and the
acceleration were recorded and transformed using
Fast Fourier Transform (FFT) to the frequency
domain. The Frequency Response Function (FRF)
as the ratio of acceleration to the excitation force
was evaluated from these signals using the vibration
analyzer and program PULSE 13.5. The test was
repeated four more times for each specimen and the
averaged function FRF was saved. From an
averaged FRF, the fundamental longitudinal Fig. 2 Test arrangement for the longitudinal
resonant frequency was determined for each vibration
specimen.

3.2 Dynamic Young’s Modulus – Flexural

Vibration
Another way how to determine Dynamic Young´s
Modulus is flexural vibration. The specimen was
simply supported in the distance 0.224 of the span
on both ends, the fundamental flexural nodal
positions (Fig. 3). The acceleration transducer was
placed at the end of the specimen on the upper
surface (Fig.3 – right side). The opposite end of the
surface was struck by the impact hammer (Fig.3 –
left side).

Fig.1 The measurement line – Brüel&Kjær

3.1 Dynamic Young’s Modulus –

Longitudinal Vibration
The specimen was supported in the fundamental
longitudinal nodal position in the middle of its span
(Fig. 2). The acceleration transducer was placed at
the centre of one end surface of the specimen (Fig.2
– right side). The opposite end surface of the
specimen was struck perpendicular to the surface by Fig. 3 Test arrangement for the flexural vibration
the impact hammer (Fig.2 – left side).
The mass and dimensions of the specimen were The mass and dimensions of the specimen were
measured, the FRF was evaluated and the dynamic measured, the FRF was evaluated and the Dynamic
Young’s modulus Edl was determined [6-7] using Young’s Modulus Edf was determined [6-7] using
the relation the relation

4lmf l 2 0.9465l 3 mf f2T1

E dl  (1) E df  (2)
bt bt 3

ISBN: 978-1-61804-142-5 164

 Recent Advances in Mechanical Engineering and Automatic Control
where: Edf is the Dynamic Young´s Modulus [Pa], ff
is the fundamental longitudinal frequency [Hz], b is
the width of the specimen [m], h is the height of the
specimen [m], l is the length of the specimen [m], m
is the mass of the specimen [kg] and T1 is the
correction factor for fundamental flexural mode to
account for finite thickness of bar and Poisson’s
ratio [6].
3.3 Dynamic Shear Modulus – Torsional
Vibration
The dynamic shear modulus was determined based
on the torsional vibration. The specimen was
supported in the middle of the span, the fundamental
torsional nodal positions (Fig. 4). The acceleration
transducer was placed at the end of the specimen in
the right upper corner of the side surface (Fig.4 –
right side). The opposite left lower corner of the
same side was struck by the impact hammer.
Fig. 4 Test arrangement for the flexural vibration
The mass and dimensions of the specimen were
measured, the FRF was evaluated and the dynamic
shear modulus Edf was determined [6-7] using the
relation
4lmf t 2
Gd  [ B /(1  A)]
bt
where: Gd is the dynamic shear modulus [Pa], ft is
the fundamental torsional frequency [Hz], b is the
width of the specimen [m], t is the height of the
specimen [m], l is the length of the specimen [m], m
is the mass of the specimen [kg], A is the empirical
correction factor dependent on the width-to-
thickness ratio of the specimen and B is defined as
 b/t t /b 
B 6
 4(t / b)  2.52(t / b)  0.21(t / b) 
2
ISBN: 978-1-61804-142-5
3.4 Poisson’s Ratio
The Poisson’s ratio can be determined from Edl or
from Edf based on the equation
d  ( Ed / 2Gd )  1 (5)
Because the correction factor T1 in the equation
(2) depends on the µd, an iterative process has to be
used for µdf and Edf determination. As the first
approximation can be used µdl calculated from Edl.
The iterative process of equations (2) and (5) can
stop when the values of µdf for the previous and the
next steps will not differ more than 1%.
4 Results
The first established value was the Dynamic
Young’s Modulus determined from the longitudinal
vibration (Fig. 5). As it can be seen in Fig. 5, value
of Dynamic Young’s Modulus Edl increased from
the average value 3.8 GPa to 3.95 GPa in the first 3
hours after production of the specimens. Then the
values a little bit decreased and during the next two
days the value of Edl remained relatively stable
between 3.7 GPa and 3.9 GPa. After two days, the
values started to increase very rapidly and they
increased to the time of seven days then they
became stable or increased very slowly.
During the last ten hours of laboratory tests, the
maximal release of free humidity occurred. Initial
weight of the specimens was about 395 g and it
decreased to the values about 270 g.
The second variant of Dynamic Young’s
Modulus determination was the flexural vibration
(Fig. 6).
As it can be seen from the comparison of the
Fig. 5 and the Fig. 6, both results correspond very
well. Only the values determined from the flexural
(3) vibration are a little bit less accurate than the results
from longitudinal vibration.
(4)
Fig. 5 Dynamic Young´s modulus Edl
165
 Recent Advances in Mechanical Engineering and Automatic Control
Fig. 6 Dynamic Young´s modulus Edf
Fig. 7 Dynamic shear modulus Edl
The changes of dynamic shear modulus are
presented in the Fig. 7. The changes at the
beginning of the gypsum hardening are not bigger
than the inaccuracy of the used method. In the final
stage of measurements, increase of Gd values occurs
similar to the case of Ed measurements. Gd values
increased by 17 % to 1.75 GPa during weight loss of
the last 10 g of free humidity from the gypsum
samples.
Based on the calculated values of Ed and Gd, the
Poisson’s ratio was calculated for each measurement
of the specimen (Fig. 8).
Fig. 8 Poisson’s ratio
ISBN: 978-1-61804-142-5
5 Conclusion
The article presents the results of an experiment
during which the development of Dynamic Young’s
Modulus, dynamic shear modulus and Poisson’s
ratio of three gypsum specimens were monitored.
Several changes occurred during evolution of
these characteristics after production of the
specimens and starting their hardening. There is
visible increase of values at the beginning (Fig. 5-7
– right side). Then the values of mechanical
properties started to decrease a little bit. At the end
of monitoring, when the specimens lost about 100
grams of their weight, the values started to increase
very rapidly. The effect, which influences
mechanical properties evolution, is the water
content in a specimen. It is necessary to add more
water to a specimen during its making because of
workability then it would be necessary for theoretic
hydration. Next processes are probably related to the
formation of the solid structure during the first day.
Resulting values of Poisson’s ratio varies
because they are very sensitive to inaccuracies
during measurement of Ed and Gd. Even small
uncertainty of determination Ed and Gd influences
Poisson’s ratio significantly. The average value
seems to be constant from the beginning of the
measurement to the end. The average value is 0.33.
The trend of mechanical properties evolution was
monitored earlier but in comparison with literature
there was not done such amount of measurements in
such short intervals. The most of previous
measurements were done using classic static
methods, the results of which are dependent on
technological discipline during specimen
preparation and during testing but they are also
dependent on conditions where they are placed
(especially on temperature). The trend of
mechanical properties evolution was verified
nondestructively on the same specimens for the
whole time. Thus the above mentioned effects and
problems during static testing were eliminated.
This outcome has been achieved with the
financial support of the Ministry of Education of the
Czech Republic under the No.: MSM 6840770031
References:
[1] Deal, B., Grove, A., General Relationship for
the Thermal Oxidation of Silicon, Journal of
Applied Physics, Vol. 36, 1965, p. 3770.
[2] Fachinger, J., Behavior of HTR Fuel Elements
in Aquatic Phases of Repository Host Rock
Formations. Nuclear Engineering & Design
Vol. 236, 2006, p. 54.
166
 Recent Advances in Mechanical Engineering and Automatic Control

[3] Quintiere, James G., Fundamentals of Fire

Phenomena, John Wiley & Sons. Ltd,
Chichester, U. K., 2006.
[4] Clark, T., Woodley, R., De Halas, D., Gas-
Graphite Systems, in “Nuclear Graphite” R.
Nightingale, Editor. Academic Press, New
York, 1962, p. 387.
[5] Samochine, D., Boyce, K., Shields, J.,.
“Investigation into staff behaviour in
unannounced evacuations of retail stores -
Implications for training and fire safety
engineering,” Fire Safety Science -
Proceedings of the 8th International
Symposium, International Association for Fire
Safety Science, 2005, pp. 519-530.
[6] ASTM E1876-01, Standard Test Method for
Dynamic Young’s Modulus, Shear Modulus,
and Poisson’s Ratio by Impulse Excitation of
Vibration, Annual Book of ASTM Standards,
American Society for Testing and Materials,
2006.
[7] ASTM C215, Standard Test Method for
Fundamental Transverse, Longitudinal, and
Torsional Resonant Frequencies of Concrete
Specimens, Annual Book of ASTM Standards,
American Society for Testing and Materials,
1991.
[8] Padevět P., Zobal O., “Change of Material
Properties of the Cement Paste CEM I,”
Proceeding of Experimental Stress Analysis
2010. Olomouc: Palacky University, 2010,
pp. 307-310.

ISBN: 978-1-61804-142-5 167